Abstract:As the basic link of livestock product supply, animal husbandry is of great significance to ensure the quality of animal products and the healthy development of industrial economic benefits. At present, the information monitoring technology in animal husbandry is still lagging behind, and the penetration rate is not high. It is impossible to accurately and effectively grasp the actual conditions of the farm animals themselves and their living environment. On the basis of sorting out and summarizing the existing research results of wearable technology, combining with the actual situation and characteristics of animal husbandry in China, the working principle, information monitoring technology and wearable monitoring methods of wearable information monitoring in animal husbandry were mainly analyzed and discussed. The research trends were summarized as follows: the development of information acquisition from manual acquisition to automatic acquisition; wearable sensors would be miniaturized and flexible; signal processing and information transmission would develop towards diversification, compounding and intellectualization; the way of information monitoring would be systematic, holistic and adaptive; and the application of wearable information monitoring technology would improve the level of aquaculture management and decision-making.

Abstract:In order to segment uneven colored apple fruits in natural environment, the fruit segmentation method based on image features extracted from superpixels was proposed for apple harvesting robot. Firstly, simple linear iterative clustering (SLIC), which was one of superpixel clustering algorithm was employed to segment original images into a set of superpixels. The color of pixels in the same superpixel was uniform relatively. Then, the color and texture features of superpixels were extracted. According to combined feature vectors, these superpixels were classified into fruit class and non-fruit class by support vector machine (SVM). Finally, the classification results were modified based on the adjacency relation of superpixels. The segmented fruits were made up of a set of superpixels in fruit class. The experiment results showed that the proposed method can classify a majority of superpixels and there were average of 2.28 superpixels in one image were classified falsely. Compared with the segmentation method based on pixel-level features and the segmentation method based on features of neighborhood pixels, the proposed method based on superpixel features had a better performance on fruit segmentation. The experiment of image segmentation with 100 images indicated that the precision and recall of proposed method can reach 0.9214 and 0.8565 respectively before modifying classification results. The running time of proposed method was 0.6087s per image.

Abstract:Aiming at the problem that the traditional path following control method of mobile robot requires manual tuning parameters and lacks of self-optimization ability, a path following control method based on heuristic dynamic programming (HDP) was proposed. Firstly, the error state equation of the path following system was established based on the structure of tracked robot system. Secondly, a path following controller based on HDP algorithm was proposed. The return function was designed by synthesizing the error performance index and the following stability index. A multi-layer feedforward neural network was used to establish the actuator and evaluation modules, and the online optimization rules of network parameters were derived. Thirdly, the path following performance of HDP method was verified by numerical simulation and system test. The experimental results showed that the controller based on the HDP algorithm of the mean absolute error was 0.04m and the root mean square error was 0.06m when following a straight line, 0.01m and 0.06m when following a obtuse angle steering curve, and 0.03m and 0.09m when following an acute angle steering curve, and the method based on HDP can obtain high precision control effect without repeated debugging control parameters, and it improved the environmental adaptability and autonomous optimization ability of the path following control method for mobile robots.

Abstract:In order to extract rice seedling rows accurately, a detection method of centerlines of rice seedling row based on sub-regional feature points clustering was proposed. 2G-R-B characteristic factor and Otsu method were used to separate seedling and background from RGB rice seedling image. By sub-regional analyzing the distribution of seedling pixels, candidate feature points of seedling row were extracted. Then feature points were clustered with the nearest neighbor relationship between feature points, and the number of seedling rows and starting points of each seedling row were determined. According to the characteristics of row planting of seedlings, trend line was introduced to refine feature points. The real feature points indicating seedling rows were obtained by comparing the shortest distance of candidate point with its corresponding trend line with a distance threshold value. Afterwards, the centerlines were detected by fitting a straight line with the least square method. The experimental results showed that the proposed method achieved good anti-noise performance. The accuracy of centerlines detection was 95.6%, but the traditional Hough method and the randomized Hough method can only reach 84.1% and 89.9%, respectively. The average processing time of a 320 pixels×237 pixels color image was less than that of the two other algorithms. It can be seen that the proposed algorithm had the advantages of high real time and high accuracy, which can accurately extract seedling row centerlines, and the research result provided navigation parameters for an automatic rice transplanter walking along the seedling row in paddy fields.

Abstract:In order to make the flow field distributed orderly in the air-assisted high speed maize precision seed metering device, thereby improving the working performance of the seed metering device, three kinds of structure type diversion turbine were designed. Through the simulation and theoretical analysis of CFD method, the diversion turbine can effectively improve the air flow inside the seeding device and increase air speed where type hole in outer ring. At the air flow rate, the pressing force was increased, and the C diversion turbine with a large windward angle and a curved structure had better disturbance and guiding, and the effect was optimal. In order to obtain the best performance parameters of the C diversion turbine installed, the operation velocity, quantity of feeding and wind pressure were used as test factors, and the qualified index, the missing index and the multiple index were used as test indicators. A three-factor quadratic rotation orthogonal combination test was performed and the multiple regression analysis and response surface analysis of the test data were carried out by Design-Expert 8.0.6 software, and the influence of various factors on the indicators was obtained. The multi-objective optimization method was used to determine the optimal combination of parameters: the operation velocity was 9.8km/h, the quantity of feeding was 1.8kg/min, and wind pressure was 8kPa and the qualified index of seed metering was the highest. At this time, the qualified index was 91.32%, the missing index was 2.83%, and the multiple index was 5.85%. The comparison verification test was carried out on the optimization results, and the comparison was made under the same conditions with the uninstalled diversion turbine seed metering device, and it was found that the installation of the diversion turbine can effectively improve the working performance of the seed metering device.

Abstract:In order to solve the problem of poor uniformity and large trench resistance in the wide seedling strip wheat planter, a wheat precision planter based on cell wheel seed-metering device and seed conveying tube with inner-quartering part was designed. The planter could implement the following function at one time: seedling belt cleaning, fertilization, wide strip seeding, active covering soil and compacting and so on. The diameter of cell wheel seed-metering device was 60mm, the number of cell columns was 4, and the number of each cell columns was 20. The cells were arranged in a horizontal staggered condition. In order to ensure uniform distribution of seed in the seedling strip, the inner-quartering seed conveying tube was designed. The tube can convey seeds in interval space. The outlet of the seed conveying tube was 32°. The seed and fertilizer were ordered covering by soil, which was thrown by rotary tillage. The rotating speed was 260r/min. To ensure the capacity of soil covering, the position of seed and fertilizer conveying tube and fertilizer was devised by kinematic analysis of soil. The distance range rotary shaft from seed and fertilizer tube was 0.36m and 0.50m, respectively. The experiment of seed-metering device and the planter was conducted. The results of laboratory test showed that the coefficient of variation of inline uniformity, the coefficient of variation of row displacement, and the seed fragmentation rate were 9.31%, 2.30% and 0.38%, respectively, under condition that the rotation speed of cell wheel was 35r/min. Field test showed that the coefficient of variation of uniformity in the wide seedling strip was 11.55%, and the average width of wide seedling strip was 8.2cm. The results also showed that the rotary tillage could clean stubble well and avoid the problem of congestion by rank grass. At the same time, by using the principle of rotary tillage and throwing soil, and soil covered the seed and fertilizer with layered, the average depths of sowing and fertilization were 3.2cm and 9.4cm, respectively. And the average vertical spacing between seed and fertilizer was 6.2cm. The variation coefficients of sowing depth and fertilizing depth were 4.15% and 2.97%, respectively, and the variation coefficient of the distance between seed and fertilizer was 5.48%, which met the national standard and agronomic requirements.

Abstract:Considering the problems that the soil surface is uneven, large area, and broad width in the process of the sowing operation, which leads to the poor uniformity and stability of sowing depth in each line of the planter, a prototype of profiling chisel sowing opener was developed based on elastic deformation of profiling roller. The working principle of the profiling chisel sowing opener was elaborated. The curvilinear equation of the chisel blade edge was determined, the mechanical model of interaction between opener and soil was established, and the relationship between the elastic deformation of the profiling mechanism, the profiling resistance and the profiling quantity was clarified. The influence of chisel surface blade edge width, profiling roller diameter, entrance angle, and sowing depth on the sowing depth stability was tested with quaternary primary regression orthogonal combination test method and high speed digital soil slot platform. The experiment results indicated that when the blade edge width was 12mm, the profiling roller diameter was 24mm, and the entrance angle was 22°, the sowing depth stability was better. The field comparative experiment results showed that under the optimum combination of parameters the depth variation coefficient of the profiling chisel opener was less than 9%, the surface smoothness was lower than 18.5mm, and the profiling chisel opener had non-tangled grass, piled-up earth. The ditching performance of the profiling chisel opener was superior to that of wing spade opener. The field seeding emergence test results indicated that as the rapeseed and wheat planting by the broad width precision no-tillage planter for rapeseed and wheat with 4.8m working breadth and sowing 24 rows, the variation coefficient of single row seedling number and variation coefficient of each row seedling number were both less than 10%, the average seedlings number of each row of the wheat, the spring rapeseed, and the winter spring was 35, 26 and 20 within a meter, respectively. The seedlings number of each row of the wheat and rapeseed could satisfy the requirement of precision seeding and seedling rate, which can provide reference for the design of planting ditch opener for rapeseed and wheat.

Abstract:Aiming at the problems of inadequate use of air pressure for air suction type metering device, friction between seed plate rotation and rubber pad, and poor uniformity of hole sowing, airflow field of rice seed metering device embedded with rotating air chamber was designed by the absorption principle. The working principle of each stage of seed metering device was introduced, and the parameters and structure of the key components were designed. Based on the basic governing equation of fluid, the negative pressure cavity was simulated by ANSYS software. The numerical simulation was carried out with the influence factors of pore diameter and position distribution of suction hole. The average pressure of the cavity and the average velocity of the suction hole were the evaluation indexes. The simulation results showed that the distribution of flow field in air cavity was relatively stable, which provided a stable negative pressure environment for seed suction. The JPS-12 test-bed was applied to verify the simulation test. Taking the rotation velocity of air chamber, the negative pressure of air chamber and the height of seed filling as the influencing factors, and taking the qualified rate of rice sowing, the rate of missed sowing and the rate of re-sowing as the evaluation indexes, the experiment of quadratic rotation combination design at five levels of factors was carried out. Using Design-Expert 8.0.6 to complete variance analysis and significance test, the regression equation and the corresponding surface map were obtained, the interaction of various factors was analyzed, and the best parameters were determined. Number combination: when the rotating velocity of the air chamber was 21.61r/min, the negative pressure of the air chamber was 4.4kPa and the seed filling height was 15.7cm, the performance of the operation was 93.6% for the qualified index, 3.47% for the missed broadcast index and 2.93% for the replay index. The experimental results of bench test were in good agreement with the optimized results, which met the requirement of direct seeding of japonica rice.

Abstract:The pulsed smoker/fogger powered by pulse engine could meet different requirements on disease and pest control for application smoke or fog during the different growth periods of plants. The working frequency of pulsed smoker/fogger is depending on the pulse frequency when the acoustic structure condition, heating condition, and feedback effect of pesticide were coupled together. Once any condition was not satisfied or the feedback disturbance was too large, an oscillating system could not be formed or the original oscillation system would stop to the pulse engine flameout. The experiments of working frequency were performed, the data were processed and analyzed by changing the throttle opening of pulse engine, replacing hot smoke pesticide by 0# diesel, replacing water fog pesticide by clear water, and changing the liquid flux entering into the exhaust pipe. By using the acoustic resonance frequency equation, it was found that the actual operating frequency was higher than the theoretical value without spraying, and the average relative error was 26.6%. In the state of spraying, when the hot smoke or water fog pesticide was injected into the exhaust pipe, the original working frequency would be decreased after the heat fuming or atomization. However, the reduction of the working frequency of water fog after atomizing was much larger than that of the hot smoke after fuming. In other words, the water fog pesticide after atomizing had more disturbances to the original pulsating combustion oscillating system than the hot smoke after fuming. The calculation formula of working frequency of the pulsed smoker/fogger under spraying smoke or water fog were established. The actual working frequency was negatively proportional to the liquid flux. The linear ratios were -0.1357Hz·h/L and -0.1157Hz·h/L, respectively. The actual working frequency was proportional to the fuel consumption rate. The linear ratios were 38.58Hz·h/L and 30.73Hz·h/L respectively. The maximum relative average errors caused by these two equations were only 2.2% and 1.4% respectively. This demonstrated that within the range of the conventional spraying amount, it was feasible to calculate the corresponding working frequency by the established equations.

Abstract:In recent years, with the development of technologies of agricultural aviation plant protection, plant protection drones was used more and more in the prevention and control of pests and diseases. Aiming at the multi-team collaborative operation mode for plant protection orders in the region, the key factors such as order time window, farmland infection status and multi-machine coordination were comprehensively considered, and a multi-objective scheduling model for UAV plant protection team with the maximum total revenue and the shortest total operation time as the optimization objective was established. The order priority sorting algorithm and the path planning algorithm based on the non-dominated sorting genetic algorithm (NSGA-Ⅱ) were designed to solve the model. A case study of plant protection operations in Wugong County, Shaanxi Province was carried out to illustrate the validity of the proposed model and algorithm. Experiments showed that the model and algorithm designed can output the Pareto optimal solution set satisfying the time window constraint, and can give reasonable transfer path and time arrangement for UAV plant protection team. Moreover, the algorithm had good search performance and stable convergence performance, which can meet the needs of real scheduling problems. At the same time, research showed that increasing the order window time was beneficial to increase the total operating income and reduce the total operation time. The research can provide a scientific basis for the deployment and decision analysis of the UAV plant protection team, and provide reference for the development of the agricultural machinery intelligent dispatching system.

Abstract:Although China is the largest country of root and stem crops production, its mechanized production technology is very backward, and the specialization, high efficiency and serialization of the working machines are still lower, which lags behind developed countries. Aiming at the extrusive problems in the domestic harvester such as low level of mechanical automation, high labor intensity, lack of auto-follow row function, low obvious rate, high injury rate and high undug rate, taking 4UGS2 double row potato harvester as the carrier and the ridge section direction as the research object, comprehensively using mechanical, hydraulic and electronic control and other key technologies, the auto-follow row device was designed, which consisted of ridge profiling mechanism, traction mechanism, hydraulic system and electronic control system. The real-time adjustment algorithm of harvester operation route was introduced based on PID control technology, and the harvester operation route adjustment model was constructed. Finally, the optimal control of track of harvester was tested to verify the effectiveness of potato harvester, which included the installation of auto-follow row device and the non-installation of auto-follow row device. The auto-follow row device was used to harvest three ridges and then relying on the tractor driver to manually match the row to harvest three ridges. The experiment was carried out by interlacing with six ridges, and the harvester speed was 5.02km/h (i.e. 1.40m/s). The results showed that 4UGS2 harvester exhibited better adaptation on the moist land with soft soil and less hardening, and the auto-follow row device and other parts of the machine had good working performance. The obvious rate, injury rate, undug rate and other indicators reached or exceeded the qualification requirements. According to the experiment and statistics, in the three ridges with manually matching the row, the obvious rate was 95.09%, the injury rate was 2.84%, and the undug rate was 3.38%. Meanwhile, in the three ridges with auto-follow row device, the potato obvious rate, injury rate and undug rate were 97.25%, 1.44% and 1.57%, respectively. Through comparison, it can be seen that when the harvester was installed with the auto-follow row device, the obvious rate was increased by 2.16 percentage points, the injury rate was decreased by 1.40 percentage points, and the undug rate was decreased by 1.81 percentage points. Due to the installation of auto-follow row device, the harvester can automatically adjust the working posture according to the change of the ground ridge, carry on the positive excavation and realize the accurate control of the harvest operation. The potato obvious rate, injury rate and undug rate were improved simultaneously. Besides, the stability of the auto-follow row device would be affected if the ridge cross section size was changed greatly and individual collapse of the ridge occurred in the course of the test. The machine was characterized by its stable and reliable work, efficient operation and better performance index. The research can provide an effective reference for the development of intelligent harvesting technology for other root and stem crops harvesting machines.

Abstract:In the view of fixed spacing and poor adaptability of the existing seedling needle，and the difficulty of ensuring the on-demand adjustment to adapt to different plugs of seedlings in the transplanting process，a needle-type seeding end-effector with adjustable insertion pitch was designed. The actuator was driven by a DC electric push rod，and the seedling needle was inserted obliquely along the four corners of the root. The adjustment of spacing among the needles was achieved by changing the position of the slider and the needle holder on the bracket. The seedlings of cabbages plugs were studied，namely 72，128 and 200 points. The effects of seedling age on the formation of root mites and the growth of cabbage seedlings were studied. The main structural parameters of the end-effector were determined，and the force analysis was carried out on the taking seedling. Under the premise that the roots were formed，single factor simulation was conducted for the seedling picking by EDEM. The specifications of plugs，the acceleration of seedling picking，the insertion ratio of the needle and the moisture content of root were regarded as the test factors. The root integrity rate and the failure rate of the seedling picking were regarded as evaluation indexes. Using L9(34) orthogonal test to investigate the influence of test factors on evaluation indexes. Through the range analysis，variance analysis and comprehensive weighting method，the primary and secondary influence of factors on the indexes were obtained as the acceleration of seedling picking，the moisture content of root，the specifications of plug，and the insertion ratio. The preferred parameter combinations were obtained as the acceleration of seedling picking was 0.1m/s2，the moisture content of root was 56.2%，the specification of plugs was 128 and the insertion ratio was 15%. At this time，the root integrity rate was 97.93%. The failure rate of the seedling picking was 0.81%.

Abstract:In order to reduce the cost of shrimp pond breeding and achieve uniform delivery of bait，a self-guided feeding boat based on paddle wheel drive was designed. The paddle-driven shrimp pond feeding boat can adapt to the complex environment of the breeding pond and meet the requirements of the whole pond throwing. Reliability was the key to its application. Fully enclosed feeding hull was designed by rotomolding process. Oil-free lubricating stainless chain and paddle wheel were used as driving mechanism to avoid pollution to pond water. Spiral bait conveying device met the requirements of stable barycenter position in the process of boat-borne bait delivery. The requirements of autonomous navigation，positioning and attitude control were realized through GPS and electronic compass. According to the feeding and control performance requirements of shrimp ponds，the feeding boat straight and spot turning motion model was constructed，and the cruise path control was carried out by PID heading and speed motion control algorithm. The results of the aquaculture pond test showed that the average speed was 0.72m/s，and the maximum yaw amount of the autonomous navigation feeding boat based on the paddle wheel drive was 0.8m and 0.5m when going straight and turning， respectively. The 40 day aquaculture pond field test results showed that the self-guided feeding boat ran smoothly under complicated path and met the feeding requirements of shrimp ponds. Fully reliability tests were carried out on harsh environments such as strong winds，heavy rains，and related problems were found and solved，enabling it to be promoted and applied.

Abstract:In deep sea area, fishing in large cage aquaculture is low efficiency and high labor intensity, and the fish injure rate of fishing appliances such as fish pump is high, so it needs to be equipped with high-power generation system, which is not suitable for small and medium sized fishing boats. An internal guide vane drum machine was designed for fishing which had lower power consumption by the screw conveying principles. And there was no such appliance in marine aquaculture. According to the structure simulation, the relationship between deformation, equivalent stress, speed and dip angle was got, as well as the position where had the maximum equivalent stress between the drum’s surface helical vane and drum’s inner wall was located. As the simulation shown, stress concentration occurred in support shell of binding force in bearing. The experimental setup was designed and optimized according to the simulation results. When the length of the drum was certain, the potential energy of transported fish was related to different dip angle. The transmission rate to fish potential energy was defined as the appliance effective transmission. A special formula can be used to be the criterion of optimizing the control scheme. The numerator of this formula was an interpolation equation of effective transmission rate at different speeds. The denominator of this formula was an interpolation equation of power consumption at different speeds. The reliability, power consumption and fish injure rate experiments were conducted. The results demonstrated that there was an optimal control scheme under certain constraints, it was the drum with parameters of 10°dip angle and 15.3r/min speed. Moreover, the optimized rotating speed of the drum was 15r/min under different dips. The experimental results showed a lower fish injure rate of 0.1%, which improved the quality of fishing.

Abstract:Rice is one of the most significant food crops all over the world. Biomass is a key phenotypic trait in rice research. A new method for nondestructive detection of rice biomass for multiple varieties at reproductive stage based on deep belief network was proposed. RGB images of 483 different rice varieties under normal growth environment and drought stress environment were captured at three time points: before stress, after stress and after rehydration. After image acquisition, the images were segmented by using fixed threshold in HSL color space and 57 image-derived features related to rice biomass were extracted. After data normalization, a rice biomass model was built based on deep belief network. The influences of visible layer type, hidden layer number, hidden layer neuron number, learning rate, epoch number and momentum on the performance of deep belief network were tested. The best model was selected based on the coefficient of determination (R2), mean absolute percent error (MAPE) and standard deviation of absolute percent error (SAPE). The deep belief network model was also compared with the stepwise linear regression model. The results showed that the biomass measurement model based on the deep belief network performed better (R2 was 0.9299, MAPE was 11.19% and SAPE was 18.36%). The research offered a new nondestructive method for accurately measuring rice biomass for multiple varieties under different growth environments, which would provide a new tool for rice research.

Abstract:With the development of rice phenomics research, it is of great significance for comprehensively analyzing, mining and applying the rice phenomics data. In order to integrate the knowledge related to rice phenomics and explore the factors affecting rice phenotypic traits，the rice phenomics knowledge graph system was implemented. Rice phenomics knowledge graph system consisted of functional modules such as entity recognition, entity query, relational query and knowledge visualization. The rice phenomics data were downloaded by a distributed data website crawler from the National Rice Data Center website, and the interactive encyclopedia website was taken as auxiliary data sources to obtain rice phenomics dataset. The dataset was preprocessed with TF-IDF and latent semantic indexing method and classified and labeling manually firstly, and then machine learning approaches were applied for training and testing. The rice phenomics entity classification was studied based on stacking ensemble learning integrated with basic learning classifier, such as K-nearest neighbor, support vector machine, random forests and gradient boosting decision tree. Based on stacking ensemble learning classifier, different types of rice phenomics data showed fine ability for entity classification. For the unbalanced rice phenomics entities, comparing with the support vector machine algorithm, the K-nearest neighbor algorithm, the random forest algorithm and the gradient boosting decision tree algorithm, the proposed method showed the best performance, i.e. the F1-Measure of Gene entities can reach 90.47%. The overall accuracy was 80.55%, and it was 6.78 percentage points higher than those of the other four basic classifiers.

Abstract:The objectives were to enhance calculation accuracy of physiological parameters of photoresponse, and assess photosynthesis mechanism and characteristics of maize under fertigation of drip irrigation in Ningxia Yellow River Irrigation Area. Using Tianci-19 (TC19) as experiment materials, the photoresponse curves of spike leaves of maize were measured by Li-6400XT gas exchange system under six nitrogen rates at silking stage of maize from 2017 to 2018. The rectangular hyperbola modified model and other four kinds of common models were used to analyze photoresponse process of spike leaves of maize, with the aim of evaluating and selecting optimal model to calculate the photoresponse parameters of maize. The results showed that there were differences in the fitting accuracy of the four models, and the right angle hyperbolic, non-orthogonal hyperbola and exponential models had poor fit to the nitrogen-deficient (N0) photoresponse curve, and the rectangular hyperbola modified model had the highest fitting degree to each rate which was used as the optimal model to calculate the photoresponse parameters of maize at silking stage. The photosynthesis ability of maize was increased and then decreased with the increase of nitrogen rates. The photoresponse parameters (compensation point, light saturation point, maximum net photosynthesis rate, apparent quantum efficiency and dark respiration rate) of N4 (360kg/hm2) were all higher than those of other nitrogen rates, and there was a slight decreasing trend in N5 (450kg/hm2). The conclusions can be drawn that the rectangular hyperbola modified model can be used to calculate the photoresponse parameters of maize, so as to diagnose the nitrogen nutrition status, regulate the optimal nitrogen application rate, improve the photosynthetic capacity at silking stage, and enhance the yield of maize.

Abstract:Crop identification and classification, as a fundamental part of modern agriculture, is an important prerequisite for ensuring regional food security and sustainable agricultural development. With the development of remote sensing technology, high-resolution visible light remote sensing imagery has become a convenient and reliable source of remote sensing data. At present, the use of these remote sensing data to carry out detailed classification research of typical crops has extremely important practical significance. However, these remote sensing images lack enough spectral information and therefore are unable to give high-accuracy recognition of the crops, it is necessary to apply deep learning techniques on the crops classification research to solve these problems. Based on composite wing unmanned aerial vehicle which equipped a remote sensing imaging equipment to obtain remote sensing data, including cotton, corn and squash, covering an area of over 867hm2. According to the characteristics of the data, the convolutional neural network (CNN) was designed to extract the crop classification information. By adjusting the network parameters and the sample spectral combination, the parameter optimization of the training process was divided into two levels: at the level of network parameters, the adjustment included the learning rate and the batch, the scale of the convolution kernel and the depth of the network;at the spectral feature level of the sample, three types of samples included single-band, dual-band and triple-band were constructed as inputs, and the model was trained in turn. The experimental results showed that the CNN can effectively extract the crop information in the image and realize the fine classification of crops. Most sensing areas had qualified classification result except the edge places planted with sparse and mixed crops. The overall classification accuracy achieved 97.75% by using the optimized CNN. Compared with the support vector machine based on radial basis kernel function and the back propagation neural network, the optimized CNN had the best effect and the highest classification accuracy. It was worthy of noting that the adjustment of network parameters would affect the final training effect. A CNN model with large learning rate (0.1), small convolutional kernel (7×7) and appropriate depth (7) was advised on the basis that the typical crops in the remote sensing images were with high density, small features and rich textures. This promised the small features in the sample would not be missed when the convergence of the training accuracy was increased. Samples with different spectral features also had an impact on the training of the model. The blue band in the visible light was more appropriate than the green and red ones on training the CNN for crops recognition. A combination of the three bands would improve the recognition accuracy and stability, but more training time was required since more input information was given. The experiments demonstrated the effectiveness and reliability of the proposed CNN on crops fine classification. This method can be regarded as a reference for the application of deep learning in agricultural remote sensing.

Abstract:Soil total nitrogen content is an important soil fertility parameter. However, the existed portable detector of soil total nitrogen content had several deficiencies, such as weak light intensity and low signal-to-noise ratio. Hence, a new portable soil total nitrogen content detector was designed based on high intensity light source and special optical fiber. Six sensitive wavelengths were used, which were 1108nm,1248nm, 1336nm, 1450nm, 1537nm and 1696nm. A high-power tungsten halogen lamp was used as the light source to improve the signal intensity, and one-way incident optical channel and six-way reflected optical channel were realized by using a “one-six” special optical fiber. The performance test of the detector showed that the conditioning circuit and the tungsten halogen lamp source worked stably, and the tungsten halogen lamp had higher measurement accuracy than the LED as the active light source. The prediction model was established by the absorbance values of the detector and the soil total nitrogen standard value. The highest model accuracy was obtained from a PLS model. The calibration R2C of the model was 0.8613 and the verification R2V of the model was 0.8042, which showed that the model had high precision and could be used for detector model embedding. A field experiment was conducted and the correlation coefficient between the real-time measurement result of the detector and the soil total nitrogen standard value reached 0.8280. The detector had high measurement accuracy and can meet the purpose of fast and accurate detection of the soil total nitrogen content in the field.

Abstract:The timely and accurate identification of estrus in dairy cows is an important basis for improving the success rate of artificial insemination and conception rate. Artificial estrus monitoring of dairy cows is time-consuming and laborious, and activity and non-contact video monitoring cannot achieve calm estrus early warning. To solve these problems, based on the physiological changes of vaginal mucus during estrus in dairy cows, an estrus monitoring scheme for dairy cows was proposed based on the changes of vaginal resistance. An implantable resistance sensor for dairy cow vagina was designed to collect the value of vaginal resistance, which consisted of a brass resistance probe and an 8-claw anti-skid device. The wireless transmission system was developed with the help of 2.4GHz ZigBee network, which transmitted the vaginal resistance value to the coordinator node wirelessly, and the coordinator was transmitted to the monitoring center via RS485 bus. A real-time monitoring system for vaginal resistance of dairy cows was developed, which realized the accurate collection and remote real-time monitoring of vaginal resistance of dairy cows. The accuracy, stability, reliability and energy availability of the implantable resistance sensor and monitoring system were tested respectively. The results showed that the implanted resistance sensor was easy to install and light in weight, and could measure resistance in the range of 1~1000Ω. The measurement accuracy was less than ±2% and the maximum fluctuation of resistance in 24h was 2Ω. In a 450m2 dairy farming area, the success rate of ZigBee network data transmission was not less than 98.5%. The implantable sensor terminal node could work continuously for 38d with 7.4V/6500mA·h lithium battery energy supply. The monitoring system ran stably and could monitor the change of vaginal resistance of dairy cows accurately, reliably and in real time. The research can provide a new method for accurate prediction of estrus degree and ovulation time of dairy cows, and also provide reference for estrus detection of other animals.

Abstract:Aiming at the problems of high throughput, high concurrency and slow response of farmland data in the process of multi-condition processing, such as high computational load and slow response speed. The data processing technology of load balancing large-scale cluster was studied, the Hbase farmland database in multi-condition retrieval was optimized, a two-level non-primary key index method was put forward based on Solr, and a large farmland data platform was buildt based on Hadoop. The 100TB data was generated by eight operations, such as subsoiling, plant protection and conservation tillage, and those data was retrieved and tested on the platform based on Hadoop. The experimental results showed that the response time of the optimized technical model was less than 1 s when the concurrent volume of farm data was 50 million, and the performance of the optimized model was improved by about four times compared with the original Hbase. When the concurrent volume of simulated users were 500000, the query per second （QPS）and transaction per second （TPS）of the system were increased by about one time, the response time （RT） of the system was increased by 2.5 times, and the average response time was 183ms. To a certain extent, this system solved the problem of low efficiency of farmland data retrieval caused by high throughput and concurrency, and improved the computing ability of real-time processing of massive farmland data.

Abstract:To effectively reduce the power consumed during data transfer between forest microclimate monitoring stations, a dictionary-toggling-based compressed sensing method was proposed. After the sample data were characterized and classified, a discrete Fourier transform fixed dictionary and K-SVD learning dictionary was switched to realize sparse expression and compression of the sample data. And then the sample data were fitted by using a Gaussian function. The coefficient of determination R2 and root-mean-square fitting error were adopted as toggling factors to define the dictionary toggling strategy. Parameters such as air temperature, air humidity, soil temperature and soil moisture content were selected for testing to verify the feasibility of the dictionary-toggling strategy. Experimental results revealed that when the sparseness and compression rate were identical for forest microclimate monitoring stations, combining the advantages of DFT and K-SVD dictionaries, the dictionary-toggling-based compressed sensing algorithm yielded smaller reconstruction errors than those based on single dictionary. Experiments on the actual power consumption of stations demonstrated that when the sparseness K was 16, the dictionary-toggling-based compressed sensing algorithm caused the average daily power consumption to be decreased by 16.35%. Thus, the proposed dictionary-toggling-based compressed sensing method ensured low-power consumption operation and reliable data transfers for forest microclimate monitoring stations.

Abstract:A comprehensive inversion of the water quality information of sewage water was realized through the combination of hyperspectral technology and water quality comprehensive evaluation method. Taking the sewage sample collected by a sewage treatment plant in Shaanxi as the research object, principal component analysis (PCA) was used to comprehensively evaluate the sewage water quality to obtain a comprehensive evaluation factor for water quality. At the same time, the original wastewater spectrum was obtained by the ASD FieldSpec 3 hyperspectral instrument. After data preprocessing and different mathematical transformations, four spectral indices were obtained: spectral reflectance (SG), reciprocal logarithm (LR), standard normal variable (SNV) and continuum removed (CR). Based on partial least squares regression (PLSR), stepwise regression (SR) and extreme learning machine (ELM), a hyperspectral model of inversion water quality comprehensive evaluation factor was constructed. The results showed that the original spectral data of this group of water samples and the spectral data modeling by standard normalization transformation had good modeling results, and the prediction effect RPD of the model was above 2.5. Among the three models, the PLSR model and the ELM model had good modeling prediction effects, while stepwise regression modeling results were declined compared with PLSR model and ELM model, the R2c and R2p of the REF-SR and SNV-SR models were all above 0.8 and 0.85, and the RPD was above 3.0, which still had a very good inversion prediction effect, and it achieved the optimization of the model and the optimization of the characteristic band, and SNV-SR-ELM (R2c=0.956, R2p=0.954, RMSE=0.500, RPD=4.651) was the best model. The establishment of SNV-SR-ELM model provided a way for the optimization of hyperspectral inversion water quality model and the rapid evaluation of sewage water quality.

Abstract:In order to study the effects of irrigation and drainage modes and nitrogen level on rice plant height and tiller growth dynamics, the water level in the paddy field was taken as the control index in the leaching-pond. Logistic curve and DMOR mathematical model were used to quantitatively analyze the effects of drought and flood alternating stress in tillering (FHL) and the drought and flood alternating stress in jointing-booting (BHL) stage and controlled irrigation and drainage treatment (CID) at different nitrogen levels on plant height and tiller dynamics. The results showed that the irrigation and drainage modes promoted the plant height and tillers mainly by changing the growth time and increasing growth rate of plant height and tillers. The maximum of plant height was shown as BHL, FHL and CID in descending order and the maximum number of tillers was shown as FHL, BHL and CID in descending order. Nitrogen fertilizer application promoted the plant height and tillers mainly by increasing growth rate of plant height and tillers. The maximum of plant height was increased with the increase of nitrogen application rate, from 98.9cm and 97.8cm with 150kg/hm2 input to 102.4cm and 101.2cm with 300kg/hm2 input, and no further significant increase was observed with more nitrogen input. Under the treatment of CID and BHL, maximum number of tillers was increased with the increase of nitrogen application rate. Under the treatment of FHL, the maximum tiller number under low nitrogen treatment and high nitrogen treatment was less than that under medium nitrogen treatment. The interaction between irrigation and drainage mode and nitrogen level had a very significant impact on the dynamic process of plant height and tillering. The decrease of plant height and tiller number caused by water stress could be alleviated by increasing nitrogen fertilizer to some extent. However, high nitrogen level can aggravate water stress, which was not conducive to the growth of plant height and tillers.

Abstract:Aiming to investigate the effects of different combinations of irrigation and nitrogen on shoot biomass, yield, total nitrogen accumulation, and water and nitrogen use efficiency of drip fertigated spring maize in sandy soil area in Ningxia. The multivariate regression analysis method was used to obtain the best combination of irrigation and nitrogen for maximum production, and the spatial analysis method was further used to comprehensively evaluate the yield, grain nitrogen accumulation, and water and nitrogen use efficiency to seek a high-yield and low-water fertigation system for the closest acceptable production of multiple targets. The plot experiments were conducted with two factors of irrigation and nitrogen. There were three irrigation levels (W0.6: 0.6KcET0;W0.8: 0.8KcET0 and W1.0：1.0KcET0, Kc was crop coefficients, ET0 was potential reference crops evapotranspiration) and four nitrogen levels (N150: 150kg/hm2;N225: 225kg/hm2;N300: 300kg/hm2 and N375：375kg/hm2). The results showed that there was a high fitting degree of shoot biomass for spring maize with Logistic function and the W1.0 irrigation treatment delayed the starting point of the rapid accumulation period of shoot biomass. The single factor of irrigation and nitrogen application had significant or extremely significant effects on the yield, plant nitrogen accumulation and water use efficiency (WUE), and irrigation amount had significant effect on nitrogen partial factor productivity (PFPN). Both irrigation and nitrogen application amounts had significant or extremely significant effects on yield, plant nitrogen accumulation and WUE, while irrigation amount had extremely marked effects on PFPN, and irrigation and nitrogen had a notable coupling effect on the nitrogen harvest index. Under the same irrigation levels, the shoot biomass, yield, plant nitrogen accumulation (except for W0.8 treatment) and WUE were increased at first, and then decreased with the increase of the nitrogen application rate. Through the multivariate regression analysis and the spatial analysis methods, considering the uneven distribution of annual rainfall in Ningxia, the results suggested that when the sum of irrigation amount and effective rainfall was 506～576mm and nitrogen application rate was 230～335kg/hm2, the yield, WUE and grain nitrogen accumulation could reach more than 95% of the maximum yield, the PFPN under this combined treatment was about 80% of the maximum value at the same time. The research provided a guidance for scientific management of water and nitrogen application for spring maize under the local drip fertigated condition in sandy soil area in Ningxia.

Abstract:A continuous in situ field experiment from 2017 to 2018 in the cold and black soil region of Heilongjiang Province was carried out to explore the effect of straw returning and different N application rates in successive years on nitrogen composition in surface soil of paddy field under controlled irrigation mode. Two straw returning levels: 6t/hm2 and 0t/hm2, and four N application rates: N0(0kg/hm2), N1(85kg/hm2), N2(110kg/hm2) and N3(135kg/hm2) were set, a total of eight treatments. The changes of ammonium nitrogen (NH+4N), nitrate nitrogen (NO-3N), soluble organic nitrogen (SON), δ15N content and total soluble nitrogen distribution ratio in paddy soil surface after straw returning were analyzed based on nitrogen stable isotope technique. The results of 2-year experiment showed that under controlled irrigation condition, straw returning increased the ratio of SON to total soluble N, N mineralization and the content of δ15N in soil surface layer. The SON content in the surface soil layer of treatments with straw was lower than those without straw at different N application rates. Among all the treatments, the NH+4N and NO-3N contents in the surface soil of N3 treatments were decreased by 40.3% and 38.7% compared with those treatments without straw at the same N application rate. Straw returning promoted the existence of total soluble nitrogen in the soil surface in a more stable form of SON. Straw returning improved soil nitrogen supply capacity compared with treatments without straw returning, there was no significant difference in N mineralization in the surface soil between N0 with straw and maximum N mineralization of the treatment without straw (P>0.05). With the increase of N application, the N mineralization in the surface layer of soil treated with straw returning was significantly higher than that without straw (P<0.05). The high content of δ15N in straw promoted the enrichment of δ15N in soil surface. There was no significant difference in the content of soil surface δ15N between the treatments of N1 and N2 with straw and treatments of N2 and N3 without straw. The content of soil surface δ15N of the treatment with straw was significantly higher than that without straw (P<0.05). Moreover, the correlation between total soluble nitrogen and NH+4N, NO-3N, SON and δ15N in the soil surface was changed after two years of straw returning. The results of this study can provide scientific basis for the feasibility of straw returning in Northeast China, which had important significance for ensuring the sustainable utilization of agricultural water and soil resources in Northeast China.

Abstract:In order to reveal the uptake and utilization of nitrogen fertilizer at different stages in rice under water and biochar management, the method of combining field plot experiment with micro-area was used, and the 15N tracer technique was used to mark the application of basal fertilizer, tiller fertilizer and panicle fertilizer respectively. Taking conventional flooding irrigation as a comparison, the uptake, utilization, accumulation and transportation of basal, tiller and panicle fertilizer in rice under different water and biochar managements and the distribution of nitrogen fertilizer applied at different stages of rice maturity in different organs under two different irrigation modes were studied. The results showed that reasonable water and biochar management could significantly increase the total nitrogen accumulation, nitrogen uptake and utilization efficiency and yield of aboveground parts of rice at maturity stage. The contribution rate of fertilizer to total nitrogen accumulation under different water and biochar managements was about 17.81%~20.60%. There was no significant difference between the two irrigation modes (P>0.05). The uptake and utilization rates of basal fertilizer, tiller fertilizer and panicle fertilizer in rice were 15.55%~23.31%, 31.68%~44.91%, 48.82%~71.18%, respectively. The application of appropriate amount of biochar could significantly improve the uptake and utilization rates of basal tiller, tiller and panicle fertilizer. The uptake and utilization rates of tiller fertilizer and panicle fertilizer under dry-wet-shallow irrigation mode were better than those under conventional flooding irrigation except basal fertilizer. The uptake and utilization rates of tillering and panicle fertilizers were positively correlated with the total nitrogen uptake and utilization rates of fertilizers (P<0.01), and the contribution rate of nitrogen transport of basal, tiller and panicle fertilizer to grain was positively correlated with the corresponding uptake and utilization rates (P<0.01). Reasonable water and biochar management can improve the contribution rate of nitrogen transport to grain and the utilization rate of nitrogen absorption, and reduce the residue of nitrogen fertilizer in soil. The results can provide theoretical basis and technical support for the application of water and biochar management model in paddy field in cold black soil region.

Abstract:Aiming to solve waterlogging stress, an obstacle that limits agricultural production under rain-shelter cultivation in southern China, and quantitatively evaluate the possible influences of biochar on alleviating damage to crops caused by waterlogging stress, tomato was taken as the experimental object, and the effects of different groundwater tables and biochar application rates on water consumption, soil redox potential and yield were analyzed systematically by soil column experiments. The results showed that the shallower the groundwater table was, the more serious the influence of waterlogging stress on tomato was, leading to less water consumption. The total water requirement was significantly reduced with biochar application. However, the soil water kept by biochar was weakened with the decrease of groundwater table. The groundwater recharge was decreased with the increase of groundwater depth. Under the same groundwater table, the application of biochar could significantly increase the contribution of groundwater to water consumption. Biochar application could increase the soil redox potential and improve the ventilation performance of soil. When the groundwater table was -80cm, 5% biochar addition could significantly improve tomato yield and water use efficiency, with an increase of 38.7% and 56.6%, respectively. Groundwater table significantly affected tomato yield, but the interaction of groundwater table and biochar had no significant effect on yield and water use efficiency.

Abstract:Aiming to reveal the fate and transformation mechanism of soil nitrogen under coastal facilities cultivation, three treatments, including subsurface drainage combined with microbial organic fertilizer application (S-OF), subsurface drainage combined with inorganic fertilizer application (S-IF) and inorganic fertilizer application without subsurface drainage (CK) were designed. The model system was grape and rape intercropping cultivation. The distribution of soil total nitrogen in vertical profile, the changes of mineral and organic nitrogen in tillage layer and their correlation with soil physical and chemical properties were observed. The results showed that soil bulk density was decreased and soil porosity was increased under the combination of subsurface drainage and microbial organic fertilizer application;subsurface drainage promoted the transfer of total nitrogen from topsoil to deep soil, and the reduction rate of total nitrogen in topsoil of S-OF treatment was smaller than that of S-IF treatment;more than 80% of total nitrogen in topsoil existed in the organic form, and the proportion of mineral nitrogen was very small, S-OF was beneficial to the increase of mineral nitrogen content in the later stage of the experiment;the mineral nitrogen content in plough layer soil was positively correlated with soil organic matter and total organic carbon. Subsurface drainage and application of microbial organic fertilizer were conductive to improving the soil structure in coastal land, increasing the content of organic matter and total organic carbon in tillage soil, and promoting the transformation of soil organic nitrogen to mineral form. The conclusions can provide scientific basis for the improvement of coastal greenhouse soil and the decision-making of irrigation and fertilization.

Abstract:Mixing Yellow River water with brackish water for drip irrigation can be an effective solution of water shortage because the mixed water contains less sediment and salinity. However, the mixed water can still cause emitter clog and significantly reduce drip irrigation system performance because of the sediment particles and salt separators. The surface morphology of the clogging substance is an integral reflection of multiple factors. Therefore, three different types of water, including the Yellow River water (YR), brackish water (SW) and mixed water (MW) of Yellow River water and brackish water at the mixing ratio of 1∶1 were used to cultivate the emitter’s clogging substance. The effect of the mixed water on the surface characteristics was studied by 3D white-light interference profilometer. The results showed that the surface of clogging substance in condition of the mixed water was less protruding and more flat, the thickness, Sq and Sdr were decreased significantly. At the end of the system operation (800h), the average thickness of the clogging substance under the mixed water was 4.39μm, which was 65.32% and 32.51% lower than that of the Yellow River water and brackish water, respectively. The result showed that the mixed water can not only effectively reduce the sediment content in the Yellow River water, but also reduce the salt content in the brackish water. The mixed water relieved emitter’s clogging effectively because of low sediment content and salinity.

Abstract:In order to simplify the structure of rocker-arm sprinkler and improve its hydraulic performance, a jet sprinkler with double nozzles was designed by using negative pressure feedback technology, including jet elements, main and auxiliary nozzles and rotary sealing mechanism. The length of the main and auxiliary nozzles were 5.6mm and 4.8mm, respectively, and the elevation angle of the nozzle was 30°. The sprinkler was innovative in its mechanism and spraying method. The pulsed oscillating water flow can be formed between the main sprinkler and the auxiliary sprinkler by using the wall-attaching effect of the jet. The step-by-step rotating sprinkler can be realized by intermittent striking of the driving plate by the pulsed jet. The main nozzle of negative pressure feedback jet nozzle provided a long range jet, while the secondary nozzle jet struck the driving plate connected with it, breaking the whole continuous jet into small droplets, providing a jet with a relatively short range, and providing a rotating driving force for the nozzle, thus realizing full circle rotation. Under the inlet pressure of 0.20MPa,0.25MPa, 0.30MPa and 0.35MPa, taking range, average sprinkler irrigation intensity and sprinkler irrigation uniformity as evaluation indexes, the hydraulic performances of six different diameter combinations of main and auxiliary nozzles (3mm×3mm, 4mm×3mm, 4mm×4mm, 5mm×3mm, 5mm×4mm and 5mm×5mm) were compared by weighted scoring method. The results showed that under the above four inlet pressures, the jet nozzles with diameters of 5mm and 4mm had the best comprehensive performance. The water distribution was “triangular”, the range was 13.2~13.7m, the average intensity of sprinkler irrigation was 3.81~4.38mm/h, and the uniformity of sprinkler irrigation was 82.5%~86.0%. The results provided certain theoretical value for studying the hydraulic characteristics of negative pressure feedback nozzle and optimizing the nozzle structure, and brought ideas and applications for the nozzle technical field.

Abstract:In order to explore the effects of different irrigation patterns on agronomic traits, photosynthetic traits, water use efficiency (WUE) and radiation use efficiency (RUE), the experiments included control irrigation, wet irrigation and flood irrigation, which were carried out in Qing’an Irrigation Experimental Center in the semi-humid area of Heilongjiang Province from May to September in 2017 and 2018. During the two growing seasons, the results showed that compared with flood irrigation, the LAI under control irrigation was increased by 7.94%~23.67% (2017) and 5%~14.47% (2018), respectively. Compared with flood irrigation, the cumulative of IPAR under control irrigation and wet irrigation was increased by 74.05μmol/（m2·s）, 23.65μmol/（m2·s） (2017) and 63.35μmol/（m2·s）,16.85μmol/（m2·s） (2018), respectively. Meanwhile, compared with flood irrigation, the accumulation of dry matter under control irrigation was also increased by 12.13% (2017) and 7.98% (2018), respectively. During the rest of the milky stage, the SPAD value of the leaves under control irrigation was significantly higher than that under wet irrigation and flood irrigation. On the other hand, Fv/Fm under the three irrigation methods was increased from tillering stage, and decreased at milky stage after Fv/Fm reached the maximum value at jointing stage and heading stage. Values of qN and qP were fluctuated slightly at early three stage, and then decreased rapidly at milky stage. ETR showed a trend of decreasing at the beginning of tillering stage. The yield under control irrigation was slightly higher than that under flood irrigation. However, the irrigation water consumption under control irrigation was decreased by 31.45% (2017) and 31.67% (2018), respectively, and compared with the flood irrigation, the WUE under control irrigation was increased by 46.45% (2017) and 46.20% (2018), respectively. Additionally, compared with the flood irrigation, the RUE under control irrigation was increased by 1.042% (2017) and 1.036% (2018), respectively. It showed that the control irrigation was the best irrigation methods in the semi-humid area of Heilongjiang Province.

Abstract:Taking the demonstration area of drip irrigation under film as the research object, and aiming at delineating management zones in large areas of farmland, in Huangyanghe Farm, Gansu Province. The topographical attributes (elevation slope and aspect), soil texture (sand, clay and silt), soil moisture content, available nitrogen, electrical conductivity and yield of maize were considered, the degree of variation and correlation of each factor were analyzed, and then the master factors of maize yield were extracted by the results of correlation analysis. Three principal components were obtained by principal component analysis (PCA) based on the master factors. Fuzzy c-means clustering algorithm (FCM) was used to delineate management zones based on the spatial variation of the principal components, the optimal partition number was determined by the fuzzy performance index (FPI), and normalized classification entropy (NCE) were minimum at the same time, and then the differences of the master factors among the management zones were analyzed. Results showed that the master factors were silt, sand, soil water content, available nitrogen, electrical conductivity and elevation, and three management zones were determined by FCM. Statistically significant differences in the master factors were found among the three management zones. Soil water content, crop height and LAI were also significantly different in different management zones during the crop growth period. The spatial variation of the factors within the same management zones was smaller than that of the factors in the whole field, and the variation between zones was large. Delineation of management zones should be adopted based on the characteristics of each zone, and the research result provided a theoretical basis for the implementation of precision agriculture.

Abstract:In order to explore the temporal and spatial variation of surface soil moisture and its influencing factors during the spring snowmelt period in cold regions, the black soil region of Songnen Plain was taken as the research area, remote sensing and GLDAS were used to simulate the assimilation dataset, and the temperature and humidity indexes were used as an auxiliary index to pass the linear regression. Correlation analysis, sliding average and relative contribution rate method were used to analyze the temporal and spatial variation characteristics of surface soil moisture during the snowmelt period, and the specific driving factors of surface soil moisture change was revealed. The results showed that the spatial distribution of soil moisture in the surface of the snowmelt was decreasing in the early stage of snowmelt, and the southward increasing and northward decreasing trend was in the late stage of snowmelt in the past 32 years. On the whole, the surface soil moisture in the early stage of snowmelt was mainly affected by snow, and the latter was mainly related to rainfall and total precipitation changes, the evaporation of temperature in areas with less snow and less rainfall would be obvious;the change of surface soil moisture in the early stage was mainly driven by the trend of temperature change. The trend of years of snow accumulation had a limiting effect on soil moisture change. The trend of rainfall and total precipitation determined the change direction of surface soil moisture in the late stage of snowmelt;from the quantitative contribution results, the change of temperature and snow in the early stage of snowmelt caused the surface soil moisture to be more than doubled, but the positive effect of rainfall trend in the late snowmelt would affect the feedback process of soil moisture warming and drying in the early stage of spring snowing, which would have a positive impact on the water exchange between land and air.

Abstract:Based on the meteorological data, remote sensing data, soil types, geological background and socio-economic data of Chishui River Basin in 2000, 2010 and 2015, the total value of ecosystem services in the basin was quantitatively assessed by carbon sequestration and oxygen release model, modified general soil loss equation (RUSLE) and InVEST water conservation model. At the same time, the ecological compensation quota of different regions in the basin was calculated by using the eco-compensation econometric model, and the compensation standard and priority of the basin were determined. The results indicated that from 2000 to 2015, the value of carbon sequestration and oxygen release per unit area, the value of soil conservation and value of water conservation in Chishui River Basin showed an inverted “V” structure. The value of carbon sequestration and oxygen release was increased from 4279.33 yuan/(hm2·a) in 2000 to 4520.16 yuan/(hm2·a) in 2010, and then decreased to 4409.69 yuan/(hm2·a) in 2015;the value of soil conservation was increased from 95.11 yuan/(hm2·a) in 2000 to 144.45 yuan/(hm2·a) in 2010, and then decreased to 89.03 yuan/(hm2·a) in 2015;and the value of water conservation was decreased from 23938.13 yuan/(hm2·a) in 2000 to 35232.99yuan/(hm2·a) in 2010, and then decreased to 20167.41 yuan/(hm2·a) in 2015. The total value of carbon sequestration and oxygen release in karst area was increased by 4.35 times as fast as that in non-karst area, while the value of soil conservation and water conservation in karst area was decreased by 1.28 and 1.59 times as fast as that in non-karst area, respectively. The value of ecological compensation per unit area was 4626.29 yuan/(hm2·a), and the average value of ecological compensation was 7.623 billion yuan. The value of compensation in the lower reaches of the basin was the largest, while that in the upper reaches was the smallest. The karst area in the upper reaches of the river basin and the primitive forest area in the lower reaches belonged to the typical ecological export area, which should be compensated first. The economically developed middle reaches of the river basin had active economic development level and should take the lead in ecological payment, which belonged to the typical ecological consumption area. Compared with the previous ecological compensation model based on value equivalence, the research quantified the ecological compensation standards of different areas in the basin on grid pixel scale, which can provide scientific basis for the ecological compensation model, operation mechanism and compensation standards of Chishui River Basin in the upper reaches of the Yangtze River.

Abstract:The essence of anaerobic fermentation (AF) is the cultivation process of microorganisms, and the carbon-nitrogen ratio (C/N) is an important factor that affects the production of biogas. To quickly detect the C/N for the AF feedstocks, such as the pretreated corn stover, the mixture of corn stover and feces, a rapid detection model was constructed based on near infrared spectroscopy (NIRS) combined with partial least squares (PLS) regression. To further improve the detection accuracy and efficiency of the model, the genetic simulated annealing interval partial least squares algorithm (GSA-iPLS) and double genetic simulated annealing partial least squares algorithm (DGSA-PLS) based on genetic simulated annealing algorithm (GSA) were proposed for selecting the efficient spectral regions and characteristic wavelength points for NIRS, respectively. Totally 1844 wavelength points of the whole spectrum were selected by GSA-iPLS, and 641 wavelength variables were obtained, and 628 wavelength variables were obtained after the characteristic wavelength points were optimized by DGSA-PLS. The coefficients of multiple determination for prediction (R2p), root mean squared error of prediction (RMSEP) and residual predictive deviation (RPD) in DGSA-PLS regressive model were 0.920, 7.178 and 3.805, respectively. Compared with the whole spectrum model, the RMSEP was decreased by 15.87% in the DGSA-PLS model. It was shown that the number of wavelengths was significantly decreased after the optimization, and the performance of regressive model was obviously higher than that of the whole wavelengths. The research improved the adaptability of the prediction model based on optimizing sensitive wavelength variables for C/N, which provided a new way for directly rapid and accurate measurement of the C/N of AF feedstock.

Abstract:Upper three layers load values of the horizontal and vertical beams were based on the proportional distribution of bearing surface area from scaled model of optimal dome greenhouse, which were stage-by-step static simulated in ANSYS software and compared with snow-loaded simulated strain test. The results showed that micro-strain was linearly related to the loading mass, correlation coefficients in tests was in the range of 0.9932～0.9999 and in simulation was in the range of 0.9948～1, relative errors of micro-strain between simulation and test were in the range of 1.840%～8.386%, which indicated that the simulation method was reliable. Static simulations on one 1ayer with the mesh size of 10mm,12mm, 16mm, 18mm and 20mm were carried out, the results showed when radius was 0.24m, most appropriate mesh size was 16～18mm. In ANSYS, 12 schemes of 6m greenhouse were analyzed step by step through primary selection, linear buckling and mechanical verification (stiffness, strength and stability) by using the same simulation loading calculation method, and optimal scheme was obtained, that was four layers, the first layer had eight beams and mixed bifurcation structure. Buckling simulations on optimal structure of dome greenhouse with radius of 12m and 18m was carried out, which obtained the first-order initial modal defects were obvious under horizontal loads, and under vertical and horizontal loads, non-linear buckling loads were 0.37 and 0.57 times of linear buckling loads on average, which was necessary to carry out nonlinear buckling analysis for long-span dome greenhouse in order to ensure sufficient structural stability. Mechanical verifications were all qualified, stability values of 6m, 12m and 18m greenhouses under combination 2 are 1.89, 2.26 and 2.33 times of that under combination 1, respectively，and there was little difference in strength under two combinations. Compared with 1152m2 Venlo multi-span glass greenhouse, steel consumptions per unit volume of three size greenhouses can be saved by 40.11%～59.34%.

Abstract:Chinese solar greenhouse (hereafter referred to as solar greenhouse) has been widely applied for growing vegetable during winter with little or without additional heating. The traditional back wall could store heat in the daytime and release heat in the solar greenhouse in the night. It plays an important role in maintaining high indoor air temperature in the solar greenhouse. However, this kind of wall has the disadvantages of high cost and uncontrollable exothermic process. The solar greenhouse with the traditional back wall was easy to show low air temperature in the late night. To solve the problems, it was proposed to replace the traditional back wall by employing the heat insulation back wall and the active heat system, which could collect heat in the daytime and release heat actively in night. The active heat system developed with capillary tube mats (AHSCTM) was developed. The heat collecting and releasing performances of AHSCTM was investigated and compared with the composite wall, which was constructed with 370mm clay brick and 100mm polystyrene board, aiming to test the above proposal. AHSCTM was composed of capillary tube mats installed on the back wall or hanged on the south roof of the solar greenhouse, water tank, water pump and pipes. By circulating water, AHSCTM could collect the heat from the solar radiation intercepted by the capillary tube mats in the daytime and stored in the water tank. And in the night when the air temperature was low, those heat could be recovered to heat the solar greenhouse by circulating water again. According to the test, the heat collected in the daytime and released in the nighttime by AHSCTM were 84.4%~111.3% and 74.8%~100.7% of those collected and released by the composite wall, respectively. The coefficient of performance（COP）of AHSCTM was 1.1~2.4. Nevertheless, during operation of AHSCTM in the nights of solar day, cloudy day and overcast day, the heat released by AHSCTM was 98.2%~172.5% of the composite wall. in the other hands, a dynamical model was developed to simulate the water temperature of the water tank. The results showed that the simulated water temperature agreed well with the measured water temperature. This model can be used to improve the heat collecting and releasing performances of AHSCTM. Based on this model, ASHCTM can be improved by lengthening the capillary tube and increasing the number of capillary tubes in each mat. Then, the heat collected in the daytime and released in the nighttime by the improved AHSCTM were 67.6%~112.1% and 69.0%~128.3% of the composite wall respectively. COP of the improved AHSCTM was 2.8~7.0. During operation of AHSCTM in the nights of solar day, cloudy day and overcast day, the heat released by AHSCTM was 2.5~5.1 times of that of the composite wall. Thus, it was feasible to replace the composite wall in the solar greenhouse by employing the improved AHSCTM and the insulation back wall.

Abstract:In order to change the binding polyphenols into free polyphenols in apple juice, improve the content of functional polyphenols, and develop products of apple juice with health care function, Lactobacillus acidophilus 6005, Lactobacillus plants 21805 and Lactobacillus fermentum 21828 were mixed and fermented with apple juice, which was mixed with three varieties of Aksu, Qinguan and Granny Smith. The changes of viable cell numbers and physicochemical composition in apple juice fermentation process were studied. In addition, the kinetic model of bacterial growth and the kinetic model of total phenolic acid incremental change were established and verified. The results showed that probiotics grew well in apple juice, and the number of viable bacteria exceeded 2.68×108CFU/mL. During fermentation, the total sugar content was decreased, the titratable acid content was increased, and the total phenolic acid content was increased. Moreover, the content of several functional phenolic polyphenol was increased, such as chlorogenic acid and gallic acid. Probiotic fermentation changed the quality of apple juice. The kinetic model of bacterial growth and the kinetic model of total phenolic acid change were established. The average error between the theoretical and experimental values of the model was less than 10%. The fitting was good and the established kinetic model could predict the fermentation process. The research results provided a theoretical basis for the development of functional juice beverage products.

Abstract:Aiming at the problem of grain moisture detection equipment not matching and the inability to quickly and accurately determine the moisture content of various coarse cereals. According to international standard for oven method, combined with the theory of grain weightlessness, the transferring multistage type quick measuring moisture content equipment was designed by using TRIZ innovative design method, which realized orderly and reliable operation of loading disc on cylindrical cam track. The simulation design and assembly verification of key components such as multi-station transfer mechanism, automatic weighing system and flash drying mechanism were completed. The PLC coupling control technology was applied to realize the cyclic switching of three working positions: quantitative feeding, flash drying measurement and rapid material removal. Accurate moisture value of grain can be obtained in real time by precise modular operation. The quadratic orthogonal rotational combination design method was used to construct a quantitative model for synergistic evaluation of moisture detection accuracy and detection time of miscellaneous grains. The parameters combination of flash drying temperature, grain size and detection quantity was optimized, and the operation efficiency and detection accuracy of the device were improved. The results showed that the device had good performance, high degree of automation, simple structure and operation, the measurement accuracy, repeatability and reproducibility of the measurement system were reliable, and the relative deviation of detection was less than 0.2%, which provided technical support for the wide application of the rapid moisture meter for the variety of coarse cereals.

Abstract:To quickly and effectively identify water injection meat, the spectral characteristics of water injection and normal meat were analyzed through the hyperspectral analysis technology. The spectrum characteristic parameters of every sample were obtained by using the Fourier transform and statistical calculation. Then the support vector machine (SVM) and neural network (BP) classification models were developed based on the full spectrum, characteristic spectrum and spectrum characteristic parameter, respectively. Finally, the two models were validated by an independent validation set and three indicators such as product’s accuracy (PA), user’s accuracy (UA) and overall accuracy (OA) were used to test the model performance. The results showed that the neural network classification recognition model based on spectrum characteristic parameters had optimal classification recognition rate for water injection pork, with the correct recognition rate of 98.8%. The neural network classification and recognition model based on the characteristic spectrum had the second best recognition, with the correct recognition rate of 96.4%. The classification recognition model of SVM based on full spectrum had the worst classification and recognition for water injection pork, and its correct recognition rate was only 84.5%. These results suggest that hyperspectral technique can be used for rapid and effective detection and identification of watered-down pork.

Abstract:Based on hyperspectral imaging technique, an identification method of star anise and its counterfeit shikimmi was proposed. The hyperspectral data in the range of 400~1000nm were collected and analyzed. Firstly, according to the different spectral characteristics of samples and background pixels, images at 850nm and 450nm were selected and subtracted, and background information was removed by threshold method. Linear stretching method was further used to remove shadow noise pixels due to sample height. Combined with the region labeling method of binary image, the automatic extraction of average spectral data from sample hyperspectral data was realized. Then based on average spectral data, four optimal wavelengths were selected by successive projections algorithm (SPA), i.e., 533nm, 617nm, 665nm and 807nm. Based on the spectral data at the optimal wavelength, a partial least square discrimination analysis (PLSDA) model was established. The classification accuracy of star anise and shikimmi was 98.4%. Using the established multi-spectral model to predict the external validation set data, the overall classification accuracy was 97.9%, and the visualization results were good. Finally, the conventional image processing technology was also used to process the same external verification set data, and the results and advantages of the two methods were compared. The results showed that the multispectral analysis method based on hyperspectral imaging technique was simple, efficient and easy to realize dynamic on-line or portable detection applications. The proposed method can provide a theoretical basis for the development of on-line or portable detection instruments.

Abstract:Effects of ultrasonic treatments with different powers (200W, 400W) for different durations (6min, 12min and 24min) on the stability of recombinant oil emulsions were investigated. Functional properties of the composite system as a function of ultrasonic conditions were explored at the same time. Circular dichroism spectrum, particle size distribution, ζ-potential, FI and emulsibility were determined. The emulsifying properties of the reconstituted oil body emulsion was significantly increased. At the same time, the volume average particle size was decreased from 1717nm to 337.5nm, the absolute value of ζ-potential was increased, and the α-helix content was decreased to 10.4%. And the solution had homogeneous distribution and stable properties. However, with the further increase of ultrasonic power, the surface protein of the recombinant oil body was insoluble, and the functional properties of the recombinant emulsion was decreased. This indicated that the ultrasonic treatment would affect the stability of the recombinant emulsion, and the ultrasonic treatment of suitable strength was beneficial to the formation of the recombinant oil emulsion.

Abstract:Hydraulic excavator driven by an internal-combustion engine is widely used in the construction field. As engine often works in inefficient area, some problems, such as high pollutant emissions, noise and low efficiency exist universally. In order to improve energy efficiency and reduce the emissions, applying hydraulic power source driven by electric motor to hydraulic excavator has become an important research direction. A power source was presented with variable speed induction motor to drive variable displacement pump, which used the fast response characteristic of variable displacement pump to make up for the slow dynamic response of variable speed induction motor, and it was applied to excavator. In order to fully understand the characteristics of the designed electro-hydraulic power source and provide a reference for electric hydraulic excavator design, the dynamic, static and disturbance characteristics of electro-hydraulic power source were studied from both the perspectives of theory and experiment. Then, load disturbance test and arm lifting test were completed on the test rig under the same working conditions. Compared with the characteristics of traditional engine used to drive hydraulic power source system, the new variable speed-variable displacement power source reduced the rotation speed fluctuation by 23.5% in the load disturbance test, the rising time of pump output pressure and flow was reduced by 14.9% and 26.3% in the arm lifting process. The new electro-hydraulic power source owed better characteristics.

Abstract:In the simultaneous localization and mapping (SLAM) technology, mobile robots generally has problems such as low state accuracy, poor stability, and complicated calculation, which can not meet the requirements of real-time and accuracy in the SLAM process. In order to improve this problem, an SLAM autonomous positioning algorithm was proposed based on iterated square root central difference Kalman filter (ISRCDKF). The central difference transform was used to deal with the nonlinear problem of SLAM, avoiding complex operations such as Jacobian matrix in the Taylor formula expansion, and directly transmitting the square root factor reduction algorithm of the covariance matrix in the filter update process. In the complexity, the Levenberg-Marquardt (L-M) optimization method was used to introduce the real-time modified covariance matrix of the adjustment parameters in the iterated observation update process to improve the accuracy and stability of the algorithm. The simulation results showed that under the same data model and noise environment, the proposed ISRCDKF-SLAM algorithm was compared with SLAM algorithm based on extended Kalman filter (EKF-SLAM),SLAM algorithm based on unscented Kalman filter (UKF-SLAM) and SLAM algorithm based on cubature Kalman filter (CKF-SLAM), the root mean square error was reduced by 47.3%, 32.7% and 25.0%, respectively. At the same time, compared with the UKF-SLAM algorithm and CKF-SLAM algorithm with the same computational complexity, the running time of the proposed algorithm was reduced by 15.1% and 10.8%, respectively, which proved the effectiveness of the algorithm. Finally, the proposed algorithm was embedded into the mobile robot platform for field experiment verification, which further proved the practicability and effectiveness of the algorithm.

Abstract:In order to meet the requirement of large stroke and high precision displacement, a giant magnetostrictive linear actuator was designed based on the principle of inchworm cumulative displacement. The front clamp mechanism, the back clamp mechanism and the driving mechanism were given a specific time series of excitation signals, and the step-by-step displacement output of the driver was realized by the cooperation of the three. The superimposed flexure hinge was used as the elastic element to improve the force condition of the flexure hinge effectively. The strength check and modal analysis of flexure hinges were carried out by using finite element method. Simplifying flexible hinges into cantilever beams, the equivalent stiffness of superimposed flexure hinges was calculated. The dynamic model of linear actuator was established based on the voltage law, magnetoresistance theory, linear piezomagnetic model and dynamic theory, and the prototype was tested experimentally. The experimental results and simulative results showed that the calculated model was consistent with the experimental results, and the maximum relative error was 1.86%. The stable working voltage range of the giant magnetostrictive actuator was 1~3V, the minimum single-step displacement was 4.55μm, and the maximum single-step displacement was 12.01μm. The maximum frequency was 150Hz, and the maximum speed was 1.34mm/s. The output state of displacement was stable, and the maximum relative error of single step displacement was 2.69%.

Abstract:An approach for the lightweight design of a 3-R-RS (R and S represented revolute and spherical joints, respectively, and R- denoted the revolute joint is active) parallel mechanism was presented. The stiffness model of the 3-R-RS parallel mechanism was expressed as the stiffness model of two sub-assemblies so as to obtain conclusion that the 3-R-RS parallel mechanism can be regarded as connected by the two sub-assemblies ‘in series’. The stiffness matching coefficients between the two sub-assemblies was obtained by maximizing the lowerorder natural frequencies, and then the rigidity constraints of the 3-R-RS parallel mechanism can be allocated to the two sub-assemblies. On this basis, the lightweight design of the 3-R-RS parallel mechanism can be developed by respectively minimizing the weights of the two subassemblies subject to technological processes, geometric interference constraints and the specified rigidity constraints attributed to them. It should be pointed out that an iteration was adopted to obtain the stiffness matching coefficients, and the response surface fitting was used to fit relationships between the structural parameters of the limb body and its stiffness and mass due to the complexity of the shape of the limb body. The proposed approach simultaneously enabled the parallel mechanism to achieve both high static rigidities and high dynamic behaviors. Numerical example showed that the differences in the stiffness matching coefficient led to the differences in the distribution of the nature frequencies, which instructed that the stiffness matching coefficient had significant influence on the dynamic characteristics.

Abstract:Rotary magnetorheological (MR) brake is a kind of semi-active damping device which can effectively generate torque and dissipate motion energy, it has a wide application prospect in the field of automobile braking systems due to its characteristics of adjustable torque and fast response speed. A rotary MR brake with multi-fluid flow channels was developed and prototyped. The magnetic flux was guided into the outside axial fluid flow channel by setting a non-magnetic material in the middle of the rotating sleeve which could improve the magnetic circuit structure greatly, the working area where the MR brake producing rheological effect was increased, and the effective damping gap of the brake was increased from two sections to four sections. The working principle and torque mathematical model of rotary MR brake were expounded. The electromagnetic field of MR brake was modeled and the distribution of magnetic flux intensity in different fluid flow channels of MR brake was analyzed by finite element method. The simulation results showed that the torque of the MR brake was 65.39N·m when the input current was 1A. The experimental test system was set up to test the dynamic performance of MR brake. The test results showed that the brake torque can reach 61.4N·m with loading current of 1.8A and rotational speed of 600r/min, the variation trend of the experimental results was basically consistent with that of the simulation results, and the speed change had no effect on the torque of rotary MR brake with multi-fluid flow channels, which was beneficial to broaden the application of MR brake.