Applications of Computer Science in Agricultural Engineering

Agricultural engineering is the branch of engineering that deals with the design and exploitation of farm machinery and devices; the location and planning of farm structures; farm drainage, soil management, and erosion control; water supply and irrigation; rural electrification; farm product processing; and deriving renewable energy from agricultural products [...] Full article

Taking an intelligent trimming device hydraulic manipulator as the research object, aiming at the uncertainty, nonlinearity and complexity of its system, a trajectory tracking control scheme is studied in this paper. In light of the virtual work principle, a coupling dynamic model of the hydraulic system and manipulator system is established. In order to improve the anti-interference and adaptive abilities of the manipulator system, a compound control strategy combining the adaptive neuro-fuzzy inference system (ANFIS) and proportional integral derivative (PID) controller is proposed. The neural adaptive learning algorithm is utilized to train the given input and output data to adjust the membership functions of the fuzzy inference system, then the PID parameters can be adjusted adaptively to accomplish trajectory tracking. Based on MATLAB/Simulink, the simulation model is established. In addition, to prove the effectiveness of the ANFIS-based PID controller (ANFIS-PID), its performance is compared with PID and fuzzy PID (FPID) controllers. The simulation results indicate that the ANFIS-PID controller is superior to the other controllers in control effect and control precision, and provides a more accurate and effective method for the control of agriculture. Full article

In this paper, a semi-Markov model for determining the optimal time for preventive replacements according to the age of technical objects is presented. In the analyzed system of transportation, due to its specific characteristics, the basic type of renewal process carried out is minimal repair. Minimal repairs of technical objects in semi-Markov models have been analyzed in the literature to date. In the system studied, the technical objects (sets of agricultural tractors with trailers), due to the continuous operation of combine harvesters, should carry out the assigned tasks of transporting agricultural crops without interruption. The damage to agricultural tractors that arises during the implementation of transport tasks should be repaired in the shortest possible time. The repairs to damaged tractors are carried out primarily by the Technical Emergency Service and, due to their purpose and scope, may be considered minimal repairs. The effectiveness of the function of the tested technical objects is analyzed by two criteria functions, which are very important for the system managers. These are profit per unit of time and availability. In the analyzed case, it is the availability to carry out the assigned transport tasks. The conditions for the existence of a maximum of criterion functions have been written for the assumptions. The analyzes carried out, which are presented in the work, are illustrated with sample calculations. It has been proven that, under general assumptions, the criterion functions considered in the paper have exactly one maximum. On the basis of the conducted analysis, sufficient conditions for the existence of a maximum of these functions were formulated. In the analyzed transport system, it is possible to increase the efficiency of the function of the technical facilities in use as a result of planning additional preventive replacements (increasing the frequency of these replacements). This is especially important for a system where transport units must have high availability. Full article

Pests and diseases are an inevitable problem in agricultural production, causing substantial economic losses yearly. The application of convolutional neural networks to the intelligent recognition of crop pest images has become increasingly popular due to advances in deep learning methods and the rise of large-scale datasets. However, the diversity and complexity of pest samples, the size of sample images, and the number of examples all directly affect the performance of convolutional neural networks. Therefore, we designed a new target-detection framework based on Cascade RCNN (Regions with CNN features), aiming to solve the problems of large image size, many pest types, and small and unbalanced numbers of samples in pest sample datasets. Specifically, this study performed data enhancement on the original samples to solve the problem of a small and unbalanced number of examples in the dataset and developed a sliding window cropping method, which could increase the perceptual field to learn sample features more accurately and in more detail without changing the original image size. Secondly, combining the attention mechanism with the FPN (Feature Pyramid Networks) layer enabled the model to learn sample features that were more important for the current task from both channel and space aspects. Compared with the current popular target-detection frameworks, the average precision value of our model ([email protected]) was 84.16%, the value of ([email protected]:0.95) was 65.23%, the precision was 67.79%, and the F1 score was 82.34%. The experiments showed that our model solved the problem of convolutional neural networks being challenging to use because of the wide variety of pest types, the large size of sample images, and the difficulty of identifying tiny pests. Full article

Convective drying is the most common drying method, and mathematical modelling of the dewatering process is an essential part of it, playing an important role in the development and optimization of drying devices. Modelling of the leek drying process can be difficult as the specific structure of this vegetable, in which the slices of leek are delaminated into uneven single rings at different times during drying and the material surface changes more than in other vegetables. This study aimed at proposing a theoretical model for leek convective drying, based on the theoretical laws of heat and mass exchange, which should take into account the observed random process disturbances in the form of random coefficients of this model. The paper presents a non-linear model of water content changes with a random coefficient n. Values of the coefficient n, which were considered to be a random variable, were obtained using the Monte Carlo method, using the inversed distribution function as a probabilistic method. The non-linear model of water content changes when a random n coefficient gives a good approximation of the measurements of water content changes to approximately 1–2 kg H₂O/kg d.m. Full article

(1) Background: The thermal environment in a pregnant sow piggery is affected by physical parameters such as air temperature, relative humidity, and airflow velocity. However, it is challenging to conduct experimental studies due to the high cost. (2) Methods: Computational fluid dynamics (CFD) was used to study the distribution characteristics of airflow in a pregnant sow piggery with centralized ventilation. (3) Results: The results show that the maximum difference between the simulated and experimental temperature was less than 1.54 °C, and the simulated and tested relative humidity difference was less than 10% RH. Incorporation of a middle air outlet is beneficial for increasing the uniformity of temperature distribution, as studied by comparing the temperature and humidity uniformity coefficient of the two air outlet locations, but the uniformity of humidity distribution will be reduced. With an increase in velocity, the temperature shows a downward trend and the relative humidity shows an upward trend. (4) Conclusions: The most suitable position for the outlet is the middle, with an associated airflow velocity of 0.5 m/s. This study revealed the variation in flow field distribution and air distribution in the pregnant sow piggery as a consequence of changes in ventilation structure, which has certain significance as a reference for the optimization of airflow in intensive pregnant sow piggeries. Full article

The rotary tillage knife roller, as one of the typical soil-touching parts of the tillage equipment cutting process, is in direct contact with the soil. During the cutting process, there are problems related to structural bending, deformation, and high power consumption, caused by impact and load, and it is difficult to observe the micro-change law of the rotary tillage tool and soil. In view of the above problems, we took the soil of the cotton experimental field in Shihezi, Xinjiang, and the soil-contacting parts of the rotary tillage equipment, specifically the rotary tiller roller, as the research subject. Using the finite-element method (FEM) to simulate the structure of the rotary tiller with different bending angle parameters, we obtained its average stress and deformation position information, and obtained a range linear relationship between the bending angle and the structural performance of the rotary tiller tool. Using discrete element method (DEM)-based simulation to build the corresponding contact model, soil particle model, and soil–rotary tillage knife roll interaction model to simulate the dynamic process of a rotary tillage knife roll cutting soil, we obtained the change rules of the soil deformation area, cutting process energy, cutting resistance, and soil particle movement. By using the orthogonal simulation test and the response surface method, we optimized the kinematic parameters of the rotary tiller roller and the key design parameters of a single rotary tiller. Taking the reduction of cutting power consumption as the optimization goal and considering the influence of the bending angle on its structural performance, the optimal parameter combination was obtained as follows: the forward speed was 900 m/h, the rotation speed was 100 rad/min, the bending angle was 115°, and the minimum power consumption of the cutter roller was 0.181 kW. The corresponding average stress and deformation were 0.983 mm and 41.826 MPa, which were 15.8%, 13%, and 7.9% lower than the simulation results of power consumption, stress, and deformation under the initial parameter setting, respectively. Finally, the effectiveness of the simulation optimization model in reducing power consumption and the accuracy of the soil-cutting simulation were verified by a rotary tilling inter-field test, which provided theoretical reference and technical support for the design and optimization of other typical soil-touching parts of tillage and related equipment, such as disc harrow, ploughshare, and sub-soiling shovel. Full article

Remote sensing estimation of crop nitrogen content allows real-time monitoring of growth to develop scientific methods. However, most of the current remote sensing estimates of crop nitrogen contents have limitations in accurately reflecting the vertical distribution of nutrients in plants. Firstly, the original hyperspectrum is first-order differential (FD), second-order differential (SD), and continuous removal (CR), and the corresponding sensitive bands were screened by correlation analysis in this paper. Then, the spectral reflectance, vegetation indices, and wavelet coefficients were used as input features to construct models for estimating nitrogen content of flag leaf, upper 1 leaf, upper 2 leaf, upper 3 leaf, and upper 4 leaf based on partial least squares regression (PLSR), support vector machine (SVM), random forest (RF), and multiple linear regression (MLR), respectively. The results showed that the accuracy of nitrogen content prediction based on wavelet coefficients was the highest. The combination of MLR and SVM with wavelet coefficients had high accuracy and robustness in the prediction of nitrogen content at different leaf positions. Additionally, the prediction accuracy of nitrogen gradually increased as the leaf position of winter wheat increased. The study can provide technical support for remote sensing estimation of nutrient elements at vertical leaf position of crops. The study can provide a reference for prediction of other crops. Full article

The mathematical description of changes of dried apples characteristics (mass gain, volume increase, dry matter loss, rehydration indices, and colour) during their rehydration was performed. The effect of conditions of both processes on model parameters were also considered. Apple slices (3 and 10 mm) and cubes (10 mm) were dried in natural convection (drying air velocity 0.01 m/s), forced convection (0.5 and 2 m/s), and fluidisation (6 m/s). Drying air temperatures (T_d) were equal to 50, 60, and 70 °C. The rehydration process was carried out in distilled water at the temperatures (T_r) of 20, 45, 70, and 95 °C. Mass gain, volume increase, and dry matter loss were modelled using the following empirical models: Peleg, Pilosof–Boquet–Batholomai, Singh and Kulshrestha, Lewis (Newton), Henderson–Pabis, Page, and modified Page. Colour changes were described through applying the first-order model. Artificial neural networks (feedforward multilayer perceptron) were applied to make the rehydration indices and colour variations (ΔE) dependent on characteristic dimension, T_d, drying air velocity, and T_r. The Page and the modified Page models can be considered to be the most appropriate in order to characterise the mass gain (RMSE = 0.0143–0.0619) and the volume increase (RMSE = 0.0142–0.1130), whereas the Peleg, Pilosof–Bouquet–Batholomai, and Singh and Kulshrestha models were found to be the most appropriate to characterise dry matter loss (RMSE = 0.0116–0.0454). The ANNs described rehydration indices and ΔE satisfactorily (RMSE = 0.0567–0.0802). Both considered process conditions influenced (although in different degree) the changes of the considered dried apple characteristics during their rehydration. Full article

Buckwheat is an important cereal crop with high nutritional and health value. Buckwheat disease greatly affects the quality and yield of buckwheat. The real-time monitoring of disease is an essential part of ensuring the development of the buckwheat industry. In this research work, we proposed an automated way to identify buckwheat diseases. It was achieved by integrating a convolutional neural network (CNN) with the image processing technology. Firstly, the proposed approach would detect the buckwheat disease area accurately. Then, to improve the accuracy of classification, a two-level inception structure was added to the traditional convolutional neural network for accurate feature extraction. It also helps to handle low-quality image problems, which includes complex imaging environment and leaf crossing in sampling buckwheat image, etc. At the same time, instead of the traditional convolution, the convolution based on cosine similarity was adopted to reduce the influence of uneven illumination during the imaging. The experiment proved that the revised convolution enabled better feature extraction within samples with uneven illumination. Finally, the experiment results showed that the accuracy, recall, and F1-measure of the disease detection reached 97.54, 96.38, and 97.82%, respectively. For identifying disease categories, the mean values of precision, recall, and F1-measure were 84.86, 85.78, and 85.4%. Our method has provided important technical support for realizing the automatic recognition of buckwheat diseases. Full article

Understanding biological materials is quite complicated. The material apple pomace is biologically unstable has been dried under certain conditions. Modeling the pomace drying is necessary to understand the heat and mass transport mechanism and is a prerequisite for the mathematical description of the entire process. Such a model plays an important role in the optimization or control of working conditions. Modeling of the pomace drying process is difficult as apple pomace is highly heterogeneous, as it consists of flesh, seeds, seed covers, and petioles of various sizes, shapes and proportions. A simple mathematical model (Page) was used, which describes well the entire course of the drying process. This is used to control the process. In turn, complex mathematical models describe the phenomena and scientifically explain the essence of drying. Mathematical modeling of the dewatering process is an indispensable part of the design, development and optimization of drying equipment. Full article

In this study, the potential of the biospeckle phenomenon for detecting fruit infestation by Drosophila suzukii was examined. We tested both graphical and analytical approaches to evaluate biospeckle activity of healthy and infested fruits. As a result of testing the qualitative approach, a generalized difference method proved to be better at identifying infested areas than Fujii’s method. Biospeckle activity of healthy fruits was low and increased with infestation development. It was found that the biospeckle activity index calculated from spatial-temporal speckle correlation of THSP was the best discriminant of healthy fruits and fruits in two different stages of infestation development irrespective of window size and pixel selection strategy adopted to create the THSP. Other numerical indicators of biospeckle activity (inertia moment, absolute value of differences, average differences) distinguished only fruits in later stage of infestation. Regular values of differences turned out to be of no use in detecting infested fruits. We found that to provide a good representation of activity it was necessary to use a strategy aimed at random selection of pixels gathered around the global maximum of biospeckle activity localized on the graphical outcome. The potential of biospeckle analysis for identification of highbush blueberry fruits infested by D. suzukii was confirmed. Full article

Fruit and vegetable processing has a significant impact on the environment due to its consumption of a significant amount of water. Water consumption mainly depends on the type of production and the technology used. Water in fruit and vegetable processing plants is used as a raw material, an energy carrier, and in hydro transport, as well as for washing raw materials and maintaining production hygiene. The variety of technological operations carried out in the production process and the seasonality of production make it difficult to objectively assess the use of water in fruit and vegetable processing plants. Few available publications in this field provide numerical values of water unit consumption indices, with none entering into the cause-and-effect relationships of water use in plants in this industry. The aim of this study was to analyze the research to date and to verify the following research hypothesis: the structure of processing and the relationship between the weights of individual products have an impact on water consumption in fruit and vegetable processing plants. For this purpose, neural models of water consumption were developed for the largest agri-food processing plants in Poland that use similar technology. Water consumption was then optimized using genetic algorithms for the processing structure. The results confirmed the hypothesis that production structure has a significant impact on the rationalization of water consumption. The optimization results show that the production of concentrates, juices, and drinks has the greatest impact on water consumption. The lowest water consumption will be achieved when the production of concentrates is at a 2 to 1 ratio to the production of juices and drinks. Full article

Deep learning proves its promising results in various domains. The automatic identification of plant diseases with deep convolutional neural networks attracts a lot of attention at present. This article extends stochastic gradient descent momentum optimizer and presents a discount momentum (DM) deep learning optimizer for plant diseases identification. To examine the recognition and generalization capability of the DM optimizer, we discuss the hyper-parameter tuning and convolutional neural networks models across the plantvillage dataset. We further conduct comparison experiments on popular non-adaptive learning rate methods. The proposed approach achieves an average validation accuracy of no less than 97% for plant diseases prediction on several state-of-the-art deep learning models and holds a low sensitivity to hyper-parameter settings. Experimental results demonstrate that the DM method can bring a higher identification performance, while still maintaining a competitive performance over other non-adaptive learning rate methods in terms of both training speed and generalization. Full article

Photosynthesis is a process that indicates the productivity of crops. The estimation of this variable can be achieved through methods based on mathematical models. Mathematical models are usually classified as empirical, mechanistic, and hybrid. To mathematically model photosynthesis, it is essential to know: the input/output variables and their units; the modeling to be used based on its classification (empirical, mechanistic, or hybrid); existing measurement methods and their invasiveness; the validation shapes and the plant species required for experimentation. Until now, a collection of such information in a single reference has not been found in the literature, so the objective of this manuscript is to analyze the most relevant mathematical models for the photosynthesis estimation and discuss their formulation, complexity, validation, number of samples, units of the input/output variables, and invasiveness in the estimation method. According to the state of the art reviewed here, 67% of the photosynthesis measurement models are mechanistic, 13% are empirical and 20% hybrid. These models estimate gross photosynthesis, net photosynthesis, photosynthesis rate, biomass, or carbon assimilation. Therefore, this review provides an update on the state of research and mathematical modeling of photosynthesis. Full article