Crop drying control method and system

Abstract

The invention relates to the technical field of crop drying control, in particular to a crop drying control method and system. The method comprises the following steps that a drying condition target value is set according to crop attributes, and a first drying environment state is obtained through a sensor; establishing a drying environment state prediction model by using the historical data of the first drying environment state, and predicting a second drying environment state according to the drying environment state prediction model; according to the drying condition target value and the first drying environment state, a drying condition difference value is obtained, and the second drying environment state is used for correcting the drying condition difference value; and PID control module parameters are set, and the drying equipment is dynamically adjusted through the PID control module based on the corrected drying condition difference value. By monitoring and adjusting the drying conditions in real time, invalid consumption of energy is avoided, the drying quality of crops is guaranteed, manual operation is reduced, and meanwhile the drying work efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of crop drying control, and particularly to a crop drying control method and system. Background Art

[0002] As a commonly used seasoning, Chinese prickly ash has a unique flavor and aroma and is deeply loved by consumers. Freshly picked Chinese prickly ash has a high water content. If it is not dried in time, it is extremely easy to mildew and rot, thus losing its economic value. At present, there are many problems in the common Chinese prickly ash drying methods. On the one hand, traditional drying methods mostly adopt fixed drying temperature and time settings. This one-size-fits-all drying method easily causes some Chinese prickly ash to be over-dried and cannot be flexibly adjusted according to the actual water content, grain fullness and batch differences of Chinese prickly ash, resulting in some Chinese prickly ash turning black in color, with serious loss of numb taste and aroma during the drying process, greatly reducing the quality of Chinese prickly ash; on the other hand, some Chinese prickly ash may be under-dried with excessive water residue, which is prone to deterioration during subsequent storage. In addition, existing drying equipment often has the problem of uneven temperature distribution during the drying process, making the drying effect of the same batch of Chinese prickly ash uneven, and it is difficult to meet the market demand for high-quality Chinese prickly ash. At the same time, unreasonable drying operations will also cause a large amount of waste of energy and increase production costs. Different types of Chinese prickly ash require different temperatures and drying times at different stages to form a temperature gradient. A reasonable temperature gradient setting can reduce energy consumption, reduce humidity difference, and improve the drying quality of Chinese prickly ash. Therefore, it is urgent to develop a method that can accurately control the drying process, improve the drying quality of Chinese prickly ash and reduce energy consumption. Summary of the Invention

[0003] In view of the deficiencies of the existing methods and the actual application requirements, in order to avoid the ineffective consumption of energy and solve the problem of accurately controlling the Chinese prickly ash drying process to improve the drying quality of Chinese prickly ash. On the one hand, the present invention provides a crop drying control method, including the following steps: setting a target value of drying conditions according to crop attributes, and obtaining the first drying environment state through sensors; establishing a drying environment state prediction model using historical data of the first drying environment state, and predicting the second drying environment state according to the drying environment state prediction model; obtaining a drying condition difference value according to the target value of the drying conditions and the first drying environment state, and correcting the drying condition difference value using the second drying environment state; setting the parameters of the PID control module, and dynamically adjusting the drying equipment through the PID control module based on the corrected drying condition difference value. The present invention comprehensively monitors the drying conditions in real time through the monitoring data of multiple sensors, then corrects the difference value from the target drying conditions and then dynamically adjusts the drying equipment through the PID control module, avoiding the ineffective consumption of energy, being beneficial to ensuring the drying quality of crops, reducing manual operations and improving the drying work efficiency at the same time.

[0004] Optionally, obtaining the first drying environment state through the sensor includes the following steps: Sort the difference values between the monitoring values and the monitoring mean values from small to large, and obtain the difference degree values of the corresponding sensors through the sorting results; use the difference degree values to assign weights to the sensors, and combine the weights and the monitoring values to obtain the first drying environment state. According to the present invention, sorting the difference values between the monitoring values of the sensors and the monitoring mean values can help to remove the influence of outliers, thereby improving the accuracy of the first drying environment state.

[0005] Optionally, obtaining the difference degree value of the corresponding sensor through the sorting result satisfies the following formula: Wherein, represents the difference degree value of the sensor ranked , represents the monitoring value of the sensor ranked , represents the monitoring average value, represents the monitoring value of the sensor ranked , represents the number of sensors.

[0006] Optionally, establishing a drying environment state prediction model using the historical data of the first drying environment state includes the following steps: Introduce an adaptive convergence factor to improve the loss function of the long short-term memory network, and construct a drying environment state prediction model according to the improved long short-term memory network; use the historical data of the first drying environment state and the historical drying equipment parameters to establish a data set, and train and verify the drying environment state prediction model through the data set. According to the present invention, introducing an adaptive convergence factor to improve the loss function of the long short-term memory network can dynamically adjust the weight or learning rate of the loss function according to the actual situation during the training process, which helps the long short-term memory network to converge to the global optimal solution or a local optimal solution close to the global optimal solution more quickly, and improves the training efficiency.

[0007] Optionally, introducing an adaptive convergence factor to improve the loss function of the long short-term memory network satisfies the following formula: Wherein, represents the improved loss function, represents the attenuation coefficient, represents the maximum number of iterations, represents the current iteration number, represents the loss function before improvement.

[0008] Optionally, the step of correcting the drying condition difference value by using the second drying environment state includes the following steps: Fuzzify the target value of the drying condition; combine the fuzzification result and the second drying environment state to correct the drying condition difference value.

[0009] Optionally, the step of combining the fuzzification result and the second drying environment state to correct the drying condition difference value satisfies the following formula: Wherein, represents the corrected drying condition difference value, represents the drying condition difference value before correction, represents the first drying environment state, represents the corresponding fuzzy membership degree of the first drying environment state, represents the target value of the drying condition, represents the second drying environment state, represents the corresponding fuzzy membership degree of the second drying environment state. By fuzzifying the target value of the drying condition, and then correcting the drying condition difference value, the present invention is beneficial to reducing the parameter adjustment amount of the drying equipment, and further reducing the energy consumption.

[0010] Optionally, the step of setting the parameters of the PID control module includes the following steps: Improve the early warning mechanism of the sparrow search algorithm; use the improved sparrow search algorithm to obtain the optimal control parameters of the PID control module.

[0011] Optionally, the improved early warning mechanism of the sparrow search algorithm satisfies the following formula: Wherein, represents the warning value, represents the fitness value of the historical global optimal solution, represents the fitness value of the current global optimal solution, represents the maximum number of iterations, represents the current iteration number, represents the safety threshold. By adaptively adjusting the warning value and the safety threshold, the present invention avoids the limitations of manual setting and improves the search ability of the sparrow in the search space.

[0012] Second aspect, to efficiently execute a crop drying control method provided by the present invention, the present invention further provides a crop drying control system, including a processor, an input device, an output device, and a memory. The processor, the input device, the output device, and the memory are interconnected. Among them, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute a crop drying control method as described in the first aspect of the present invention. The crop drying control system of the present invention has a compact structure and stable performance, and can stably execute a crop drying control method provided by the present invention, further improving the overall applicability and practical application ability of the present invention. Description of the Drawings

[0013] Figure 1 It is a flowchart of a crop drying control method provided by an embodiment of the present invention; Figure 2 It is a framework diagram of a crop drying control system provided by an embodiment of the present invention. Detailed Embodiments

[0014] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described here are only for illustrative purposes and are not used to limit the present invention. In the following description, in order to provide a thorough understanding of the present invention, a large number of specific details are set forth. However, it will be apparent to those of ordinary skill in the art that: it is not necessary to employ these specific details to practice the present invention. In other instances, well-known circuits, software, or methods have not been specifically described in order to avoid obscuring the present invention.

[0015] Throughout the specification, the reference to "one embodiment", "an embodiment", "one example" or "an example" means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment of the present invention. Thus, the phrases "in one embodiment", "in an embodiment", "one example" or "an example" appearing throughout the specification do not necessarily all refer to the same embodiment or example. In addition, the specific features, structures, or characteristics may be combined in any suitable combination and / or sub-combination in one or more embodiments or examples. In addition, those of ordinary skill in the art should understand that the drawings provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0016] Please refer to Figure 1 In order to avoid ineffective consumption of energy and solve the problem of precisely controlling the drying process of Chinese prickly ash to improve the drying quality of Chinese prickly ash. The present invention provides a crop drying control method. As Figure 1 shown, in one embodiment, the method includes the following steps: S1. Set the target value of the drying conditions according to the properties of the crops, and obtain the first drying environment state through sensors.

[0017] In the embodiment, according to the different raw material characteristics and drying requirements of the Chinese prickly ash to be dried, the drying temperature and humidity parameters are set as the target values of the drying conditions.

[0018] Specifically, before the Chinese prickly ash enters the drying equipment, a capacitive moisture sensor is used to detect the moisture content of the Chinese prickly ash sample. By measuring the capacitance change between the Chinese prickly ash and the sensor electrode, the moisture content of the Chinese prickly ash is quickly and accurately calculated; machine vision technology is adopted to take pictures of the Chinese prickly ash through a high-definition camera to obtain the image information of the Chinese prickly ash, and image processing algorithms are used to analyze the characteristics such as the contour, area, and internal texture of the Chinese prickly ash, so as to evaluate the fullness of the Chinese prickly ash grains. Furthermore, according to the moisture content and fullness of the Chinese prickly ash grains and other relevant information, the drying temperature and humidity parameters are set.

[0019] It can be understood that the present invention can also be applied to the drying of other crops such as woody crops, forages, seeds, etc. Further, the present invention can also be used in other occasions where it is necessary to control the temperature and humidity to be constant.

[0020] Further, the obtaining of the first drying environment state through sensors includes the following steps: S11. Sort the difference values between the monitoring values and the monitoring average values from small to large, and obtain the difference degree values of the corresponding sensors through the sorting results.

[0021] First, obtain the monitoring values at the same monitoring moment and obtain the average value of these monitoring values.

[0022] Further, sort the difference values between the monitoring values and the monitoring average values from small to large, that is, the smaller the difference value between the monitoring value and the monitoring average value, the higher the ranking. The smaller the difference value between the monitoring value and the monitoring average value, the more in line with the actual situation the monitoring value of the corresponding monitor is.

[0023] Further, obtain the difference degree values of the corresponding sensors through the sorting results. Specifically, the difference degree values of the corresponding sensors obtained through the sorting results satisfy the following formula: Among them, represents the difference degree value of the sensor ranked , represents the monitoring value of the sensor ranked , represents the monitoring average value, represents the monitoring value of the sensor ranked , represents the number of sensors.

[0024] S12. Use the difference degree value to assign weights to the sensors, and combine the weights and the monitoring values to obtain the first drying environment state.

[0025] Specifically, using the difference degree value to assign weights to the sensors satisfies the following formula: Where, represents the weight of the sensor ranked , represents the difference degree value of the sensor ranked , represents the number of sensors.

[0026] Further, combining the weights and the monitoring values to obtain the first drying environment state satisfies the following formula: Where, represents the first drying environment state, represents the weight of the sensor ranked , represents the monitoring value of the sensor ranked , represents the number of sensors. The first drying environment state refers to the temperature and humidity monitoring values obtained by multiple monitors.

[0027] In some other embodiments, since data acquisition is affected by system noise and high humidity, this may cause the sensor data to change suddenly within an extremely short time, resulting in noise peaks and large data fluctuations. For spike noise, that is, occasional extreme values, Kalman filtering is also required to process abnormal data.

[0028] Kalman filtering is a recursive filter for state estimation. It can effectively process dynamic systems with random noise. The Kalman filter is based on Bayesian filtering theory. By fusing the dynamic model of the system and the observed data, it estimates the state of the system and provides information about the optimal estimate of the state and its uncertainty.

[0029] S2. Use the historical data of the first drying environment state to establish a drying environment state prediction model, and predict the second drying environment state according to the drying environment state prediction model.

[0030] In the embodiment, the step of using the historical data of the first drying environment state to establish a drying environment state prediction model in step S2 includes the following steps: S21. Introduce an adaptive convergence factor to improve the loss function of the long short-term memory network, and construct a drying environment state prediction model according to the improved long short-term memory network.

[0031] The Long Short-Term Memory (LSTM) is a special type of Recurrent Neural Network (RNN) designed to address the vanishing or exploding gradient problems in traditional RNNs when dealing with long sequence data, enabling the network to learn long-term dependencies.

[0032] LSTM introduces three gating mechanisms that allow the network to dynamically decide what information to retain and forget. These three gates are: Forget Gate: The forget gate determines what information to discard from the cell state. It generates a value between 0 and 1 through a sigmoid function, indicating the retention degree of each state value.

[0033] Input Gate: The input gate determines how much new information can be added to the memory cell. It consists of two parts: a sigmoid layer that determines which values will be updated, and a tanh layer that generates a vector of new candidate values. The outputs of the sigmoid layer and the tanh layer of the input gate are multiplied to obtain the updated candidate values.

[0034] Output Gate: The output gate determines how much the output of the memory cell can affect the final output. It determines which cell states will be output through a sigmoid layer, then generates candidate values for the output state through a tanh layer, and finally combines these two parts to form the final output.

[0035] Each cell of LSTM contains the following four main parts: Forgetting Phase: LSTM decides how much information from the previous moment to forget based on the input through the forget gate.

[0036] Input Phase: Decides how much of the current input information needs to be added to the memory through the input gate.

[0037] Update Phase: Combines the forgotten and input information to update the state of the memory cell. This is obtained by adding the output of the forget gate and the output of the input gate.

[0038] Output Phase: Decides what to output based on the updated state of the memory cell through the output gate.

[0039] Further, to construct a drying environment state prediction model based on a long short-term memory network, it is necessary to set parameters in multiple aspects, including the input feature dimension, the number of hidden layer features, the number of LSTM stacking layers, the bias term, the batch processing dimension setting, the Dropout layer, the bidirectional LSTM, as well as the initial hidden state and cell state, and the loss function and the internal optimizer of the model.

[0040] The loss function is a function used in machine learning and deep learning to measure the difference between the model's prediction results and the true labels. It is the optimization objective of the model. By minimizing the loss function, the model parameters are adjusted so that the model can better fit the training data. The internal optimizer is an algorithm used to minimize the loss function and update the internal parameters of the model. It adjusts the parameters according to the training data and the internal parameters of the model to minimize the loss function, thereby improving the performance of the model.

[0041] Common loss functions include the mean squared error loss function, the mean absolute error loss function, and the cross-entropy loss function. However, these functions cannot adaptively adjust the weight or learning rate of the loss function according to the actual situation during the training process. It can be understood that introducing an adaptive convergence factor to improve the loss function of the long short-term memory network helps the LSTM network converge to the global optimal solution or a local optimal solution close to the global optimal solution more quickly, improving the training efficiency.

[0042] Further, the improved loss function of the long short-term memory network by introducing an adaptive convergence factor satisfies the following formula: where represents the improved loss function, represents the decay coefficient, represents the maximum number of iterations, represents the current iteration number, represents the loss function before improvement. The decay coefficient can be set according to the learning situation. In this embodiment, the mean squared error loss function is used as the loss function, .

[0043] S22. Establish a data set using the historical data of the first drying environment state and the historical drying equipment parameters, and train and verify the drying environment state prediction model through the data set.

[0044] Specifically, the drying equipment includes devices such as a fan and a steam regulating valve for controlling and adjusting the temperature and humidity. By controlling the frequency conversion of the fan, the wind speed and wind pressure during drying are changed, and the opening size of the steam regulating valve is adjusted, thereby controlling and adjusting the temperature and humidity changes.

[0045] Further, a data set is established using the historical data of the first drying environment state and the historical drying equipment parameters. The data is divided into a training set and a test set. The training set is used to train the model, and the test set is used to evaluate the model performance. 90% of the data is used as the training set, and the training set is normalized to reduce the impact of different eigenvalue ranges on the model. The standard score function is used to calculate the mean and standard deviation of the training data, and these parameters are applied to normalize the test data.

[0046] S3. Obtain a drying condition difference value according to the target value of the drying condition and the first drying environment state, and correct the drying condition difference value by using the second drying environment state.

[0047] Specifically, the difference between the target value of the drying condition and the first drying environment state is the obtained drying condition difference value.

[0048] Further, the step of correcting the drying condition difference value by using the second drying environment state in step S3 includes the following steps: S31. Fuzzify the target value of the drying condition.

[0049] In the embodiment, the target value of the drying condition is the optimal temperature and humidity value for drying Chinese prickly ash. However, when drying Chinese prickly ash, there is little difference in the effect between being in a preferred temperature and humidity range and being at the optimal temperature and humidity value. Therefore, controlling the temperature and humidity within the preferred temperature and humidity range during the drying of Chinese prickly ash can better reduce energy consumption.

[0050] Further, perform fuzzification according to the preferred temperature and humidity range and the optimal temperature and humidity value, satisfying the following formula: Where represents the fuzzy membership degree of the temperature and humidity value, represents the monitoring value of the th sensor, represents the optimal temperature and humidity value, represents the maximum value of the temperature and humidity range, represents the minimum value of the temperature and humidity range.

[0051] S32. Combine the fuzzification result and the second drying environment state to correct the drying condition difference value.

[0052] Specifically, combining the fuzzification result and the second drying environment state to correct the drying condition difference value satisfies the following formula: Where represents the corrected drying condition difference value, Denote the difference value of the drying conditions before correction, Denote the first drying environment state, Denote the corresponding fuzzy membership degree of the first drying environment state, Denote the target value of the drying conditions, Denote the second drying environment state, Denote the corresponding fuzzy membership degree of the second drying environment state.

[0053] S4. Set the parameters of the PID control module. Based on the corrected difference value of the drying conditions, dynamically adjust the drying equipment through the PID control module.

[0054] In the control of Chinese prickly ash drying, the constant temperature control of multiple temperature zones requires the accurate opening of the steam regulating valve. After adding the feedback signal of the temperature sensor, the opening of the steam regulating valve can be controlled through PID to achieve closed-loop control, making the drying accuracy of multiple temperature zones higher. In addition, for humidity control, precise control of the exhaust fan is also required. According to the difference between the value collected by the humidity sensor and the set value of the PLC program, the start-stop and frequency of the exhaust fan are controlled through PID operation.

[0055] The PID (Proportion Integration on Differentiation, PID) control method determines the new input value by comparing the input data with the pre-set reference data to maintain the stable control of the system. The PID control takes the system deviation as the input, and calculates the control quantity through three links: proportional control, differential control and integral control, and acts on the controlled object.

[0056] The realization of PID control for regulating the controlled object is inseparable from the close cooperation among the proportional link, integral link and differential link. Among them, the control key points of the three links are as follows: Proportional link. Generally speaking, the proportional controller is equivalent to an amplifier, which is used to reflect the deviation between the measured value and the given value. After the deviation occurs, the controller immediately adjusts to reduce the deviation. Reasonably setting the proportional control coefficient Kp can, to a certain extent, eliminate the interference of the system.

[0057] Integral link. The integral link of the controller uses the control coefficient Ki to reduce the error generated between the true value and the pre-set value of the system, so as to solve the steady-state error problem of the system and improve the control effect.

[0058] Differential link. The differential control link is mainly used to overcome the lag of the controlled object, reduce the oscillation in the control process, and improve the predictability of the system. In PID control, the differential link is generally independent of the magnitude of the deviation, but is proportional to the speed and time of the deviation change.

[0059] Further, setting the parameters of the PID control module, that is, setting the proportional coefficient, integral coefficient, and differential coefficient of the PID control module, includes the following steps: S41. Improve the early warning mechanism of the sparrow search algorithm.

[0060] The advantages of the sparrow search algorithm are its high efficiency, flexibility, low memory occupancy, and easy implementation. It is especially suitable for large-scale search problems. Starting from the initial solution, it continuously performs local searches, selects local operations that can be improved each time, and updates the current solution. The algorithm selects the optimal local operation in the current state at each step until a certain stopping condition is reached. In addition, it also has the characteristics of high convergence accuracy, fast convergence speed, and strong robustness, and is superior to other swarm intelligence algorithms in function optimization problems. The core idea of the sparrow is to continuously search for the optimal solution of the objective function through local operations and can find an approximate optimal solution in a short time. However, the sparrow algorithm also has some limitations, such as being easily trapped in local optimal solutions and having certain restrictions on the solution space of the problem.

[0061] Based on this, the present invention improves the early warning mechanism of the sparrow search algorithm. The improved early warning mechanism of the sparrow search algorithm satisfies the following formula: where, represents the warning value, represents the fitness value of the historical global optimal solution, represents the fitness value of the current global optimal solution, represents the maximum number of iterations, represents the current iteration number, represents the safety threshold. By adaptively adjusting the warning value and safety threshold according to the iteration situation, the limitations of manual setting are avoided, and the search ability of the sparrow in the search space is improved.

[0062] S42. Use the improved sparrow search algorithm to obtain the optimal control parameters of the PID control module.

[0063] Specifically, using the improved sparrow search algorithm to obtain the optimal control parameters of the PID control module includes the following steps: Initialize the population. According to the set population size, randomly generate a set of PID parameters as the initial population; Set the fitness evaluation function, evaluate the fitness of each individual, and calculate its fitness value; Update the discoverer. According to the fitness value, select the individual with a higher fitness as the discoverer. The discoverer has a larger search range in the search space and is responsible for exploring new solution spaces; The joiner updates, and the remaining individuals act as joiners. They are updated according to the position and behavior of the discoverer. The joiners can follow the discoverer to forage or compete with the discoverer for food resources, thereby improving their fitness; Reconnaissance and early warning mechanism. Introduce a reconnaissance and early warning mechanism, select a certain proportion of individuals for reconnaissance and early warning to avoid falling into a local optimum. When the fitness value of the discoverer or joiner has not been improved for a long time, trigger the reconnaissance and early warning mechanism to re-initialize the positions of these individuals.

[0064] Iterative update. Continuously iterate and update the individuals in the population until the maximum number of iterations is reached or other stopping conditions are met.

[0065] Furthermore, based on the corrected drying condition difference value, the drying equipment is dynamically adjusted by the PID control module.

[0066] In the drying equipment, the PID control module is combined with devices such as the actuator to form a closed-loop control system. The PID control module calculates the control output according to the corrected drying condition difference value, and the actuator dynamically adjusts the parameters of the drying equipment according to the control output.

[0067] Please refer to Figure 2 , in the embodiment, in order to efficiently execute a crop drying control method provided by the present invention, the present invention also provides a crop drying control system, including: an input device, an output device, a processor, and a memory. The input device, output device, processor, and memory are interconnected. The memory contains program instructions, and the program instructions are used for the steps of the crop drying control method. The crop drying control system of the present invention has a compact structure and stable performance, can stably execute a crop drying control method of the present invention, and further improves the overall applicability and practical application ability of the present invention.

[0068] In an embodiment, the so-called processor may be a Central Processing Unit (CPU), and the processor may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The input device may be used to obtain data information. The output device may be used to output the result obtained from the program instructions included in the computer program stored in the memory provided by the present invention. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include a non-volatile random access memory.

[0069] In a possible implementation manner, the memory may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function, etc.; the data storage area may store the data created during use. In addition, the memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include NVRAM. The memory stores an operating system, operation instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof. Among them, the operation instructions may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0070] The embodiment also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned crop drying control method are implemented.

[0071] The storage medium may include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs.

[0072] In summary, the present invention monitors the drying conditions in real time by integrating the monitoring data of multiple sensors, corrects the difference value from the target drying conditions, and then dynamically adjusts the drying equipment through the PID control module, avoiding the ineffective consumption of energy, being beneficial to ensuring the drying quality of crops, reducing manual operations and improving the drying work efficiency at the same time.

[0073] Therefore, the present invention effectively overcomes various drawbacks in the prior art and has high industrial utilization value.

[0074] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope recorded in the present invention.

Claims

1. A method for controlling the drying of crops, characterized in that, Including the following steps: Set the target value of the drying condition according to the crop attributes, and obtain the first drying environment state through sensors; Establish a drying environment state prediction model using the historical data of the first drying environment state, and predict the second drying environment state according to the drying environment state prediction model; Obtain the drying condition difference value according to the target value of the drying condition and the first drying environment state, and correct the drying condition difference value using the second drying environment state; Set the parameters of the PID control module, and dynamically adjust the drying equipment through the PID control module based on the corrected drying condition difference value.

2. The crop drying control method according to claim 1, characterized in that The obtaining of the first drying environment state through sensors includes the following steps: Sort the difference values between the monitoring values and the monitoring mean values from small to large, and obtain the difference degree values of the corresponding sensors through the sorting results; Allocate weights to the sensors using the difference degree values, and combine the weights and the monitoring values to obtain the first drying environment state.

3. The crop drying control method according to claim 2, characterized in that The obtaining of the difference degree values of the corresponding sensors through the sorting results satisfies the following formula: Among them, represents the difference degree value of the th sensor, represents the monitoring value of the th sensor, represents the monitoring average value, represents the monitoring value of the th sensor, represents the number of sensors.

4. The crop drying control method according to claim 1, characterized in that The establishing of the drying environment state prediction model using the historical data of the first drying environment state includes the following steps: Introduce an adaptive convergence factor to improve the loss function of the long short-term memory network, and construct a drying environment state prediction model according to the improved long short-term memory network; Establish a data set using the historical data of the first drying environment state and the historical drying equipment parameters, and train and verify the drying environment state prediction model through the data set.

5. The crop drying control method according to claim 4, wherein, The introducing of the adaptive convergence factor to improve the loss function of the long short-term memory network satisfies the following formula: Among them, represents the improved loss function, represents the attenuation coefficient, represents the maximum number of iterations, represents the number of iterations for the current time, represents the loss function before improvement.

6. The crop drying control method according to claim 1, characterized in that, The correcting of the drying condition difference value using the second drying environment state includes the following steps: Fuzzify the target value of the drying condition; Combine the fuzzification result and the second drying environment state to correct the drying condition difference value.

7. The crop drying control method according to claim 6, characterized in that, The combining of the fuzzification result and the second drying environment state to correct the drying condition difference value satisfies the following formula: Among them, represents the difference value of the drying conditions after correction, represents the difference value of the drying conditions before correction, represents the first drying environment state, represents the corresponding fuzzy membership degree of the first drying environment state, represents the target value of the drying conditions, represents the second drying environment state, represents the corresponding fuzzy membership degree of the second drying environment state.

8. The crop drying control method according to claim 1, characterized in that, The setting of the parameters of the PID control module includes the following steps: Improve the early warning mechanism of the sparrow search algorithm; Use the improved sparrow search algorithm to obtain the optimal control parameters of the PID control module.

9. The crop drying control method according to claim 8, wherein, The improving of the early warning mechanism of the sparrow search algorithm satisfies the following formula: Among them, represents the warning value, represents the fitness value of the historical global optimal solution, represents the fitness value of the current global optimal solution, represents the maximum number of iterations, represents the current number of iterations, represents the safety threshold.

10. A crop drying control system, characterized in that, The crop drying control system includes: an input device, an output device, a processor, and a memory. The input device, the output device, the processor, and the memory are interconnected. The memory includes program instructions for executing the crop drying control method according to any one of claims 1-9.

Images