Feedback control-based drip irrigation fertilization method and system for field potatoes

Abstract

The present invention provides a field potato drip fertilization method and system based on feedback control, relating to the technical field of agricultural irrigation. The present invention forms an individual vector from the gain of a PID controller, generates an initial population, constructs and trains an optimal compound fertilizer flow setting model, intelligently determines a target compound fertilizer flow setting value according to potato tuber feedback data through the compound fertilizer flow setting model, and constructs a first fitness function in combination with the actual compound fertilizer flow output of the system; selects individuals in the initial population based on the fitness function for a splicing and crossover operation to obtain spliced ​​and crossover individuals; performs equivalence on a signal transmission process of a motor system, and establishes a second fitness function based on the equivalence result; recombines the spliced ​​and crossover individuals into 8 individuals, selects recombinations based on the value of the second fitness function, places the selected recombined individuals in an iterative population, and continuously iterates to complete potato drip fertilization control.

Description

Technical Field

[0001] The invention relates to the technical field of agricultural irrigation, in particular to a field potato drip fertilization method based on feedback control. Background Art

[0002] The potato growth and maturation process consists of five stages: germination, seedling, tuber formation, tuber expansion, and maturity. The tuber formation stage determines the number of tubers each potato plant will produce. Compound fertilizers provide nutrients such as nitrogen, phosphorus, and potassium. Phosphorus promotes energy metabolism and cell division, nitrogen promotes plant growth, and potassium improves tuber quality. These synergistic effects contribute to high-quality potato yields. Potatoes require different ratios of nitrogen, phosphorus, and potassium at different growth stages. Phosphorus is the primary requirement during the germination stage, while nitrogen promotes growth during the seedling stage. Potassium and compound fertilizers are required during the tuber expansion stage. Traditional fertilization systems generally rely on PID algorithms to regulate compound fertilizer flow. However, its limitations are significant. First, PID parameter tuning often relies on empirical methods such as the critical proportion method, which is cumbersome and difficult to obtain the optimal parameter combination. Second, traditional PID cannot effectively handle the large time lag, time-varying, and nonlinear characteristics of fertilization systems, making it difficult to effectively integrate with actual fertilization feedback during the tuber formation period. Furthermore, it lacks the adaptive ability to dynamically adjust fertilization targets based on plant physiological status. Furthermore, precise control of compound fertilizer flow is a complex system involving complex dynamic processes and signal transmission. This not only involves the PID controller's own adjustment capabilities, but more critically, the actual response characteristics of the compound fertilizer flow control process: first, the drive frequency must be adjusted by the frequency converter, which then activates the motor to provide power, ultimately ensuring a stable flow of fertilizer liquid from the fertilization device outlet. This makes it difficult for traditional compound fertilizer flow feedback control based on static PID parameters to effectively compensate for the interference caused by these combined factors, resulting in large systematic errors between the actual compound fertilizer flow and the set value. This ultimately leads to inaccurate fertilization, waste of resources, and potential environmental risks.

[0003] In the prior art, publication number CN119689839A discloses a liquid fertilizer control method based on EDBKA-PID, a soil testing and formula fertilization system and method. By making multiple processes of the system flow transfer process equivalent and establishing corresponding origin functions, the system error is calculated based on the corresponding transfer function, and the system error is adjusted by the PID algorithm to achieve the problem of accurately controlling the fertilizer flow rate. However, the prior art still has defects. The prior art is limited to the fertilization method and ignores the response of the fertilization object.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The object of the present invention is to provide a field potato drip fertilization method based on feedback control to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for drip irrigation and fertilization of potatoes in a field based on feedback control, comprising the following steps:

[0008] Step 1: The proportional gain coefficient, differential gain coefficient and integral gain coefficient of the PID controller are used to form an individual vector and generate the initial population.

[0009] Step 2: Obtain historical potato tuber absorption feedback data and corresponding compound fertilizer flow control signals for yields that meet standards. Using the potato tuber absorption feedback data as input and the compound fertilizer flow control signals as output, construct and train an optimal compound fertilizer flow setting model.

[0010] Step 3: Obtain historical potato tuber absorption feedback data and corresponding compound fertilizer flow control signals for yields that meet standards. Using the potato tuber absorption feedback data as input and the compound fertilizer flow control signals as output, construct and train an optimal compound fertilizer flow setting model.

[0011] Step 4: Equivalent the transmission process of the compound fertilizer flow control signal motor system, obtain the system error signal based on the input compound fertilizer flow control signal and the compound fertilizer flow control signal equivalent to the motor system after individual gain adjustment, and construct the second fitness function based on the system error signal;

[0012] Step 5: Reorganize the splicing crossover individuals into 8 recombined individuals, select the individual with the lowest second fitness function value among the recombined individuals, and place the selected individuals in the iterative population. Repeat the splicing crossover and recombination operations until the iterative population reaches half of the initial population size. Randomly select half of the individuals in the initial population and the iterative population to form a new initial population;

[0013] Step 6: Repeat steps 3-5 until the preset number of iterations is reached, select the individual with the smallest first fitness function in the initial population, and complete the potato drip irrigation fertilization control.

[0014] Furthermore, each individual in the initial population is represented as ,in, is the initial population Individuals, is the first The proportional gain coefficient of each individual, is the first The differential gain coefficient of each individual, is the first The integral gain coefficient of each individual, is the index of the individual in the initial population, and , is the total number of individuals in the initial population, and A multiple of 4.

[0015] A yield threshold is preset, and historical potato tuber absorption feedback data and a corresponding compound fertilizer flow control signal for yields greater than the yield threshold are obtained. The potato tuber absorption feedback data is used as input and the compound fertilizer flow control signal is used as output to construct and train an optimal compound fertilizer flow setting model; the potato tuber absorption feedback data includes the impedance change rate of the potato tuber.

[0016] Furthermore, the first fitness function is:

[0017] ;

[0018] ;

[0019] in, To predict compound fertilizer flow deviation signal, is the target compound fertilizer flow control signal, is the compound fertilizer flow output after individual gain adjustment, is the first fitness function, is the time variable, is the length of the compound fertilizer flow control signal.

[0020] Furthermore, 2 individuals are randomly selected from the initial population each time for the trophy selection operation, and the first fitness function values ​​of the two selected individuals are compared. The individual with the smaller first fitness function value is called a candidate individual. The trophy selection operation is repeated until the number of candidate individuals is 4, and the 4 candidate individuals are subjected to a splicing crossover operation.

[0021] Furthermore, the specific logic of the crossover operation is: randomly select two individuals from the selected four individuals for splicing operation, and also perform splicing operation on the remaining two individuals to obtain two spliced ​​individuals, which are expressed as , For the selected The proportional gain coefficient of each individual, For the selected The differential gain coefficient of each individual, For the selected The integral gain coefficient of each individual, For the selected The proportional gain coefficient of each individual, For the selected The differential gain coefficient of each individual, For the selected The integral gain coefficient of each individual, , is the index of the selected individual; 3 elements from the two spliced ​​individuals are randomly selected for crossover.

[0022] Furthermore, the compound fertilizer flow control signal transmission process is equivalent in the frequency domain, and an equivalent transfer function of the compound fertilizer flow control signal in the motor system is established. The equivalent transmission process of the compound fertilizer flow control signal in the motor system includes the equivalence of the frequency conversion transmission process, the equivalence of the motor transmission process, and the equivalence of the fertilizer outlet transmission process of the fertilizing device;

[0023] The specific formula for the equivalent basis of the frequency conversion transmission process is:

[0024] ;

[0025] in, is the equivalent transfer function of the variable frequency transfer process, is the equivalent gain of the inverter, is the frequency of the compound fertilizer flow control signal;

[0026] The specific formula for equivalent motor transfer process is:

[0027] ;

[0028] in, is the equivalent transfer function of the motor transfer process, is the inertia time constant of the motor, is the equivalent gain of the motor;

[0029] The equivalent formula for the fertilizer delivery process of the fertilizing device is:

[0030] ;

[0031] in, is the equivalent transfer function of the fertilizer outlet transfer process of the fertilizing device, is the equivalent gain of the fertilizer outlet, is the time inertia constant of the fertilizer outlet, is the time lag constant of the fertilizer outlet of the compound fertilizer flow control signal;

[0032] The equivalent transfer function of compound fertilizer flow is:

[0033] ;

[0034] in, is the equivalent transfer function of compound fertilizer flow of the motor system.

[0035] Furthermore, the system error signal is calculated, and the system error signal is converted to the time domain through inverse Fourier transform. The second fitness function is constructed according to the system error in the time domain, specifically:

[0036] ;

[0037] ;

[0038] ;

[0039] in, is the second fitness function, is the system error signal in the time domain, is the system error signal in the frequency domain, It is the compound fertilizer flow control signal output from the fertilizer outlet. is the input compound fertilizer flow control signal in the frequency domain, is the equivalent transfer function of compound fertilizer flow of the motor system, is the PID control function after Laplace transform, is the time variable, is the length of the compound fertilizer flow control signal.

[0040] The present invention further provides a field potato drip fertigation system based on feedback control, the system being used to implement the field potato drip fertigation method based on feedback control, specifically comprising:

[0041] The population building module is used to construct individual vectors of the proportional gain coefficient, differential gain coefficient and integral gain coefficient of the PID controller and generate the initial population.

[0042] The flow prediction module is used to obtain historical potato tuber absorption feedback data and the corresponding compound fertilizer flow control signal when the yield reaches the target. The module uses the potato tuber absorption feedback data as input and the compound fertilizer flow control signal as output to build and train an optimal compound fertilizer flow setting model.

[0043] A crossover module is selected to obtain potato tuber absorption feedback data in real time and input the data into an optimal compound fertilizer flow setting model to obtain a target compound fertilizer flow control signal. A first fitness function is constructed based on the target compound fertilizer flow control signal and the input compound fertilizer flow control signal after individual gain adjustment. Individuals in the initial population are selected based on the first fitness function to perform a splicing crossover operation to obtain splicing crossover individuals.

[0044] A system equivalent module is used to perform equivalence on the compound fertilizer flow control signal motor system transmission process, obtain a system error signal based on the input compound fertilizer flow control signal and the compound fertilizer flow control signal that has been adjusted by individual gains and is equivalent to the motor system, and construct a second fitness function based on the system error signal;

[0045] The population recombination module is used to recombine the splicing crossover individuals into 8 recombined individuals, select the individual with the lowest second fitness function value among the recombined individuals, and place the selected individual in the iterative population. The splicing crossover and recombination operations are repeated until the iterative population reaches half the size of the initial population. Half of the individuals in the initial population are randomly selected to form a new initial population with the iterative population;

[0046] The iterative update module is used to repeatedly select the crossover module-population recombination module operation until the preset number of iterations is reached, and select the individual with the smallest first fitness function in the initial population to complete the potato drip irrigation fertilization control.

[0047] The present invention establishes a connection between the real-time feedback of potatoes and the control flow, and predicts the optimal flow. By optimizing the error between the system control flow and the predicted flow through the PID algorithm, the fertilization flow is controlled. By establishing a connection between the actual feedback of potatoes and the flow control, the actual growth needs of potatoes can be accurately matched, and the efficiency of resource utilization can be improved.

[0048] The present invention introduces genetic algorithms into PID controller optimization, which can significantly break through the limitations of traditional parameter tuning methods and achieve better control accuracy, robustness and adaptability in complex dynamic systems. In the process of optimizing the gain parameters of the PID controller using genetic algorithms, the present invention adopts a multi-parent co-evolution strategy to address the problem of limited crossover diversity caused by low individual dimensions (containing only proportional, integral and differential gains): first, four parent individuals are selected and spliced ​​in pairs to form spliced ​​crossover individuals; then, a crossover operation is performed on the spliced ​​crossover individuals to generate spliced ​​recombinant individuals; finally, the optimal recombinant solution is selected through fitness evaluation to participate in iterative updates, which significantly improves the global exploration capability of the solution space, breaks through the low-dimensional crossover limitation, and avoids premature convergence.

[0049] The present invention achieves accurate characterization of the compound fertilizer flow control process by establishing a full-link equivalent model covering variable frequency drive, motor dynamics and fluid transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0051] Figure 2 This is a flow control structure diagram of the present invention;

[0052] Figure 3 It is a schematic diagram of the overall system structure of the present invention. DETAILED DESCRIPTION

[0053] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0054] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0055] Example:

[0056] See also Figure 1-2 , the present invention provides a technical solution:

[0057] A field potato drip fertigation method based on feedback control, comprising the following steps:

[0058] Step 1: The proportional gain coefficient, differential gain coefficient and integral gain coefficient of the PID controller are used to form an individual vector and generate the initial population.

[0059] Furthermore, each individual in the initial population is represented as ,in, is the initial population Individuals, is the first The proportional gain coefficient of each individual, is the first The differential gain coefficient of each individual, is the first The integral gain coefficient of each individual, is the index of the individual in the initial population, and , is the total number of individuals in the initial population, and A multiple of 4.

[0060] In the PID control algorithm, the proportional gain coefficient directly affects the system's response speed and steady-state error. The integral gain coefficient is used to eliminate steady-state error, but it may increase overshoot or oscillation. The differential gain coefficient improves the system's dynamic characteristics and suppresses overshoot and oscillation. The subsequent splicing crossover operation is based on 4 individuals, so set A multiple of 4.

[0061] Step 2: Obtain historical potato tuber absorption feedback data and corresponding compound fertilizer flow control signals for yields that meet standards. Using the potato tuber absorption feedback data as input and the compound fertilizer flow control signals as output, construct and train an optimal compound fertilizer flow setting model.

[0062] A yield threshold is preset, and historical potato tuber absorption feedback data and a corresponding compound fertilizer flow control signal for yields greater than the yield threshold are obtained. The potato tuber absorption feedback data is used as input and the compound fertilizer flow control signal is used as output to construct and train an optimal compound fertilizer flow setting model; the potato tuber absorption feedback data includes the impedance change rate of the potato tuber.

[0063] The tuber impedance change rate of a potato tuber is the ratio of the impedance change value of the potato tuber within one minute to time. The impedance of the tuber can be obtained by an impedance probe array, and the impedance value of the center or other landmark positions within 1 minute is obtained, and the change rate is calculated.

[0064] The impedance probe array constitutes a feedback data acquisition system for collecting potato tuber absorption feedback data.

[0065] The synergistic effects of nitrogen, phosphorus, and potassium in compound fertilizers affect cell membrane permeability, ion channel activity, and cell wall structure, leading to comprehensive changes in the electrical impedance of tuber cells. Phosphorus promotes ATP synthesis and influences ion pump activity; nitrogen influences protein synthesis and alters membrane structure; and potassium directly participates in ion balance. These three factors work together to ensure that impedance changes more comprehensively reflect the plant's nutrient absorption status. This impedance change exhibits rapid response and specificity. The feedback signal is stable and fluctuates widely, making it suitable for feedback on compound fertilizer effectiveness.

[0066] The optimal compound fertilizer flow setting model adopts a feedforward neural network. The system dynamically calculates the optimal compound fertilizer demand under the current growth status based on the real-time collected potato tuber impedance change rate and the neural network model trained with historical yield compliance data, thereby generating a personalized compound fertilizer flow control signal. Specifically, the potato tuber absorption feedback data is used as input, and the corresponding optimal compound fertilizer flow setting value is the label. The model training adopts the existing technology, which specifically includes: input layer, hidden layer, output layer and activation function. The input layer is responsible for receiving the potato tuber absorption feedback data; the hidden layer is used to process the potato tuber absorption feedback data; it is composed of multiple layers, each layer contains multiple time nodes, and the time nodes of each hidden layer are connected to the previous layer through weights, which are used to perform feature abstraction and nonlinear transformation on the input potato tuber absorption feedback data; learn the complex mapping relationship between plant physiological state and optimal fertilizer application amount; by using the ReLu activation function, the nonlinear relationship is introduced so that the model can fit the complex feature relationship; an independent neuron is set in the output layer, which is responsible for converting the local and high-level feature representations extracted by the hidden layer for outputting compound fertilizer flow control signals; the root mean square error loss function is used; the input data is calculated once through the network to obtain the output result, the loss function is calculated according to the predicted value and the true value, the gradient of the loss function for each weight and bias is calculated by the chain rule, and the weights and biases of the network are updated using the gradient descent algorithm to minimize the loss function.

[0067] Step 3: Acquire potato tuber absorption feedback data in real time and input it into the optimal compound fertilizer flow setting model to obtain a target compound fertilizer flow control signal. Construct a first fitness function based on the target compound fertilizer flow control signal and the input compound fertilizer flow control signal after individual gain adjustment. Based on the first fitness function, select individuals from the initial population for a splicing and crossover operation to obtain splicing and crossover individuals.

[0068] Further, ;

[0069] ;

[0070] in, To predict compound fertilizer flow deviation signal, is the target compound fertilizer flow control signal, is the compound fertilizer flow output after individual gain adjustment, is the first fitness function, is the time variable, is the length of the compound fertilizer flow control signal.

[0071] The predicted compound fertilizer flow deviation signal reflects the deviation between the current compound fertilizer flow control signal and the ideal compound fertilizer flow control signal. , indicating that the control is completely accurate. Large, indicating poor control effect, is the absolute deviation accumulated during the entire control process, that is, the total error of the PID control system. The smaller it is, the better the PID parameters are. The better the control effect, the more stable and accurate the compound fertilizer can be controlled in the ideal state. This is to prevent positive and negative deviations from canceling each other out, ensuring that all deviations are factored into fitness evaluation. This prevents situations where a PID parameter causes fertilization to overshoot (positive deviation) followed by undershoot (negative deviation). Direct integration may appear to have a small error, but in reality, the control effect is suboptimal.

[0072] Each time, two individuals are randomly selected from the initial population for the trophy selection operation. The first fitness function values ​​of the two selected individuals are compared. The individual with the smaller first fitness function value is called a candidate individual. The trophy selection operation is repeated until the number of candidate individuals is 4, and the 4 candidate individuals are subjected to the splicing crossover operation.

[0073] The technical effect of the selection here is to select individuals with small deviations from the predicted compound fertilizer flow control signal.

[0074] Furthermore, the specific logic of the crossover operation is: randomly select two individuals from the selected four individuals for splicing operation, and also perform splicing operation on the remaining two individuals to obtain two spliced ​​individuals, which are expressed as , For the selected The proportional gain coefficient of each individual, For the selected The differential gain coefficient of each individual, For the selected The integral gain coefficient of each individual, For the selected The proportional gain coefficient of each individual, For the selected The differential gain coefficient of each individual, For the selected The integral gain coefficient of each individual, , is the index of the selected individual; 3 elements from the two spliced ​​individuals are randomly selected for crossover.

[0075] Step 4: Equivalent the transmission process of the compound fertilizer flow control signal motor system, obtain the system error signal based on the input compound fertilizer flow control signal and the compound fertilizer flow control signal equivalent to the motor system after individual gain adjustment, and construct the second fitness function based on the system error signal;

[0076] The transmission process of the compound fertilizer flow control signal in the motor system is equivalent in the frequency domain, and an equivalent transfer function of the compound fertilizer flow control signal in the motor system is established. The equivalent transmission process of the compound fertilizer flow control signal in the motor system includes the equivalence of the frequency conversion transmission process, the equivalence of the motor transmission process, and the equivalence of the fertilizer outlet transmission process of the fertilizing device;

[0077] The specific formula for the equivalent basis of the frequency conversion transmission process is:

[0078] ;

[0079] in, is the equivalent transfer function of the variable frequency transfer process, is the equivalent gain of the inverter, is the frequency of the compound fertilizer flow control signal;

[0080] By making a series of simplifications, approximations and using linearization methods near its static operating point, the specific formula for equivalent motor transfer process is:

[0081] ;

[0082] in, is the equivalent transfer function of the motor transfer process, is the inertia time constant of the motor, is the equivalent gain of the motor;

[0083] During the drip irrigation process, the flow rate at the fertilizer outlet of the fertilizing device generally remains in a relatively stable state. This position can be considered as a first-order inertia link, which is approximately equivalent to a pure lag first-order inertia link. The equivalent formula for the fertilizer outlet transfer process of the fertilizing device is:

[0084] ;

[0085] in, is the equivalent transfer function of the fertilizer outlet transfer process of the fertilizing device, is the equivalent gain of the fertilizer outlet, is the time inertia constant of the fertilizer outlet, is the time lag constant of the fertilizer outlet of the compound fertilizer flow control signal;

[0086] The equivalent transfer function of compound fertilizer flow is:

[0087] ;

[0088] in, is the equivalent transfer function of compound fertilizer flow of the motor system.

[0089] Furthermore, the system error signal is calculated, and the system error signal is converted to the time domain through inverse Fourier transform. The second fitness function is constructed according to the system error in the time domain, specifically:

[0090] ;

[0091] ;

[0092] ;

[0093] in, is the second fitness function, is the system error signal in the time domain, is the system error signal in the frequency domain, It is the compound fertilizer flow control signal output from the fertilizer outlet. is the input compound fertilizer flow control signal in the frequency domain, is the equivalent transfer function of compound fertilizer flow of the motor system, is the PID control function after Laplace transform, is the time variable, is the length of the compound fertilizer flow control signal.

[0094] During the water and fertilizer regulation process, factors such as the response time, overshoot, and error of the flow control system will affect the evaluation of the control effect. Therefore, this paper adopts the time multiplied absolute error integral criterion (ITAE) to construct the second fitness function to reflect the speed and accuracy of the control system.

[0095] The PID control function is:

[0096] ;

[0097] in, is the controller output, is the proportional gain coefficient, is the integral gain coefficient, is the differential gain coefficient, is the systematic error, Represents a time variable.

[0098] After Laplace transformation, it is expressed as:

[0099] ;

[0100] Step 5: Reorganize the splicing crossover individuals into 8 recombined individuals, select the individual with the lowest second fitness function value among the recombined individuals, and place the selected individuals in the iterative population. Repeat the splicing crossover and recombination operations until the iterative population reaches half of the initial population size. Randomly select half of the individuals in the initial population and the iterative population to form a new initial population;

[0101] The form of the recombined individuals is the same as that of the individuals in the initial population. Each recombined individual still contains the proportional gain coefficient, the integral gain coefficient, and the differential gain coefficient. The spliced ​​crossover individual contains two of each of the three types of gain coefficients. When recombining, one individual of each type is selected to form a recombined individual. The number of recombined individuals is ;

[0102] The optimal recombinant individual obtained by recombining two spliced ​​crossover individuals is placed in the iterative population through the second fitness function, and individuals in the initial population are selected again based on the first fitness function for splicing and crossover operations to obtain new splicing and crossover individuals. The new splicing and crossover individuals are recombined, and the optimal recombinant individual is selected from the new recombinant individuals based on the second fitness function. The new recombinant individuals are placed in the iterative population, and this process is repeated until the iterative population reaches half the size of the initial population.

[0103] This embodiment introduces a genetic algorithm into PID controller optimization, which can significantly break through the limitations of traditional parameter tuning methods and achieve better control accuracy, robustness and adaptability in complex dynamic systems. In the process of optimizing the PID controller gain parameters using a genetic algorithm, to address the problem of limited crossover diversity caused by low individual dimensions (containing only proportional, integral and differential gains), the present invention adopts a multi-parent co-evolution strategy: first, four parent individuals are selected and spliced ​​in pairs to form spliced ​​individuals; then, a crossover operation is performed on the spliced ​​individuals to generate spliced ​​recombinant individuals; finally, the optimal recombinant solution is selected through fitness evaluation to participate in iterative updates, significantly improving the global exploration capability of the solution space, breaking through the low-dimensional crossover limitation, and avoiding premature convergence.

[0104] On this basis, since this embodiment performs splicing, crossover and recombination operations, compared with general genetic algorithms, this embodiment actually performs two selections, and these two selections are in a progressive relationship. The first is to select the error size between the input control signal and the predicted control signal through the first fitness function. During the recombination selection process, the selected individuals select the system error size of the system through the second fitness function. The purpose of this selection is to mainly adjust the error between the input control signal and the predicted control signal, and adjust the system error of the system under this premise.

[0105] Step 6: Repeat steps 3-5 until the preset number of iterations is reached, select the individual with the smallest first fitness function in the initial population, and complete the potato drip irrigation fertilization control.

[0106] See also Figure 3 The present invention further provides a field potato drip fertigation system based on feedback control, wherein the system is used to implement the field potato drip fertigation method based on feedback control, specifically comprising:

[0107] The population building module is used to construct individual vectors of the proportional gain coefficient, differential gain coefficient and integral gain coefficient of the PID controller and generate the initial population.

[0108] The flow prediction module is used to obtain historical potato tuber absorption feedback data and the corresponding compound fertilizer flow control signal when the yield reaches the target. The module uses the potato tuber absorption feedback data as input and the compound fertilizer flow control signal as output to build and train an optimal compound fertilizer flow setting model.

[0109] Select a cross module to obtain historical potato tuber absorption feedback data and corresponding compound fertilizer flow control signals for yields that meet standards. Using the potato tuber absorption feedback data as input and the compound fertilizer flow control signals as output, build and train an optimal compound fertilizer flow setting model.

[0110] The system equivalent module is used to obtain potato tuber absorption feedback data in real time and input it into the optimal compound fertilizer flow setting model to obtain a target compound fertilizer flow control signal; a first fitness function is constructed based on the target compound fertilizer flow control signal and the input compound fertilizer flow control signal after individual gain adjustment; individuals in the initial population are selected based on the first fitness function to perform a splicing crossover operation to obtain splicing crossover individuals;

[0111] The population recombination module is used to recombine the splicing crossover individuals into 8 recombined individuals, select the individual with the lowest second fitness function value among the recombined individuals, and place the selected individual in the iterative population. The splicing crossover and recombination operations are repeated until the iterative population reaches half the size of the initial population. Half of the individuals in the initial population are randomly selected to form a new initial population with the iterative population;

[0112] The iterative update module is used to repeatedly select the crossover module-population recombination module operation until the preset number of iterations is reached, and select the individual with the smallest first fitness function in the initial population to complete the potato drip irrigation fertilization control.

[0113] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0114] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0115] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0116] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technical personnel familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, and they should all be covered by the scope of protection of the present application.

Claims

1. A field potato drip fertilization method based on feedback control, characterized in that: The specific steps include: Step 1: The proportional gain coefficient, differential gain coefficient and integral gain coefficient of the PID controller are used to form an individual vector and generate an initial population; Step 2: Obtain historical potato tuber absorption feedback data and corresponding compound fertilizer flow control signals for yields that meet standards. Using the potato tuber absorption feedback data as input and the compound fertilizer flow control signals as output, construct and train an optimal compound fertilizer flow setting model. Step 3: Acquire potato tuber absorption feedback data in real time and input it into the optimal compound fertilizer flow setting model to obtain a target compound fertilizer flow control signal. Construct a first fitness function based on the target compound fertilizer flow control signal and the input compound fertilizer flow control signal after individual gain adjustment. Based on the first fitness function, select individuals from the initial population for a splicing and crossover operation to obtain splicing and crossover individuals. Step 4: Equivalent the transmission process of the compound fertilizer flow control signal motor system, obtain the system error signal based on the input compound fertilizer flow control signal and the compound fertilizer flow control signal equivalent to the motor system after individual gain adjustment, and construct the second fitness function based on the system error signal; Step 5: Reorganize the splicing crossover individuals into 8 recombined individuals, select the individual with the lowest second fitness function value among the recombined individuals, and place the selected individuals in the iterative population. Repeat the splicing crossover and recombination operations until the iterative population reaches half of the initial population size. Randomly select half of the individuals in the initial population and the iterative population to form a new initial population; Step 6: Repeat steps 3-5 until the preset number of iterations is reached, select the individual with the smallest first fitness function in the initial population, and complete the potato drip irrigation fertilization control; Each individual in the initial population is represented by ,in, is the initial population Individuals, is the first The proportional gain coefficient of each individual, is the first The differential gain coefficient of each individual, is the first The integral gain coefficient of each individual, is the index of the individual in the initial population, and , is the total number of individuals in the initial population, and is a multiple of 4; The first fitness function is: in, To predict compound fertilizer flow deviation signal, is the target compound fertilizer flow control signal, is the compound fertilizer flow output after individual gain adjustment, is the first fitness function, is the time variable, is the length of the compound fertilizer flow control signal.

2. The method for drip irrigation and fertilization of field potatoes based on feedback control according to claim 1, characterized in that: A yield threshold is preset, and historical potato tuber absorption feedback data and a corresponding compound fertilizer flow control signal for yields greater than the yield threshold are obtained. The potato tuber absorption feedback data is used as input and the compound fertilizer flow control signal is used as output to construct and train an optimal compound fertilizer flow setting model; the potato tuber absorption feedback data includes the impedance change rate of the potato tuber.

3. The method for drip irrigation and fertilization of field potatoes based on feedback control according to claim 1, characterized in that: Each time, two individuals are randomly selected from the initial population for the trophy selection operation. The first fitness function values ​​of the two selected individuals are compared. The individual with the smaller first fitness function value is called a candidate individual. The trophy selection operation is repeated until the number of candidate individuals is 4, and the 4 candidate individuals are subjected to the splicing crossover operation.

4. The method for drip fertilization of field potatoes based on feedback control according to claim 3, characterized in that: The specific logic of the crossover operation is: randomly select two individuals from the selected four individuals for splicing operation, and also perform splicing operation on the remaining two individuals to obtain two spliced ​​individuals, which are expressed as , For the selected The proportional gain coefficient of each individual, For the selected The differential gain coefficient of each individual, For the selected The integral gain coefficient of each individual, For the selected The proportional gain coefficient of each individual, For the selected The differential gain coefficient of each individual, For the selected The integral gain coefficient of each individual, , is the index of the selected individual; 3 elements from the two spliced ​​individuals are randomly selected for crossover.

5. The method for drip fertigation of potato in a field based on feedback control according to claim 1, characterized in that: The transmission process of the compound fertilizer flow control signal in the motor system is equivalent in the frequency domain, and an equivalent transfer function of the compound fertilizer flow control signal in the motor system is established. The equivalent transmission process of the compound fertilizer flow control signal in the motor system includes the equivalence of the frequency conversion transmission process, the equivalence of the motor transmission process, and the equivalence of the fertilizer outlet transmission process of the fertilizing device; The specific formula for the equivalent basis of the frequency conversion transmission process is: in, is the equivalent transfer function of the variable frequency transfer process, is the equivalent gain of the inverter, is the frequency of the compound fertilizer flow control signal; The specific formula for equivalent motor transfer process is: in, is the equivalent transfer function of the motor transfer process, is the inertia time constant of the motor, is the equivalent gain of the motor; The equivalent formula for the fertilizer delivery process of the fertilizing device is: in, is the equivalent transfer function of the fertilizer outlet transfer process of the fertilizing device, is the equivalent gain of the fertilizer outlet, is the time inertia constant of the fertilizer outlet, is the time lag constant of the fertilizer outlet of the compound fertilizer flow control signal; The equivalent transfer function of compound fertilizer flow is: in, is the equivalent transfer function of compound fertilizer flow of the motor system.

6. The method for drip fertilization of field potatoes based on feedback control according to claim 5, characterized in that: Calculate the system error signal, convert the system error signal to the time domain through inverse Fourier transform, and construct the second fitness function based on the system error in the time domain, specifically: in, is the second fitness function, is the system error signal in the time domain, is the system error signal in the frequency domain, It is the compound fertilizer flow control signal output from the fertilizer outlet. is the input compound fertilizer flow control signal in the frequency domain, is the equivalent transfer function of compound fertilizer flow of the motor system, is the PID control function after Laplace transform, is the time variable, is the length of the compound fertilizer flow control signal.

7. A field potato drip fertigation system based on feedback control, characterized by: The system is used to implement the field potato drip fertilization method based on feedback control according to any one of claims 1 to 6, specifically comprising: The population building module is used to construct individual vectors of the proportional gain coefficient, differential gain coefficient and integral gain coefficient of the PID controller and generate the initial population. The flow prediction module is used to obtain historical potato tuber absorption feedback data and the corresponding compound fertilizer flow control signal when the yield reaches the target. The module uses the potato tuber absorption feedback data as input and the compound fertilizer flow control signal as output to build and train an optimal compound fertilizer flow setting model. A crossover module is selected to obtain potato tuber absorption feedback data in real time, and input the data into an optimal compound fertilizer flow setting model to obtain a target compound fertilizer flow control signal. A first fitness function is constructed based on the target compound fertilizer flow control signal and the input compound fertilizer flow control signal after individual gain adjustment. Individuals in the initial population are selected based on the first fitness function to perform a splicing crossover operation to obtain splicing crossover individuals. A system equivalent module is used to perform equivalence on the compound fertilizer flow control signal motor system transmission process, obtain a system error signal based on the input compound fertilizer flow control signal and the compound fertilizer flow control signal that has been adjusted by individual gains and is equivalent to the motor system, and construct a second fitness function based on the system error signal; The population recombination module is used to recombine the splicing crossover individuals into 8 recombined individuals, select the individual with the lowest second fitness function value among the recombined individuals, and place the selected individual in the iterative population. The splicing crossover and recombination operations are repeated until the iterative population reaches half the size of the initial population. Half of the individuals in the initial population are randomly selected to form a new initial population with the iterative population; The iterative update module is used to repeatedly select the crossover module-population recombination module operation until the preset number of iterations is reached, and select the individual with the smallest first fitness function in the initial population to complete the potato drip irrigation fertilization control.

Images