Method, device, equipment and medium for determining working parameters of seed supply device

Through the method of combining multi-objective regression model and wolf pack search algorithm, the problem of low working parameters in the air suction seed ejaculation device is solved, the speed and stability of parameter determination are improved, local optimal solution is avoided, and more Efficient seed delivery device performance.

CN118551503BActive Publication Date: 2024-11-26NANJING AGRICULTURAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411018129.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-26
Estimated Expiration
2044-07-29

AI Technical Summary

Technical Problem

In the existing air suction seed dischargers, when determining the working parameters based on the non-dominant sorting genetic algorithm, the calculation amount is large and the convergence speed is slow, resulting in low efficiency of the working parameters of the seed supply device.

Method used

The multi-objective regression model is used to combine the wolf pack search algorithm to determine the Pareto frontier direction of the initial population through genetic evolution, and the target population is optimized by the wolf pack search algorithm, thereby determining the target working parameters of the seed supply device.

Benefits of technology

The speed and stability of the determination of the working parameters of the seed delivery device are improved, and the genetic algorithm is avoided from falling into the local optimal solution at the end, which is enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118551503B_ABST
    Figure CN118551503B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, device, equipment and medium for determining working parameters of a seed supply device. The method comprises obtaining the equipment operating parameters and the seed supply working parameters of the seed supply device during the working process; constructing a multi-objective regression model based on the equipment operating parameters and the seed supply working parameters; constructing an initial population based on the equipment operating parameters and the seed supply working parameters, wherein the genotype of the initial individuals included in the initial population is the independent variable of the multi-objective regression model, and the phenotype is the equipment operating parameter; in the process of genetically evolving the initial population based on the multi-objective regression model, determining the Pareto frontier direction based on the evolving population, and obtaining a population to be optimized that meets the Pareto frontier direction; optimizing the population to be optimized based on a wolf pack search algorithm to obtain a target population; and determining the target working parameters corresponding to the seed supply device based on the target population, thereby improving the speed and stability of determining the working parameters of the seed supply device.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery, and in particular to a method, device, equipment and medium for determining working parameters of a seed supply device. Background Art

[0002] The air-suction seed metering device has become the key development direction of precision seeding technology for large seeds (e.g., soybeans) due to its high speed and precision. The air-suction seed metering device includes a variety of working parameters, and the determination of the working parameter combination is the key point of the air-suction seed metering device design.

[0003] At present, multiple working parameters are usually determined based on a multi-objective optimization method. In the prior art, a multi-objective optimization method usually performs multi-objective optimization based on a non-dominated sorting genetic algorithm to obtain a series of non-dominated solution sets.

[0004] However, the non-dominated sorting genetic algorithm has a large amount of computation and a slow convergence speed, which leads to low efficiency of the working parameters of the seed supply device. Summary of the invention

[0005] The present invention provides a method, device, equipment and medium for determining working parameters of a seed supply device, so as to solve the problem of low performance of determining target working parameters based only on genetic algorithm, and improve the speed and stability of determining working parameters of the seed supply device.

[0006] According to one aspect of the present invention, a method for determining working parameters of a seeding device is provided, comprising:

[0007] Obtaining equipment operating parameters and seed supply working parameters of the seed supply device during the working process; the equipment operating parameters include one or more items of seed plate rotation speed, vacuum degree and seed pile center of gravity height; and the seed supply working parameters include one or more items of empty seed frequency, seeding completion time and seed average adsorption time;

[0008] Constructing a multi-objective regression model based on the equipment operation parameters and the seed supply operation parameters, wherein the independent variables of the multi-objective regression model include at least part of the seed supply operation parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters;

[0009] An initial population is constructed based on equipment operation parameters and seed supply working parameters. The initial population includes multiple initial individuals. The genotype of the initial individual is the independent variable of the multi-objective regression model, and the phenotype of the initial individual is the equipment operation parameter.

[0010] Based on the multi-objective regression model, the initial population is genetically evolved to obtain the population to be optimized; wherein, in the process of genetic evolution, the Pareto frontier direction is determined based on the evolving population to obtain the population to be optimized that satisfies the Pareto frontier direction;

[0011] Based on the wolf pack search algorithm, the target population is optimized by optimizing the population to be optimized; wherein the target population satisfies the end search condition;

[0012] The target working parameters corresponding to the seed supply device are determined based on the target population.

[0013] According to another aspect of the present invention, there is provided a device for determining working parameters of a seeding device, comprising:

[0014] A parameter acquisition module is used to acquire the equipment operation parameters and seed supply working parameters of the seed supply device during the working process; the equipment operation parameters include one or more items of seed plate rotation speed, vacuum degree and seed pile gravity center height; the seed supply working parameters include one or more items of empty seed frequency, seeding completion time and seed average adsorption time;

[0015] A multi-objective regression model building module, used to build a multi-objective regression model based on equipment operation parameters and seed supply operation parameters, wherein the independent variables of the multi-objective regression model include at least part of the seed supply operation parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters;

[0016] The initial population construction module is used to construct the initial population based on the equipment operation parameters and the seed supply working parameters. The initial population includes multiple initial individuals. The genotype of the initial individual is the independent variable of the multi-objective regression model, and the phenotype of the initial individual is the equipment operation parameter.

[0017] The genetic evolution module is used to perform genetic evolution on the initial population based on the multi-objective regression model to obtain the population to be optimized; wherein, in the process of genetic evolution, the Pareto front direction is determined based on the evolving population to obtain the population to be optimized that satisfies the Pareto front direction;

[0018] The target population determination module is used to optimize the to-be-optimized population based on the wolf pack search algorithm to obtain the target population; wherein the target population satisfies the end search condition;

[0019] The target working parameter determination module is used to determine the target working parameters corresponding to the seed supply device based on the target population.

[0020] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0021] at least one processor; and

[0022] a memory communicatively connected to at least one processor; wherein,

[0023] The memory stores a computer program that can be executed by at least one processor. The computer program is executed by at least one processor so that the at least one processor can execute the method for determining the working parameters of the seeding device according to any embodiment of the present invention.

[0024] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining working parameters of a seeding device according to any embodiment of the present invention when executed.

[0025] Beneficial effects:

[0026] The technical solution of the embodiment of the present invention determines the direction of the Pareto front based on the evolving population in the process of genetically evolving the initial population based on a multi-objective regression model, obtains a population to be optimized that satisfies the direction of the Pareto front, and optimizes the population to be optimized based on a wolf pack search algorithm to obtain a target population, so as to determine the target working parameters corresponding to the seed supply device based on the target population, instead of genetically evolving the population to be optimized to obtain the target population after obtaining the population to be optimized. This solves the problem of low performance of determining the target working parameters based only on the genetic algorithm, and by combining the genetic algorithm and the wolf pack search algorithm, it is possible to avoid falling into a local optimal solution at the end of the genetic evolution. The wolf pack search algorithm improves the probability of finding the global optimal solution by combining local search with global search, and improves the speed and stability of determining the working parameters of the seed supply device.

[0027] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 is a flow chart of a method for determining working parameters of a seed supply device provided in Embodiment 1 of the present invention;

[0030] Figure 2 It is a structural schematic diagram of a seed supply device provided in Embodiment 1 of the present invention;

[0031] Figure 3 It is a structural schematic diagram of a seeding disc provided in the first embodiment of the present invention;

[0032] Figure 4 is a flow chart of a method for determining working parameters of a seed supply device provided in Embodiment 2 of the present invention;

[0033] Figure 5It is a structural diagram of a device for determining working parameters of a seed supply device provided in Embodiment 3 of the present invention;

[0034] Figure 6 It is a structural schematic diagram of an electronic device for implementing the method for determining working parameters of a seed supply device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0036] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0037] Embodiment 1

[0038] Figure 1 This is a flow chart of a method for determining working parameters of a seed supply device provided in Embodiment 1 of the present invention. This embodiment can be applied to designing working parameters of a seed supply device. The method can be executed by a device for determining working parameters of a seed supply device. The device for determining working parameters of a seed supply device can be implemented in the form of hardware and / or software. The device for determining working parameters of a seed supply device can be configured in electronic devices such as computers and servers.

[0039] In this embodiment, the seed supply device is an automatic seed supply device of an air-suction seed metering device. Figure 2 is a structural schematic diagram of a seed supply device provided in Embodiment 1 of the present invention, such as Figure 2As shown, the seed supply device 20 includes a seed box 21, a seed carrying tube 22, a seed guide turntable 23, a turntable support 24, a seed drop tube 25, a seed collection tube 26 and a bracket 27. Among them, the seed guide turntable 23 is provided with a notch, and the seed supply device supplies seeds to the seed meter through the rotation of the seed guide turntable 23. The cavity of the seed meter includes a seeding disk, and the seeding speed of the seed meter can be adjusted by adjusting the rotation speed of the seeding disk. Figure 3 Schematic diagram of the structure of a seeding disc provided in the first embodiment of the present invention. Figure 3 As shown, the seeding disc 30 includes a front shell 31 and a shaped hole 32, wherein the best seeding area 33 is Figure 3 The area enclosed by the dotted line frame. The seed supply device is used to transport seeds from the seed box to the seed meter, and the seed meter is used to discharge the seeds into the soil at equal intervals.

[0040] like Figure 1 As shown, the method includes:

[0041] S110, obtaining equipment operating parameters and seed supply working parameters of the seed supply device during the working process; the equipment operating parameters include one or more items of seed plate rotation speed, vacuum degree and seed pile center of gravity height; the seed supply working parameters include one or more items of empty seed frequency, seeding completion time and seed average adsorption time.

[0042] In this embodiment, the equipment operation parameters are parameters used to affect the seed supply performance of the seed supply device, and the seed supply working parameters are parameters used to characterize the seed supply performance of the seed supply device. The seed disc speed is the rotational angular velocity of the seed disc in the seed meter; the vacuum degree is the vacuum degree of the seed disc hole, characterizing the adsorption strength of the hole to the seed; the seed pile gravity center height is the gravity center height of the seed pile in the seed cavity of the seed meter; the empty seed frequency is used to characterize the number of holes that do not adsorb seeds in the optimal seeding area when the seed disc rotates through M holes on average, where the number of holes in the optimal seeding area is M; the seeding completion time is the time required for the number of seeds in the seed cavity of the seed meter to drop to 20% of the initial number of seeds; the average seed adsorption time is the quotient of the total time from the seed entering the effective seeding area to the seed completely exiting the effective seeding area in the seed meter and the number of holes in the optimal seeding area. In some embodiments, after 80% of the seeds on the plane at the center of gravity of the seed pile are adsorbed, the number of seeds in the seed cavity of the seed meter is scarce, resulting in reduced seeding efficiency. Optionally, the time it takes for 80% of the seeds on the plane at the center of gravity of the seed pile to be adsorbed is determined as the seeding completion time to avoid the influence of the number of seeds on the determination of the working parameters of the seed supply device.

[0043] Exemplarily, the rotation speed of the seed disc is 35 rpm, the vacuum degree is 4 Pascals, the height of the center of gravity of the seed pile is 10 cm, the empty seed frequency is 6 seconds, the seeding completion time is 5 seconds, and the average seed adsorption time is 0.1 seconds.

[0044] Specifically, the equipment operation parameters and the seed supply working parameters can be real data or simulation data, and this embodiment does not limit this. In some embodiments, based on sensors, experimental data is collected during the actual working process of the seed supply device to obtain multiple groups of real experimental data, each group of real experimental data includes equipment operation parameters and seed supply working parameters; based on the real experimental data, the equipment operation parameters and seed supply working parameters are extracted. Exemplarily, the sensors include but are not limited to angular velocity sensors, vacuum sensors and laser sensors. In some embodiments, the seed supply device is three-dimensionally modeled based on a modeling application to generate a model file of the seed supply device; the generated model file is imported into a finite element application, and the model file is meshed to generate a corresponding mesh file; based on the generated mesh file, a fluid coupling simulation is performed to obtain multiple groups of simulation experimental data, each group of simulation experimental data includes equipment operation parameters and seed supply working parameters; based on the simulation experimental data, the equipment operation parameters and seed supply working parameters are extracted. Exemplarily, the fluid coupling simulation can be based on the discrete element method modeling application and the fluid simulation application to perform coupling simulation.

[0045] S120. Construct a multi-objective regression model based on the equipment operation parameters and the seed supply operation parameters, wherein the independent variables of the multi-objective regression model include at least part of the seed supply operation parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters.

[0046] In this embodiment, the multi-objective regression model is a mathematical model used to describe the relationship between each parameter in the seed supply working parameters and multiple parameters in the equipment operation parameters.

[0047] Specifically, by fitting the relationship between each parameter in the seed supply working parameter and multiple parameters in the equipment operation parameter based on real experimental data and / or simulation experimental data, a regression equation corresponding to each parameter in the seed supply working parameter is obtained, and a multi-objective regression model is constructed based on the regression equations corresponding to the multiple parameters in the seed supply working parameter. The regression equation is a multivariate quadratic equation.

[0048] Exemplarily, the multi-objective regression model can be expressed as:

[0049] ,

[0050] in, Characterize the seed supply working parameters parameters, Characterize the equipment operating parameters parameters, Characterize the seed supply working parameters The constant term in the regression equation corresponding to the parameter is Characterize the seed supply working parameters The regression equation corresponding to the parameter coefficients of non-constant terms, where , , ,For example, Characterize the frequency of the empty species, Indicates the completion time of the previous seed. Characterizes the average adsorption time of seeds, Characterizes the rotation speed of the seed disc, Characterize the vacuum degree, Characterizes the height of the center of gravity of the seed pile.

[0051] It should be noted that, for any parameter in the seed supply working parameters, when the influence of any parameter in the equipment operating parameters on the parameter in the seed supply working parameters can be ignored, the data item corresponding to the parameter in the equipment operating parameters in the regression equation corresponding to the parameter in the seed supply working parameters can be eliminated, so that the independent variables of the multi-objective regression model include at least part of the seed supply working parameters.

[0052] In some embodiments, a simplified multi-objective regression model is constructed based on response surface analysis. Optionally, before constructing the multi-objective regression model, it also includes: based on supplying working parameters as target variables, performing response surface analysis on equipment operating parameters; determining significance data of each parameter in the equipment operating parameters based on the response surface analysis results; and determining independent variables of the multi-objective regression model based on the significance data of each parameter in the equipment operating parameters.

[0053] In this embodiment, the target variable is a variable used to describe the performance of the seed supply device in response surface analysis.

[0054] Specifically, by inputting real experimental data and / or simulation experimental data into the response surface analysis application, and setting each parameter in the supply working parameters as the target variable, setting each parameter in the equipment operation parameters as the response surface analysis test factor, and setting the factor variation range for multiple response surface analysis test factors, the response surface analysis result is obtained by performing response surface analysis on the target variable and the response surface analysis test factor, wherein the response surface analysis result includes the initial regression equations corresponding to the multiple target variables and the significance data of each response surface analysis test factor on each target variable. For any target variable, when the significance data of the effect of any response surface analysis test factor on the target variable does not meet the preset significance data threshold, it can be considered that the effect of the response surface analysis test factor on the target variable can be ignored, then the target regression equation corresponding to the target variable is obtained by eliminating the data item corresponding to the response surface analysis test factor in the initial regression equation corresponding to the target variable. Wherein, the preset significance data threshold is the threshold of the significance data corresponding to the response surface analysis test factor, which is used to determine that the effect of any response surface analysis test factor on any target variable cannot be ignored, for example, the significance data is the P value, and the preset significance data threshold is 0.05. Based on the target regression equations corresponding to the multiple target variables, a multi-target regression model is constructed. It can be understood that the significance data corresponding to the independent variables retained in each target regression equation in the multi-target regression model all meet the preset significance data threshold.

[0055] The technical solution of this embodiment determines the independent variables of the multi-objective regression model based on the significance data of each parameter in the equipment operating parameters in the response surface analysis results of the equipment operating parameters relative to the target variables. Based on the determined independent variables of the multi-objective regression model, a simplified multi-objective regression model can be constructed to reduce the complexity of the multi-objective regression model, which helps to improve the efficiency of determining the working parameters of the subsequent seed supply device.

[0056] In some embodiments, by performing variance analysis on the target regression equations corresponding to multiple target variables in the response surface analysis results, the significance indexes corresponding to the constant term, the first term, the second term and the square term in each target regression equation are determined, wherein the significance index is the data used to characterize the degree of influence of each data item in the target regression equation corresponding to each target variable on each target variable. It can be understood that when the target regression equation includes the seeding disc speed, the vacuum degree and the seed pile gravity center height, the variance analysis is performed by a three-factor three-level orthogonal test, wherein the three factors include the seeding disc speed, the vacuum degree and the seed pile gravity center height, and the three levels are low, medium and high levels determined based on the parameter values ​​of the seeding disc speed, the vacuum degree and the seed pile gravity center height. For example, Table 1 shows the relationship between the seeding disc speed, the vacuum degree and the seed pile gravity center height and the three levels. As shown in Table 1, the low level is represented by a logical identifier -1, the medium level is represented by a logical identifier 0, and the high level is represented by a logical identifier 1.

[0057] Table 1

[0058]

[0059] When the significance index corresponding to any data item in the target regression equation corresponding to any target variable does not meet the preset significance index threshold, it can be considered that the influence of the data item in the target regression equation on the target variable can be ignored. Then, by eliminating the data item in the initial regression equation corresponding to the target variable, the updated target regression equation corresponding to the target variable is obtained, and a multi-objective regression model is generated based on the updated target regression equations corresponding to multiple target variables. A multi-objective regression model is constructed based on the updated target regression equations corresponding to the set of multiple target variables. The significance index can be a P value or an F value, which is not limited in this embodiment. The preset significance index threshold is the threshold of the significance index, which is used to determine that any data item in the target regression equation cannot be ignored. For example, the significance index is a P value, and the preset significance data threshold is 0.05.

[0060] Exemplarily, the multi-objective regression model is:

[0061] .

[0062] The technical solution implemented in this embodiment simplifies the target regression equation based on variance analysis to obtain an updated target regression equation, which can further reduce the complexity of the multi-target regression model and further improve the efficiency of determining the working parameters of the subsequent seed supply device.

[0063] S130: constructing an initial population based on equipment operation parameters and seed supply operation parameters.

[0064] The initial population includes multiple initial individuals, the genotypes of the initial individuals are the independent variables of the multi-objective regression model, and the phenotypes of the initial individuals are the equipment operation parameters.

[0065] In this embodiment, the initial individual is an individual generated based on the equipment operation parameters and the seed supply operation parameters, the genotype of the initial individual includes each parameter in the independent variable of the multi-objective regression model, and the phenotype of the initial individual includes each parameter in the equipment operation parameters. In some embodiments, the genotype of the initial individual is the independent variable of the normalized multi-objective regression model, and the phenotype is the normalized seed supply operation parameter.

[0066] Specifically, based on the values ​​of each parameter in the acquired equipment operation parameters, the numerical ranges corresponding to the multiple parameters in the independent variables of the multi-objective regression model are determined. In the process of constructing any initial individual, for each parameter in the independent variables of the multi-objective regression model, numerical sampling is performed based on the numerical range corresponding to the parameter to obtain a set of candidate equipment operation parameters; by inputting the values ​​of multiple parameters in the set of equipment operation parameters into the multi-objective regression model for calculation, the candidate seed supply working parameters corresponding to the set of equipment operation parameters are obtained, and then an initial individual with the set of equipment operation parameters as the genotype and the candidate seed supply working parameters corresponding to the set of equipment operation parameters as the phenotype is constructed; by constructing multiple initial individuals, an initial population is obtained. Among them, numerical sampling includes but is not limited to random sampling, equal interval sampling and sampling based on a good point set, which is not limited in this embodiment. It should be noted that the number of initial individuals in the initial population can be preset.

[0067] Exemplarily, assume that the independent variables of the multi-objective regression model include seed disk speed, vacuum degree and seed pile center of gravity height, wherein the value range of the seed disk speed is 15 to 60 rpm, the value range of the vacuum degree is 2 to 6 Pascals, and the height of the seed pile center of gravity is 5 to 16 centimeters.

[0068] S140. Based on a multi-objective regression model, genetic evolution is performed on the initial population to obtain a population to be optimized; wherein, in the process of genetic evolution, the direction of the Pareto frontier is determined based on the evolving population to obtain a population to be optimized that satisfies the direction of the Pareto frontier.

[0069] It should be noted that genetic evolution is a process of iterative optimization of the initial population through operations such as selection, crossover and mutation. Through iterative optimization, the individuals in the population gradually approach the optimal solution. The Pareto front is used to characterize a set of optimal solutions in the seed supply working parameters. The optimal solution is better than other solutions in at least one parameter of the seed supply working parameters.

[0070] In this embodiment, the Pareto front direction refers to the optimization direction from the genetically evolved population corresponding to the current number of genetic iterations to the genetically evolved population corresponding to the Pareto front. The population to be optimized is the population obtained after the initial population is gradually optimized during the genetic evolution process, and the individual representations in the population to be optimized represent the solution when the Pareto front direction is determined.

[0071] Specifically, the initial population is subjected to genetic iteration based on a preset non-dominated sorting genetic algorithm to obtain an evolving population. The preset non-dominated sorting genetic algorithm includes but is not limited to a second non-dominated sorting genetic algorithm (Non-dominated Sorting Genetic Algorithm II, NSGA-II) and an improved NSGA-II algorithm. In the case where the Pareto front direction can be determined based on the evolving population, the evolving population is determined as the population to be optimized.

[0072] Exemplarily, assuming that in a population that satisfies the Pareto front direction, the number of individuals in the first Pareto front level is greater than the number of individuals in the non-first Pareto front level, then in an evolving population, if the number of individuals in the first Pareto front level is greater than the number of individuals in the non-first Pareto front level, it can be considered that the Pareto front direction can be determined based on the evolving population, and the evolving population is determined as the population to be optimized.

[0073] S150. Based on the wolf pack search algorithm, the target population is optimized to obtain the target population; wherein the target population satisfies the search end condition.

[0074] It should be noted that the wolf pack search algorithm is optimized by simulating the behavior of wolves, where each individual in the wolf pack represents a solution in the solution space. During the algorithm's iteration process, the wolf pack individuals search and update according to certain strategies, gradually approaching the optimal solution.

[0075] In this embodiment, the target population is a population obtained after the wolf pack individuals are gradually optimized during the search process, and the individuals in the target population represent the optimal solution during the search process.

[0076] Specifically, based on the population to be optimized, wolf group individuals are constructed, wherein each wolf group individual corresponds to an individual in the population to be optimized. The wolf group individuals are optimized based on the wolf group search algorithm. During the wolf group optimization process, if any optimized wolf group meets the end search condition, a target population is constructed based on the wolf group that meets the end search condition, and each individual in the target population corresponds to each wolf group individual in the wolf group that meets the end search condition, wherein the end search condition is a condition for determining the end of the wolf group search. Exemplarily, the end search condition is that the number of individuals in the non-first Pareto front level in the population constructed based on the optimized wolf group can be ignored.

[0077] In some embodiments, the end search condition is that the number of dominant individuals in the target population is greater than the second preset multiple of the number of individuals in the target population, wherein the second preset multiple is greater than the first preset multiple. In this embodiment, the dominant level is the first Pareto front level. The first preset multiple and the second preset multiple are both pre-set multiples, and it should be noted that the second preset multiple is greater than the first preset multiple. Correspondingly, optionally, the following steps are iteratively performed, and when the updated population to be optimized corresponding to the current search iteration number satisfies the end search condition, the updated population to be optimized corresponding to the current search iteration number is determined as the target population: an initial wolf group is constructed based on the current population to be optimized; the current population to be optimized is the population to be optimized or the updated population to be optimized corresponding to the previous search iteration number; based on the prey concentration function values ​​corresponding to the multiple artificial wolves contained in the initial wolf group, the positions of the multiple artificial wolves are updated; the prey concentration function value is determined based on the parameters in the seed supply working parameters corresponding to the artificial wolves; based on the multiple artificial wolves after the position update, the updated population to be optimized corresponding to the current search iteration number is constructed.

[0078] In this embodiment, the current search iteration number is any search iteration number. When the current search iteration number is the first search iteration number, the current population to be optimized is the population to be optimized; when the current search iteration number is the In the case of searching iterations, the current population to be optimized is The updated population to be optimized corresponds to the number of search iterations, where Among them, the updated population to be optimized is a population constructed based on the updated wolf pack.

[0079] Specifically, at the current search iteration number, an artificial wolf is constructed based on each individual in the current population to be optimized, and multiple artificial wolves form an initial wolf pack. The position of each artificial wolf is determined based on the genotype of the individual corresponding to the artificial wolf. For example, The position of the artificial wolf is represented by ,in, Respectively The seed disc speed, vacuum degree and seed pile center of gravity height in the genotype of the individual corresponding to the artificial wolf, among which, ,in, Characterizes the number of artificial wolves in the initial wolf pack.

[0080] The phenotype of each individual corresponding to the artificial wolf included in the initial wolf pack is input into the prey concentration function to obtain the prey concentration function value corresponding to the artificial wolf. In some embodiments, the independent variables in the prey concentration function include one or more of the average adsorption time of seeds, the frequency of empty seeds, and the completion time of the last seed. For example, The prey concentration function value of an artificial wolf is expressed as ,in, Expressed as The phenotype of the individual corresponding to the artificial wolf, Characterize the conversion relationship between phenotype and prey concentration function value.

[0081] It should be noted that the larger the prey concentration function value of any artificial wolf is, the closer the artificial wolf is to the prey. In some embodiments, a preset number of artificial wolves with the smallest prey concentration function value in the initial wolf pack are eliminated, and a preset number of new artificial wolves are randomly generated based on the solution space, thereby updating the initial wolf pack, shortening the average distance between multiple artificial wolves in the initial wolf pack and the prey, helping to improve the search efficiency, and thus speeding up the determination of the working parameters of the seed supply device.

[0082] The artificial wolf corresponding to the maximum prey concentration function value in the initial wolf pack is determined as the alpha wolf. When the maximum prey concentration function value corresponds to multiple artificial wolves, any artificial wolf corresponding to the maximum prey concentration function value is determined as the alpha wolf; the artificial wolf with the largest number of scout wolves other than the alpha wolf in the initial wolf pack is determined as a scout wolf, and the artificial wolves other than the alpha wolf and the scout wolf in the initial wolf pack are determined as fierce wolves, thereby realizing the division of the initial wolf pack. The number of scout wolves is determined based on a preset scout wolf ratio factor and the number of artificial wolves in the initial wolf pack. Exemplarily, assuming that the preset scout wolf ratio factor is , then the number of wolves is , the number of fierce wolves is .

[0083] Each scout wolf performs a wandering behavior, and moves forward in the direction of the solution space where the prey concentration function value is the largest and greater than the prey concentration function value at the current position. For example, assuming that the current position of the scout wolf is , the wolf is in Where Explore in different directions. In the direction The prey concentration function value at the exploration position in the direction is the largest, then the exploration wolf moves to the first One step in each direction to reach the exploration location , for example, assuming the search step is expressed as , then The exploration position in each direction is expressed as , among which, The exploration position of the solution space dimension is expressed as: , , for A random number between .

[0084] Repeat the above wandering behavior. If the prey concentration function value of any scout wolf is greater than the prey concentration function value of the alpha wolf, it can be considered that the scout wolf is closer to the prey than the alpha wolf. Then the scout wolf is updated to the alpha wolf, and the updated alpha wolf initiates a summoning behavior to the fierce wolf, so that the summoned fierce wolf moves to the position of the updated alpha wolf. During the rushing process, if the prey concentration function value of any fierce wolf is greater than the prey concentration function value of the alpha wolf, it can be considered that the fierce wolf is closer to the prey than the alpha wolf, then the fierce wolf is updated to the alpha wolf and initiates a summoning behavior; if the prey concentration function value of any fierce wolf is less than or equal to the prey concentration function value of the alpha wolf, it can be considered that the alpha wolf is closer to the prey than the fierce wolf, then the fierce wolf continues to respond to the call of the alpha wolf and moves to the position of the alpha wolf until it enters the range of the siege prey position and captures the prey. During the hunting process, both the fierce wolf and the scout wolf determine the target moving position based on the current position, prey position and attack step length. If the prey concentration function value of any artificial wolf at the target moving position is greater than the prey concentration function value at the current position, it can be considered that the artificial wolf can attack the prey better at the target moving position than at the Dangxiang position. The artificial wolf will be moved to the target moving position, otherwise the current position will remain unchanged.

[0085] For example, assume that during the siege process, the siege step length is expressed as , No. The current position of the artificial wolf is expressed as , Represents the number of iterations in the siege process, and the current position of the prey is , then The target moving position of the artificial wolf is ,in, , , for A random number between .

[0086] It should be noted that the artificial wolf in the wolf pack updates its position once every time it moves one step, and obtains an artificial wolf with an updated position. For each artificial wolf with an updated position, the genotype corresponding to the artificial wolf is generated based on the parameters in the updated position, and the phenotype corresponding to the artificial wolf is obtained by inputting the genotype into the multi-objective regression model. Based on the genotype and phenotype corresponding to the artificial wolf, an updated individual corresponding to the artificial wolf is generated. By collecting the updated individuals corresponding to the artificial wolves with multiple updated positions, an updated population to be optimized corresponding to the current number of search iterations is constructed.

[0087] By performing non-dominated sorting on the updated population to be optimized corresponding to the current search iteration number, the number of dominant individuals in the updated population to be optimized is determined, wherein the dominant individuals are updated individuals belonging to the first Pareto front level. If the number of dominant individuals in the updated population to be optimized is greater than the second preset multiple of the number of individuals in the updated population to be optimized, it can be considered that the updated population to be optimized corresponding to the current search iteration number meets the end search condition, and the updated population to be optimized corresponding to the current search iteration number is determined as the target population; if the number of dominant individuals in the updated population to be optimized is less than or equal to the second preset multiple of the number of individuals in the updated population to be optimized, it can be considered that the updated population to be optimized corresponding to the current search iteration number does not meet the end search condition, and enter the next search iteration number. Exemplarily, assuming that the second preset multiple is 0.95, then when the number of dominant individuals in the updated population to be optimized is greater than 0.95 times the number of individuals in the updated population to be optimized, the updated population to be optimized is determined as the target population.

[0088] In some embodiments, the wolf pack search algorithm includes a preset maximum number of search iterations. When the number of dominant individuals in the updated population to be optimized is greater than a second preset multiple of the number of individuals in the updated population to be optimized, the next number of search iterations is continued until the current number of search iterations is greater than the preset maximum number of search iterations. Then, the updated population to be optimized that represents the global optimal solution (for example, the one with the largest number of dominant individuals) in the updated population to be optimized corresponding to multiple search iterations is determined, and the updated population to be optimized that represents the global optimal solution is determined as the target population, which helps to ensure that the working parameters of the seed supply device determined based on the target population are globally optimal.

[0089] The technical solution of this embodiment is to construct an updated population to be optimized corresponding to the current search iteration number based on multiple artificial wolves after position update at the current search iteration number, and determine the updated population to be optimized corresponding to the current search iteration number as the target population when the updated population to be optimized corresponding to the current search iteration number meets the end search condition, wherein the end search condition is that the number of dominant individuals in the target population is greater than a second preset multiple of the number of individuals in the target population, which can ensure that the parameters in the expressiveness of multiple individuals in the target population converge and are the global optimal solution, and help to improve the accuracy of determining the working parameters of the seed supply device.

[0090] S160. Determine target operating parameters corresponding to the seed supply device based on the target population.

[0091] In this embodiment, the target operating parameters include target equipment operating parameters and seed supply operating parameters corresponding to the target equipment operating parameters, wherein the target equipment operating parameters are equipment operating parameters that characterize the optimal working performance of the seed supply device.

[0092] Specifically, based on the phenotypes of multiple individuals in the target population, at least one candidate individual is determined, and the phenotype of the candidate individual, compared with the phenotypes of other individuals in the target population, represents the best working performance of the seed supply device. It can be understood that the smaller the empty seed frequency, the shorter the seeding completion time, and the longer the average seed adsorption time, the better the working performance of the seed supply device.

[0093] In the case of a single candidate individual, the parameters in the candidate individual's genotype are determined as the parameters in the target device operating parameters, and the parameters in the candidate individual's phenotype are determined as the parameters in the seed supply operating parameters corresponding to the target device operating parameters. In the case of multiple candidate individuals, the target individual is determined based on the genotypes of the multiple candidate individuals, wherein the genotype of the target individual, compared with the genotypes of other candidate individuals, characterizes the best seed supply device operating performance.

[0094] It can be understood that among the parameters in the genotype, the main influencing factor on the working performance of the seed supply device is the height of the center of gravity of the seed pile, the rotation speed of the seeding disk is a secondary influencing factor, and the vacuum degree is the least influencing factor. In the case where the genotype includes the height of the center of gravity of the seed pile, the candidate individual corresponding to the smallest height of the center of gravity of the seed pile among multiple candidate individuals is determined as the target individual; in the case where the genotype includes the height of the center of gravity of the seed pile and the rotation speed of the seeding disk, and there are multiple candidate individuals corresponding to the smallest height of the center of gravity of the seed pile, the candidate individual corresponding to the smallest rotation speed of the seeding disk among multiple candidate individuals corresponding to the smallest height of the center of gravity of the seed pile is determined as the target individual; in the case where the genotype includes the height of the center of gravity of the seed pile, the rotation speed of the seeding disk and the vacuum degree, and there are multiple candidate individuals corresponding to the smallest height of the center of gravity of the seed pile and the smallest rotation speed of the seeding disk, the candidate individual corresponding to the smallest vacuum degree among multiple candidate individuals corresponding to the smallest height of the center of gravity of the seed pile and the smallest rotation speed of the seeding disk is determined as the target individual.

[0095] For example, assume that there are 2 candidate individuals, among which the seed pile center of gravity height of candidate individual 1 is 5 cm, the seed disk rotation speed is 35 rpm and the vacuum degree is 5 Pascals, and the seed pile center of gravity height of candidate individual 2 is 5 cm, the seed disk rotation speed is 35 rpm and the vacuum degree is 4 Pascals, then candidate individual 2 is determined as the target individual.

[0096] The target equipment operating parameters are obtained by setting each parameter in the genotype of the target individual as the target operating parameter and setting each parameter in the phenotype of the target individual as the seed supply operating parameter corresponding to the target operating parameter.

[0097] The technical solution of this embodiment determines the direction of the Pareto front based on the evolving population in the process of genetically evolving the initial population based on the multi-objective regression model, obtains the population to be optimized that satisfies the direction of the Pareto front, and optimizes the population to be optimized based on the wolf pack search algorithm to obtain the target population, so as to determine the target working parameters corresponding to the seed supply device based on the target population, instead of genetically evolving the population to be optimized to obtain the target population after obtaining the population to be optimized. This solves the problem of low performance of determining the target working parameters based on the genetic algorithm alone, and by combining the genetic algorithm and the wolf pack search algorithm, it is possible to avoid falling into the local optimal solution at the end of the genetic evolution. The wolf pack search algorithm improves the probability of finding the global optimal solution by combining local search with global search, and improves the speed and stability of determining the working parameters of the seed supply device.

[0098] Embodiment 2

[0099] Figure 4 1 is a flow chart of a method for determining working parameters of a seeding device provided in the second embodiment of the present invention. The technical solution of the embodiment of the present invention is improved on the basis of the above embodiment. The explanations of the terms that are the same or corresponding to the above embodiments are not repeated here. Figure 4 As shown, the method includes:

[0100] S210, obtaining equipment operating parameters and seed supply working parameters of the seed supply device during the working process; the equipment operating parameters include one or more items of seed plate rotation speed, vacuum degree and seed pile center of gravity height; the seed supply working parameters include one or more items of empty seed frequency, seeding completion time and seed average adsorption time.

[0101] S220. Construct a multi-objective regression model based on the equipment operation parameters and the seed supply operation parameters, wherein the independent variables of the multi-objective regression model include at least part of the seed supply operation parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters.

[0102] S230, constructing an initial population based on the equipment operation parameters and the seed supply working parameters, wherein the initial population includes a plurality of initial individuals, the genotypes of the initial individuals are the independent variables of the multi-objective regression model, and the phenotypes of the initial individuals are the equipment operation parameters.

[0103] S240. Perform partial elimination of initial individuals in the initial population based on a preset elimination mechanism to obtain an updated initial population.

[0104] In this embodiment, the preset elimination mechanism is a pre-set mechanism for eliminating initial individuals in the initial population to update the initial population. Exemplarily, the preset elimination mechanism includes but is not limited to a competition elimination mechanism between similar individuals, a multi-indicator comprehensive evaluation elimination mechanism, and an elite strategy.

[0105] Specifically, the initial individuals to be eliminated in the initial population are determined based on a preset elimination mechanism, and the updated initial population is obtained by removing the initial individuals to be eliminated from the initial population, wherein the initial individuals to be eliminated include at least some of the initial individuals in the initial population.

[0106] In some embodiments, the preset elimination mechanism includes a deviation-weighted population elimination mechanism. It should be noted that the deviation-weighted population elimination mechanism is a deviation-based multi-index comprehensive evaluation elimination mechanism. Optionally, based on the deviation limits corresponding to each parameter in the seed supply working parameter, the deviation value corresponding to the initial individual is determined, and the initial individual is eliminated based on the deviation value to obtain an updated initial population; the deviation values ​​corresponding to multiple initial individuals in the updated initial population are all less than the deviation threshold corresponding to the updated initial population; the deviation value is the distance between the phenotype of the initial individual and the preset optimal phenotype, and the deviation threshold is a preset multiple of the half interval length.

[0107] In this embodiment, the deviation limit is the median of the numerical range of each parameter in the phenotype of multiple initial individuals in the current initial population, wherein the current initial population is the initial population or the initial population that has been processed by the deviation weighted population elimination mechanism once or twice. Exemplarily, assuming that the obtained seed supply working parameters include the empty seed frequency, the seed completion time and the average seed adsorption time, the numerical range corresponding to the empty seed frequency is 0 to 2 Hz, the numerical range corresponding to the seed completion time is 5 to 10 seconds, and the numerical range corresponding to the average seed adsorption time is 0 to 0.1 seconds, then the deviation limit corresponding to the empty seed frequency is 1 Hz, the deviation limit corresponding to the seed completion time is 7.5 seconds, and the deviation limit corresponding to the average seed adsorption time is 0.05 seconds. In some embodiments, each parameter in the phenotype of the initial individuals in the initial population is normalized. Exemplarily, when the numerical range of any parameter in the phenotype is 0 to 1, the deviation limit corresponding to the parameter in the phenotype is 0.5.

[0108] Specifically, for each parameter in the phenotype of the initial individual, the absolute value of the difference between the parameter and the deviation limit corresponding to the parameter is determined as the deviation value corresponding to the parameter. Based on the deviation weights corresponding to the multiple parameters in the phenotype, the deviation values ​​corresponding to the multiple parameters in the phenotype of each initial individual are weighted and summed to obtain the deviation value corresponding to the initial individual. It can be understood that the sum of the deviation weights corresponding to the multiple parameters in the phenotype is 1.

[0109] For example, suppose The deviation function corresponding to the initial individuals is expressed as ,in, Characterizes the deviation limit corresponding to the empty frequency, Characterizes the deviation limit corresponding to the previous completion time, Characterizes the deviation limit corresponding to the average adsorption time of seeds, Characterizes the deviation weight corresponding to the empty frequency, Characterizes the deviation weight corresponding to the previous completion time, Characterizes the deviation weight corresponding to the average adsorption time of seeds, where , by Each parameter in the phenotype of the initial individual is input into the deviation function to obtain the The deviation value corresponding to the initial individuals.

[0110] The deviation threshold is the threshold of the deviation value corresponding to the initial individual, which is used to determine whether to eliminate the initial individual. The deviation threshold is a preset multiple of the half-interval length. The deviation threshold is a preset multiple of the half-interval length corresponding to the current initial population, where the half-interval length is the weighted sum of the half-value ranges of each parameter in the phenotype of multiple initial individuals in the current initial population.

[0111] For example, assuming that the preset multiple is 110%, the deviation threshold corresponding to the current initial population is: , where the numerical range of the empty species frequency in the phenotypes of multiple initial individuals in the current initial population is , the value range corresponding to the above completion time is , the value range corresponding to the average adsorption time of seeds is .

[0112] The next initial population is obtained by eliminating the initial individuals whose corresponding deviation values ​​in the current initial population are greater than or less than the deviation threshold corresponding to the current initial population. If the deviation values ​​corresponding to each initial individual in the next initial population are all less than the deviation threshold corresponding to the current initial population, it can be considered that there are no extremely deviated initial individuals in the next initial population, and the next initial population is determined as the updated initial population; if there are initial individuals in the next initial population whose corresponding deviation values ​​are all less than the deviation threshold corresponding to the current initial population, the next initial population is eliminated based on the deviation weighted population elimination mechanism until the updated initial population is determined.

[0113] The technical solution of this embodiment determines the deviation value corresponding to the initial individual based on the deviation limits corresponding to each parameter in the seed supply working parameters, and eliminates the initial individuals based on the deviation value to obtain an updated initial population; the deviation values ​​corresponding to multiple initial individuals in the updated initial population are all smaller than the deviation threshold corresponding to the updated initial population, and the initial individuals in the initial population can be eliminated based on the deviation weighted population elimination mechanism, so that there are no extremely deviated initial individuals in the updated initial population, which helps to increase the number of individuals on the Pareto frontier generated by subsequent genetic evolution based on the updated initial population, thereby accelerating the subsequent optimization speed based on the wolf pack search algorithm and reducing the amount of calculation.

[0114] In some embodiments, the preset elimination mechanism includes a secondary weighted population elimination mechanism. It should be noted that the secondary weighted population elimination mechanism is a competitive elimination mechanism between individuals of the same type. Optionally, the initial parent population is subjected to a fast non-dominated sort to obtain an initial child population; wherein the initial parent population is an initial population or an updated initial population obtained after at least one elimination of the initial population; the objective function values ​​of the initial parent individuals in the initial parent population and the objective function values ​​of the initial child individuals in the initial child population are determined based on a multi-objective regression model; for initial parent individuals and initial child individuals with the same objective function values, local individuals are eliminated based on the priority of the independent variables of the multi-objective regression model to obtain an updated initial population.

[0115] In this embodiment, the objective function value is a numerical value determined based on the phenotype of the initial individual, and is used to characterize the performance of the seed supply device corresponding to the initial individual.

[0116] Specifically, the initial population or the updated initial population obtained after at least one elimination of the initial population is determined as the initial parent population, and the initial parent population is quickly non-dominated sorted based on the preset non-dominated sorting genetic algorithm to obtain the initial offspring population. Based on the target weights corresponding to the multiple parameters in the phenotype, the multiple parameters in the phenotype of each initial parent individual are weighted summed to obtain the target function value corresponding to each initial parent individual, and the multiple parameters in the phenotype of each initial offspring individual are weighted summed to obtain the target function value corresponding to each initial offspring individual, wherein the initial parent individual is the initial individual constituting the initial parent population, and the initial offspring individual is the initial individual constituting the initial offspring population. In some embodiments, the weights corresponding to the multiple parameters in the phenotype can be determined based on response surface analysis, and in the response surface analysis process, each parameter in the seed supply working parameter is set as the target variable; in some embodiments, the weights corresponding to the parameters in the phenotype can be preset weights, wherein the preset weights are set by the user based on experience, and this embodiment does not limit this.

[0117] For example, suppose The objective function value of the initial individual is expressed as ,in, To characterize the target weight corresponding to the empty frequency, To represent the target weight corresponding to the above completion time, is the target weight corresponding to the average adsorption time of seeds.

[0118] The initial parent individuals and initial offspring individuals with the same objective function value are determined as individuals of the same type, and the initial parent individuals and initial offspring individuals of the same type are partially eliminated based on the secondary weighted population elimination mechanism, so that the same type of individuals only includes one initial parent individual or one initial offspring individual. The initial population is updated based on the initial parent individuals and / or initial offspring individuals that have not been eliminated.

[0119] In some embodiments, for the initial parent individuals and initial offspring individuals with the same objective function value, the elimination sequence numbers corresponding to the initial parent individuals and initial offspring individuals with the same objective function value are determined according to the priority of the independent variables of the multi-objective regression model. Among them, the priority of the independent variables of the multi-objective regression model is the height of the center of gravity of the seed pile, the rotation speed of the seeding disk, and the vacuum degree.

[0120] It should be noted that, when the phenotype includes one or more of the seed disk rotation speed, vacuum degree and seed pile gravity center height, for any two initial parent individuals or initial offspring individuals, or any initial parent individual and initial offspring individual, when the phenotype includes the seed pile gravity center height, the larger the seed pile gravity center height is, the larger the corresponding elimination number is; when the seed pile gravity center height is the same or the phenotype does not include the seed pile gravity center height, the larger the corresponding elimination number is; when the seed pile gravity center height is the same and the seed disk rotation speed is the same, the phenotype does not include the seed pile gravity center height and the seed disk rotation speed is the same, or the phenotype does not include the seed pile gravity center height and the seed disk rotation speed, the larger the vacuum degree is, the larger the elimination number is. The initial population is updated by eliminating the initial parent individuals or initial offspring individuals corresponding to the other elimination numbers except the minimum elimination number.

[0121] The technical solution of this embodiment, by eliminating local individuals based on the priority of the independent variables of the multi-objective regression model for initial parent individuals and initial offspring individuals with the same objective function value, obtains an updated initial population, which can ensure that each objective function value in the updated initial population corresponds to only one initial parent individual or one initial offspring individual, and for any objective function value, the performance of the seed supply device represented by the retained initial parent individual or initial offspring individual is better than the performance of the seed supply device represented by the eliminated initial parent individual or initial offspring individual, which helps to increase the number of individuals on the Pareto front generated by subsequent genetic evolution based on the updated initial population, thereby accelerating the subsequent optimization speed based on the wolf pack search algorithm and reducing the amount of calculation.

[0122] In some embodiments, the initial individuals in the initial population are eliminated based on the bias weighted population elimination mechanism to obtain a first updated initial population. The initial individuals in the first updated population are eliminated based on the elite strategy to obtain a second updated initial population. The initial offspring individuals and initial parent individuals corresponding to the second updated initial population are eliminated based on the secondary weighted population elimination mechanism to obtain a third updated initial population, wherein the third updated initial population is an updated initial population for genetic evolution. Among them, the initial individuals in the first updated population are eliminated based on the elite strategy to obtain the second updated initial population, including: performing a fast non-dominated sort on the individuals in the first updated population to obtain the Pareto front level of each individual. It can be understood that the smaller the Pareto front level of any individual, the closer the individual is to the optimal solution. When the Pareto front levels of any two individuals are equal, the crowding distance and crowding degree of the two individuals on the same Pareto front are determined, and the individual with the smaller crowding distance of the two individuals is eliminated, and the individual with the larger crowding distance of the two individuals is retained.

[0123] S250. Based on the multi-objective regression model, the initial population is iteratively updated to obtain a population to be optimized.

[0124] Specifically, the initial population is genetically iterated based on a preset non-dominated sorting genetic algorithm to obtain an evolving population. In some embodiments, the following steps are optionally performed iteratively, and when the number of dominant individuals in the update population corresponding to the current genetic iteration number is greater than the first preset multiple of the number of individuals in the update population corresponding to the current genetic iteration number, the update population corresponding to the current genetic iteration number is determined to be the population to be optimized: the fitness corresponding to each individual in the current population to be updated is determined based on a preset fitness function; the current population to be updated is the update population corresponding to the updated initial population or the previous genetic iteration number; based on the fitness corresponding to multiple individuals in the current population to be updated, local individuals in the current population to be updated are selected for hybridization and mutation to obtain multiple offspring individuals corresponding to the current population to be updated; individuals in the current population to be updated are merged with multiple offspring individuals corresponding to the current population to be updated to obtain the current merged population; the Pareto frontier level and crowding distance corresponding to multiple individuals in the current merged population are determined; the dominant individuals in the current merged population are determined based on the Pareto frontier level and crowding distance; based on the multiple dominant individuals, the update population corresponding to the current genetic iteration number is constructed.

[0125] In this embodiment, when the current number of genetic iterations is 1, the current population to be updated is the updated initial population; when the current number of genetic iterations is greater than 1, the current population to be updated is the updated population corresponding to the previous number of genetic iterations. For example, , the current population to be updated is the number of genetic iterations The corresponding updated population, where .

[0126] Specifically, under the current number of genetic iterations, the preset fitness function is a preset fitness function, which is used to calculate the fitness based on one or more of the empty seed frequency, the seed completion time, and the average seed adsorption time. By inputting at least one of the multiple parameters in the phenotype of each individual in the current population to be updated into the preset fitness function, the fitness corresponding to each individual in the current population to be updated is obtained. Exemplarily, the preset fitness function is expressed as , that is, the first The fitness corresponding to each individual is Average adsorption time of seeds per individual The opposite number of , which makes the larger the average adsorption time of seeds, the smaller the fitness, ensuring that the smaller the fitness, the closer the corresponding individual is to the optimal solution.

[0127] By selecting two or more individuals whose corresponding fitness in the current population to be updated is less than the preset fitness threshold, multiple parent individuals are obtained. By performing hybridization mutation on multiple parent individuals, multiple new individuals are obtained, and the new individuals are determined as the offspring individuals corresponding to the current population to be updated. Among them, the preset fitness threshold is a preset fitness threshold, which is used to determine that an individual in the current population to be updated is a parent individual. Exemplarily, the preset fitness threshold can be the average fitness of multiple individuals in the current population to be updated.

[0128] By combining multiple individuals in the current population to be updated and multiple offspring individuals corresponding to the current population to be updated, the parent individuals and the offspring individuals are merged to obtain the current merged population. The current merged population is a population including parent individuals and offspring individuals corresponding to the current number of genetic iterations. By performing fast non-dominated sorting on multiple individuals in the current merged population, the Pareto front levels corresponding to multiple individuals in the current merged population are determined, and the crowding distance of each individual in the current generation merged population is determined. Based on the order from small to large, the Pareto front levels are traversed, and based on the order from large to small crowding distances, the individuals corresponding to the currently traversed Pareto front levels in the current generation merged population are selected in turn until the number of selected individuals is equal to the number of individuals in the first generation population, and each selected individual is determined as a dominant individual. By combining multiple dominant individuals, an update population corresponding to the current number of genetic iterations is constructed.

[0129] By performing fast non-dominated sorting on multiple individuals in the update population corresponding to the current number of genetic iterations, the Pareto front level of each individual is obtained, and the individuals belonging to the first Pareto front level in the update population are determined as dominant level individuals.

[0130] If the number of dominant individuals in the update population corresponding to the current genetic iteration number is less than or equal to the first preset multiple of the number of individuals in the update population, it can be considered that the update population corresponding to the current genetic iteration number cannot determine the direction of the Pareto frontier, and then enter the next genetic iteration number. If the number of dominant individuals in the update population corresponding to the current genetic iteration number is greater than the first preset multiple of the number of individuals in the update population, it can be considered that the update population corresponding to the current genetic iteration number can determine the direction of the Pareto frontier, and then do not enter the next genetic iteration number, but determine the update population corresponding to the current genetic iteration number as the population to be optimized. Among them, the first preset multiple is less than the second preset multiple, and exemplarily, the first preset multiple is 0.7.

[0131] The technical solution of this embodiment is to iteratively execute the genetic evolution steps of updating the initial population, and when the number of dominant individuals in the updated population corresponding to the current number of genetic iterations is greater than the first preset multiple of the number of individuals in the updated population corresponding to the current number of genetic iterations, determine that the updated population corresponding to the current number of genetic iterations is the population to be optimized. When the number of dominant individuals in the updated population is greater than the first preset multiple of the number of individuals in the updated population, the genetic evolution can be stopped, which can avoid falling into a local optimal solution and reduce the consumption of computing resources. It can also provide sufficient dominant individuals for the subsequent optimization process based on the wolf pack search algorithm, which is helpful to improve the performance of determining the working parameters of the seed supply device.

[0132] S260. Based on the wolf pack search algorithm, the target population is optimized to obtain the target population; wherein the target population satisfies the search end condition.

[0133] S270. Determine target operating parameters corresponding to the seed supply device based on the target population.

[0134] The technical solution of this embodiment is to obtain an updated initial population by locally eliminating initial individuals in the initial population based on a preset elimination mechanism, and to obtain a population to be optimized by genetically evolving the updated initial population based on a multi-objective regression model. Compared with genetically evolving the initial population directly based on the multi-objective regression model, the number of initial appropriate individuals in the updated initial population is less than the number of initial appropriate individuals in the initial population, thereby reducing the amount of calculation. Compared with the initial population, the updated initial population is closer to the optimal solution as a whole, which helps to increase the number of individuals on the Pareto frontier generated by subsequent genetic evolution based on the updated initial population, thereby accelerating the subsequent optimization speed based on the wolf pack search algorithm and reducing the amount of calculation.

[0135] Embodiment 3

[0136] Figure 5Schematic diagram of a device for determining working parameters of a seed supply device provided in Embodiment 3 of the present invention. Figure 5 As shown, the device comprises:

[0137] Parameter acquisition module 310, used to acquire equipment operation parameters and seed supply working parameters of the seed supply device during the working process; the equipment operation parameters include one or more items of seed plate rotation speed, vacuum degree and seed pile gravity center height; the seed supply working parameters include one or more items of empty seed frequency, seeding completion time and seed average adsorption time;

[0138] A multi-objective regression model building module 320, for building a multi-objective regression model based on the equipment operation parameters and the seed supply operation parameters, wherein the independent variables of the multi-objective regression model include at least part of the seed supply operation parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters;

[0139] An initial population construction module 330 is used to construct an initial population based on equipment operation parameters and seed supply working parameters. The initial population includes multiple initial individuals. The genotypes of the initial individuals are independent variables of the multi-objective regression model, and the phenotypes of the initial individuals are equipment operation parameters.

[0140] The genetic evolution module 340 is used to perform genetic evolution on the initial population based on the multi-objective regression model to obtain the population to be optimized; wherein, in the process of genetic evolution, the Pareto front direction is determined based on the evolving population to obtain the population to be optimized that satisfies the Pareto front direction;

[0141] The target population determination module 350 is used to optimize the population to be optimized based on the wolf pack search algorithm to obtain the target population; wherein the target population meets the end search condition;

[0142] The target working parameter determination module 360 ​​is used to determine the target working parameters corresponding to the seed supply device based on the target population.

[0143] The technical solution of this embodiment determines the direction of the Pareto front based on the evolving population in the process of genetically evolving the initial population based on the multi-objective regression model, obtains the population to be optimized that satisfies the direction of the Pareto front, and optimizes the population to be optimized based on the wolf pack search algorithm to obtain the target population, so as to determine the target working parameters corresponding to the seed supply device based on the target population, instead of genetically evolving the population to be optimized to obtain the target population after obtaining the population to be optimized. This solves the problem of low performance of determining the target working parameters based on the genetic algorithm alone, and by combining the genetic algorithm and the wolf pack search algorithm, it is possible to avoid falling into the local optimal solution at the end of the genetic evolution. The wolf pack search algorithm improves the probability of finding the global optimal solution by combining local search with global search, and improves the speed and stability of determining the working parameters of the seed supply device.

[0144] On the basis of the above-mentioned embodiment, optionally, the device for determining the working parameters of the seed supply device also includes: a multi-objective regression model independent variable determination module; wherein the multi-objective regression model independent variable determination module is used to perform a response surface analysis on the equipment operating parameters based on the seed supply working parameters as the target variables; determine the significance data of each parameter in the equipment operating parameters based on the response surface analysis results; and determine the independent variables of the multi-objective regression model based on the significance data of each parameter in the equipment operating parameters.

[0145] On the basis of the above embodiment, optionally, the genetic evolution module 340 is specifically used to perform local elimination of initial individuals in the initial population based on a preset elimination mechanism to obtain an updated initial population; based on a multi-objective regression model, perform genetic evolution on the updated initial population to obtain a population to be optimized.

[0146] On the basis of the above embodiment, optionally, the genetic evolution module 340 is also used to determine the deviation value corresponding to the initial individual based on the deviation limits corresponding to each parameter in the seed supply working parameters, and eliminate the initial individuals based on the deviation value to obtain an updated initial population; the deviation values ​​corresponding to multiple initial individuals in the updated initial population are all smaller than the deviation threshold corresponding to the updated initial population; the deviation value is the distance between the phenotype of the initial individual and the preset optimal phenotype, and the deviation threshold is a preset multiple of the half interval length.

[0147] On the basis of the above embodiment, optionally, the genetic evolution module 340 is also used to perform fast non-dominated sorting on the initial parent population to obtain an initial child population; wherein the initial parent population is the initial population or an updated initial population obtained after at least one elimination of the initial population; the objective function values ​​of the initial parent individuals in the initial parent population and the objective function values ​​of the initial child individuals in the initial child population are determined based on a multi-objective regression model; for the initial parent individuals and initial child individuals with the same objective function values, local individuals are eliminated based on the priority of the independent variables of the multi-objective regression model to obtain an updated initial population.

[0148] On the basis of the above embodiment, optionally, the genetic evolution module 340 is further used to iteratively execute the following steps: when the number of dominant individuals in the update population corresponding to the current genetic iteration number is greater than the first preset multiple of the number of individuals in the update population corresponding to the current genetic iteration number, the update population corresponding to the current genetic iteration number is determined to be the population to be optimized: the fitness corresponding to each individual in the current population to be updated is determined based on a preset fitness function; the current population to be updated is the update population corresponding to the updated initial population or the previous genetic iteration number; based on the fitness corresponding to multiple individuals in the current population to be updated, local individuals in the current population to be updated are selected for hybridization and mutation to obtain multiple offspring individuals corresponding to the current population to be updated; individuals in the current population to be updated are merged with multiple offspring individuals corresponding to the current population to be updated to obtain the current merged population; Pareto frontier levels and crowding distances corresponding to multiple individuals in the current merged population are determined; dominant individuals in the current merged population are determined based on the Pareto frontier level and the crowding distance; based on the multiple dominant individuals, an update population corresponding to the current genetic iteration number is constructed.

[0149] On the basis of the above embodiment, optionally, the end search condition is that the number of dominant individuals in the target population is greater than a second preset multiple of the number of individuals in the target population, wherein the second preset multiple is greater than the first preset multiple; accordingly, the target population determination module 350 is specifically used to iteratively execute the following steps, and when the updated population to be optimized corresponding to the current search iteration number satisfies the end search condition, the updated population to be optimized corresponding to the current search iteration number is determined as the target population: an initial wolf pack is constructed based on the current population to be optimized; the current population to be optimized is the population to be optimized or the updated population to be optimized corresponding to the previous search iteration number; based on the prey concentration function values ​​corresponding to the multiple artificial wolves included in the initial wolf pack, the positions of the multiple artificial wolves are updated; the prey concentration function value is determined based on the parameters in the seed supply working parameters corresponding to the artificial wolves; based on the multiple artificial wolves after the position update, an updated population to be optimized corresponding to the current search iteration number is constructed.

[0150] The device for determining the working parameters of a seed supply device provided in an embodiment of the present invention can execute the method for determining the working parameters of a seed supply device provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0151] Embodiment 4

[0152] Figure 61 is a schematic diagram of the structure of an electronic device for implementing the method for determining the working parameters of the device of the embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0153] like Figure 6 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0154] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0155] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a method for determining working parameters of a device.

[0156] In some embodiments, the method for determining the working parameters of the seed supply device can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for determining the working parameters of the seed supply device described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the method for determining the working parameters of the seed supply device by any other appropriate means (for example, by means of firmware).

[0157] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0158] The computer program for implementing the method for determining the working parameters of the seed device of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0159] Embodiment 5

[0160] Embodiment 5 of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute a method for determining working parameters of a seeding device, the method comprising:

[0161] The equipment operation parameters and seed supply working parameters of the seed supply device in the working process are obtained; the equipment operation parameters include one or more of the rotation speed of the seed plate, the vacuum degree and the height of the center of gravity of the seed pile, and the seed supply working parameters include one or more of the empty seed frequency, the seeding completion time and the average adsorption time of the seeds; a multi-objective regression model is constructed based on the equipment operation parameters and the seed supply working parameters, the independent variables of the multi-objective regression model include at least part of the seed supply working parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters; an initial population is constructed based on the equipment operation parameters and the seed supply working parameters, the initial population includes a plurality of initial individuals, the genotype of the initial individual is the independent variable of the multi-objective regression model, and the phenotype of the initial individual is the equipment operation parameter; based on the multi-objective regression model, the initial population is genetically evolved to obtain a population to be optimized; wherein, in the process of genetic evolution, the Pareto front direction is determined based on the evolving population, and the population to be optimized that meets the Pareto front direction is obtained; based on the wolf pack search algorithm, the population to be optimized is optimized to obtain a target population; wherein, the target population meets the end search condition; based on the target population, the target working parameters corresponding to the seed supply device are determined.

[0162] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0163] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0164] The systems and techniques described herein may be implemented in a computing system that includes a backend (e.g., as a data server), or a computing system that includes middleware (e.g., an application server), or a computing system that includes a frontend (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backends, middleware, or frontends. The systems may be interconnected by digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0165] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0166] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0167] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining working parameters of a seed supply device, characterized in that: include: Obtaining equipment operating parameters and seed supply working parameters of the seed supply device during the working process; the equipment operating parameters include one or more of the rotation speed of the seeding disc, the vacuum degree and the height of the center of gravity of the seed pile; the seed supply working parameters include one or more of the empty seed frequency, the seeding completion time and the average seed adsorption time; Constructing a multi-objective regression model based on the equipment operation parameters and the seed supply operation parameters, wherein the independent variables of the multi-objective regression model include at least part of the seed supply operation parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters; Constructing an initial population based on the equipment operation parameters and the seed supply working parameters, wherein the initial population includes a plurality of initial individuals, the genotypes of the initial individuals are the independent variables of the multi-objective regression model, and the phenotypes of the initial individuals are the equipment operation parameters; Based on the multi-objective regression model, the initial population is genetically evolved to obtain a population to be optimized; wherein, in the process of the genetic evolution, the Pareto front direction is determined based on the evolving population to obtain a population to be optimized that satisfies the Pareto front direction; Based on the wolf pack search algorithm, the population to be optimized is optimized to obtain a target population; wherein the target population satisfies an end search condition; the end search condition is that the number of dominant individuals in the target population is greater than a second preset multiple of the number of individuals in the target population, wherein the second preset multiple is greater than the first preset multiple; Determining target working parameters corresponding to the seed supply device based on the target population; The step of optimizing the population to be optimized to obtain a target population based on a wolf pack search algorithm includes: The following steps are iteratively performed, and when the updated population to be optimized corresponding to the current search iteration number satisfies the end search condition, the updated population to be optimized corresponding to the current search iteration number is determined as the target population: Constructing an initial wolf pack based on the current population to be optimized; the current population to be optimized is the population to be optimized or the updated population to be optimized corresponding to the previous search iteration number; Based on the prey concentration function values ​​corresponding to the multiple artificial wolves included in the initial wolf pack, the positions of the multiple artificial wolves are updated; the prey concentration function values ​​are determined based on the parameters of the seed supply working parameters corresponding to the artificial wolves; Based on the multiple artificial wolves after the positions are updated, the updated population to be optimized corresponding to the current search iteration number is constructed.

2. The method according to claim 1, characterized in that Before building a multi-objective regression model, it also includes: Based on the seed supply working parameters as target variables, a response surface analysis is performed on the equipment operating parameters; The significance data of each parameter in the equipment operation parameters are determined based on the response surface analysis results; and the independent variables of the multi-objective regression model are determined based on the significance data of each parameter in the equipment operation parameters.

3. The method according to claim 1, characterized in that The step of performing genetic evolution on the initial population based on the multi-objective regression model to obtain a population to be optimized includes: Partially eliminating initial individuals in the initial population based on a preset elimination mechanism to obtain an updated initial population; Based on the multi-objective regression model, the updated initial population is genetically evolved to obtain the population to be optimized.

4. The method according to claim 3, characterized in that The locally eliminating the initial individuals in the initial population based on the preset elimination mechanism to obtain an updated initial population includes: Based on the deviation limits corresponding to the parameters in the seed supply working parameters, the deviation values ​​corresponding to the initial individuals are determined, and the initial individuals are eliminated based on the deviation values ​​to obtain the updated initial population; the deviation values ​​corresponding to the multiple initial individuals in the updated initial population are all smaller than the deviation threshold corresponding to the updated initial population; The deviation value is the distance between the phenotype of the initial individual and the preset optimal phenotype, and the deviation threshold is a preset multiple of the half-interval length.

5. The method according to claim 3 or 4, characterized in that: The locally eliminating the initial individuals in the initial population based on the preset elimination mechanism to obtain an updated initial population includes: Performing a fast non-dominated sort on the initial parent population to obtain an initial child population; wherein the initial parent population is the initial population or an updated initial population obtained after the initial population is eliminated at least once; Based on the multi-objective regression model, the objective function values ​​of the initial parent individuals in the initial parent population and the objective function values ​​of the initial offspring individuals in the initial offspring population are determined; for the initial parent individuals and the initial offspring individuals with the same objective function values, local individuals are eliminated based on the priority of the independent variables of the multi-objective regression model to obtain the updated initial population.

6. The method according to claim 3, characterized in that: The step of performing genetic evolution on the updated initial population based on the multi-objective regression model to obtain the population to be optimized includes: The following steps are iteratively performed, and when the number of dominant individuals in the update population corresponding to the current genetic iteration number is greater than a first preset multiple of the number of individuals in the update population corresponding to the current genetic iteration number, the update population corresponding to the current genetic iteration number is determined to be the population to be optimized: Determine the fitness corresponding to each individual in the current population to be updated based on a preset fitness function; the current population to be updated is the updated initial population or the updated population corresponding to the last genetic iteration number; Based on the fitness values ​​of the multiple individuals in the current population to be updated, local individuals in the current population to be updated are selected for hybridization and mutation to obtain multiple offspring individuals corresponding to the current population to be updated; Merging the individuals in the current population to be updated with a plurality of offspring individuals corresponding to the current population to be updated to obtain a current merged population; Determine the Pareto frontier levels and crowding distances corresponding to the multiple individuals in the current merged population; Determining a dominant individual in the current merged population based on the Pareto front level and the crowding distance; Based on the plurality of dominant individuals, the updated population corresponding to the current number of genetic iterations is constructed.

7. A device for determining working parameters of a seed supply device, characterized in that: include: A parameter acquisition module is used to acquire equipment operation parameters and seed supply working parameters of the seed supply device during the working process; the equipment operation parameters include one or more of the rotation speed of the seeding disc, the vacuum degree and the height of the center of gravity of the seed pile; the seed supply working parameters include one or more of the empty seed frequency, the seeding completion time and the average seed adsorption time; A multi-objective regression model building module, used to build a multi-objective regression model based on the equipment operation parameters and the seed supply operation parameters, wherein the independent variables of the multi-objective regression model include at least part of the seed supply operation parameters, and the dependent variables of the multi-objective regression model include the equipment operation parameters; An initial population construction module, used to construct an initial population based on the equipment operation parameters and the seed supply working parameters, wherein the initial population includes a plurality of initial individuals, the genotypes of the initial individuals are the independent variables of the multi-objective regression model, and the phenotypes of the initial individuals are the equipment operation parameters; A genetic evolution module, used to perform genetic evolution on the initial population based on the multi-objective regression model to obtain a population to be optimized; wherein, in the process of the genetic evolution, the Pareto front direction is determined based on the evolving population to obtain a population to be optimized that satisfies the Pareto front direction; A target population determination module is used to optimize the population to be optimized based on a wolf pack search algorithm to obtain a target population; wherein the target population satisfies a search termination condition; the search termination condition is that the number of dominant individuals in the target population is greater than a second preset multiple of the number of individuals in the target population, wherein the second preset multiple is greater than the first preset multiple; A target working parameter determination module, used to determine the target working parameters corresponding to the seed supply device based on the target population; The target population determination module is specifically used to iteratively perform the following steps, and when the updated population to be optimized corresponding to the current search iteration number satisfies the end search condition, the updated population to be optimized corresponding to the current search iteration number is determined as the target population: Constructing an initial wolf pack based on the current population to be optimized; the current population to be optimized is the population to be optimized or the updated population to be optimized corresponding to the previous search iteration number; Based on the prey concentration function values ​​corresponding to the multiple artificial wolves included in the initial wolf pack, the positions of the multiple artificial wolves are updated; the prey concentration function values ​​are determined based on the parameters of the seed supply working parameters corresponding to the artificial wolves; Based on the multiple artificial wolves after the positions are updated, the updated population to be optimized corresponding to the current search iteration number is constructed.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the working parameters of the seed supply device according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for determining working parameters of a seed supply device according to any one of claims 1 to 6 when executed.