Large-scale farmland scale design method based on double-target constraint

Abstract

The invention provides a large-scale farmland scale design method based on double-target constraint, and relates to the technical field of farmland standardized construction, and the method comprises the steps: carrying out the fitting of a farmland actual irrigation sample and an agricultural machinery operation actual sample, and obtaining an irrigation performance prediction model and an agricultural machinery operation efficiency prediction model; judging actual constraint conditions, and if the actual constraint conditions do not exist, performing simulation by utilizing the paddy field specification prediction model to obtain a paddy field specification prediction data sample so as to obtain a paddy field specification suitable range; and if the actual constraint condition exists, based on the irrigation performance prediction data sample and the agricultural machine operation efficiency prediction data sample, performing multi-objective optimization and global random search by using the paddy field specification optimization model to obtain a suitable paddy field specification, and analyzing a paddy field specification suitable range and the suitable paddy field specification to obtain a farmland scale design result. The problem that an existing farmland design technology lacks large scale, whole-process mechanization and high intensification is solved.

Description

TECHNICAL FIELD

[0001] The present specification relates to the technical field of standardization construction of farmland, and particularly relates to a large-scale farmland scale design method based on double-target constraints. BACKGROUND

[0002] Agricultural water resource shortage and low production efficiency are the main problems of regional irrigation agriculture development. Urbanization development makes rural labor force increasingly scarce. Based on the current situation, it has become an inevitable trend for the agricultural industry in the main grain-producing areas to develop towards "moderate scale, full mechanization and high intensity". At present, the reasonable layout and standardization construction of large-scale paddy field irrigation systems lack corresponding technical support. SUMMARY

[0003] In view of the above shortcomings in the prior art, the large-scale farmland scale design method based on double-target constraints provided by the present application solves the problem that the existing farmland design technology lacks scale, full mechanization and high intensity.

[0004] In order to achieve the above-mentioned purpose of the application, the technical scheme adopted by the present application is as follows: a large-scale farmland scale design method based on double-target constraints, comprising: S1: using an artificial neural network model to fit the actual irrigation samples of farmland and the actual operation samples of agricultural machinery, respectively obtaining an irrigation performance prediction model and an agricultural machinery operation efficiency prediction model; wherein the irrigation performance prediction model is used to obtain irrigation performance prediction data samples by simulation, and the agricultural machinery operation efficiency prediction model is used to obtain agricultural machinery operation efficiency prediction data samples by simulation; S2: judging the actual constraint condition, if there is no actual constraint condition, constructing a paddy field specification prediction model and entering S3; if there is an actual constraint condition, constructing a paddy field specification optimization model and entering S5; S3: normalizing and setting weight coefficients for the irrigation performance indicators in the irrigation performance prediction data samples and the agricultural machinery operation efficiency indicators in the agricultural machinery operation efficiency prediction data samples, respectively, simulating using the paddy field specification prediction model to obtain paddy field specification prediction data samples with paddy field system performance; S4: selecting data exceeding the index threshold in the paddy field specification prediction data samples as non-inferior solutions, combining the non-inferior solutions to obtain a suitable range of paddy field specifications; S5: based on the irrigation performance prediction data samples and the agricultural machinery operation efficiency prediction data samples, using the paddy field specification optimization model to perform multi-objective optimization and global random search to obtain a suitable paddy field specification; S6: analyzing the suitable range of paddy field specifications and the suitable paddy field specification to obtain a farmland scale design result, and completing the design of the scale of the large-scale farmland.

[0005] The application has the beneficial effects that the application provides a large-scale farmland scale design method based on double-target constraints, combines an irrigation performance prediction model and a farm machine operation efficiency prediction model to determine optimization design variables, for a scenario without constraint conditions, a paddy field specification prediction model is constructed, and a suitable range of paddy field specifications with high system performance is proposed, the result can be widely applied to rice planting areas in cold regions, for a specific region, a local field channel system and a road network status are taken as constraint conditions, a paddy field specification optimization model is constructed, and a paddy field specification with high system performance suitable for the local region is proposed based on an intelligent optimization algorithm. In this way, the farmland scale design can be more in line with the actual situation.

[0006] Further, the expression of the paddy field specification system performance is: ; Among them, represents the paddy field system performance, represents the weight of the normalized irrigation performance index, represents the normalized irrigation performance index, represents the weight of the normalized farm machine operation efficiency index, represents the normalized farm machine operation efficiency index.

[0007] Further, the expressions of the normalized irrigation performance index and the normalized farm machine operation efficiency index are respectively: ; ; Among them, represents the normalized irrigation performance index, represents the maximum value of the irrigation performance index, represents the irrigation performance index value under a typical working condition, represents the minimum value of the irrigation performance index, represents the normalized farm machine operation efficiency index, represents the farm machine operation efficiency index value under a typical working condition, represents the minimum value of the farm machine operation efficiency index, represents the maximum value of the farm machine operation efficiency index.

[0008] The two different evaluation indexes are normalized respectively, which is helpful for subsequent construction and solution of the objective function.

[0009] Further, the expression of the multi-objective optimization is: ; Among them, represents a maximum value function, represents the paddy field system performance, represents irrigation efficiency, represents field length, represents field width, represents irrigation flow, represents farm operation efficiency, represents farm width, represents maximum field length, represents maximum field width, represents maximum irrigation flow.

[0010] Further, the suitable paddy field specifications are solved by an intelligent optimization algorithm, by tracking the historical optimal position found by itself and the global optimal position shared by the group, wherein the expression of the velocity and position of particle optimization is: ; ; wherein, represents the velocity of the idth particle in the t+1th iteration, represents an inertia factor, represents the velocity of the idth particle in the tth iteration, represents a first learning factor, represents a first random constant, represents a population individual extreme value, represents the position of the idth particle in the tth iteration, represents a second learning factor, represents a second random constant, represents a population global extreme value, represents the position of the gdth particle in the tth iteration.

[0011] Further, the expression of the inertia weight and the learning factor is respectively: ; ; wherein, represents an inertia weight, represents a maximum weight, represents a minimum weight, represents a current iteration number, represents a maximum iteration number, represents the jth learning factor, represents the jth maximum learning factor, represents the jth minimum learning factor. BRIEF DESCRIPTION OF DRAWINGS

[0012] The present specification will be further explained in the way of example embodiments, which will be described in detail with reference to the attached drawings. These embodiments are not limiting, in these embodiments, the same numbers are used for the same structures and / or the same functions, wherein: Figure 1 is an exemplary flowchart of a method for designing scale of large-scale farmland based on double-target constraints according to some embodiments of the present specification; Figure 2 is an exemplary schematic diagram of a technical route for optimizing water field specification according to some embodiments of the present specification; Figure 3 is an exemplary schematic diagram of a suitable range of water field specification under no constraints according to some embodiments of the present specification; Figure 4 is an exemplary schematic diagram of fitness variation trend of global random search algorithm training process according to some embodiments of the present specification. DETAILED DESCRIPTION

[0013] The specific embodiments of the present application are described below to facilitate the understanding of the present application for those skilled in the art, but it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, it is obvious that various changes are within the spirit and scope of the present application defined and determined by the appended claims, and all the applications utilizing the concept of the present application are within the scope of protection.

[0014] EMBODIMENTS Figure 1 is an exemplary flowchart of a method for designing scale of large-scale farmland based on double-target constraints according to some embodiments of the present specification. As shown in Figure 1 , the flow includes the following steps. In some embodiments, the flow can be executed by a processor.

[0015] In some embodiments, as Figure 2As shown, the scale of the farmland scale needs to be targeted at water saving, energy saving, emission reduction and labor saving, among which the irrigation performance reflects water saving and energy saving (power saving), and the agricultural machinery operation efficiency reflects energy saving (diesel saving), emission reduction (emission reduction) and labor saving (labor saving). Therefore, the optimal design of the paddy field specification is a double target optimization design of the field technical elements including irrigation performance and agricultural machinery operation efficiency. The field technical elements involved in irrigation performance include field length, field width and irrigation flow, the field technical elements involved in agricultural machinery operation efficiency include field length, field width and agricultural machinery width, and irrigation performance and agricultural machinery operation efficiency are connected through field length and field width. For the scenarios of unclear constraint conditions, new farm or irrigation area planning, etc., the paddy field specification prediction model is constructed, and the non-inferior solution is used to determine the suitable range of the paddy field specification with high system performance. For the scenarios of clear constraint conditions, the paddy field specification optimization model is constructed, the weight coefficient method is used to construct a single target mathematical model, and the related intelligent optimization algorithm (particle swarm optimization algorithm) for solving multi-target optimization problems is used to obtain the paddy field specification with high system performance.

[0016] S1: fitting the actual irrigation sample of the farmland and the actual sample of the agricultural machinery operation by using the artificial neural network model to obtain the irrigation performance prediction model and the agricultural machinery operation efficiency prediction model respectively; wherein the irrigation performance prediction model is used to obtain the irrigation performance prediction data sample by simulation, and the agricultural machinery operation efficiency prediction model is used to obtain the agricultural machinery operation efficiency prediction data sample by simulation.

[0017] The actual irrigation sample of the farmland is the simulation value of the water dynamics model of paddy field irrigation.

[0018] The water dynamics model of paddy field irrigation is an artificial neural network model (ANN model) for simulating the irrigation scene of the paddy field.

[0019] The actual sample of agricultural machinery operation is the simulation value of the agricultural machinery operation efficiency model.

[0020] The agricultural machinery operation efficiency model is an artificial neural network model for simulating the agricultural machinery operation scene.

[0021] The irrigation performance prediction model is a mathematical model for predicting irrigation performance. For example, the expression of the irrigation performance prediction model can be: ; wherein, represents the irrigation performance prediction index value, represents the irrigation performance function, represents the field length, represents the field width, represents the irrigation flow.

[0022] In some embodiments, the processor can use the simulation values of the water field irrigation hydrodynamic model as actual samples to fit an irrigation performance prediction model about the field length, the field width, the irrigation flow rate, and the corresponding irrigation performance by using an ANN model.

[0023] The agricultural machine operation efficiency prediction model is a mathematical model for predicting the agricultural machine operation efficiency. For example, the expression of the agricultural machine operation efficiency model can be: ; wherein, represents the agricultural operation efficiency prediction index value, represents the agricultural machine operation efficiency function.

[0024] In some embodiments, the processor can use the simulation values of the agricultural machine operation efficiency model as actual samples to fit an agricultural machine operation efficiency prediction model about the field length, the field width, the agricultural machine width, and the corresponding agricultural machine operation efficiency by using an ANN model.

[0025] The irrigation performance prediction data sample is the predicted field length, field width, and irrigation flow rate data.

[0026] The agricultural machine operation efficiency prediction data sample is the predicted field length, field width, and agricultural machine width data.

[0027] In some embodiments, a water field irrigation process monitoring test is carried out, the constructed water field irrigation hydrodynamic model is calibrated and verified, the water field irrigation process of different water field specifications (length and width), irrigation flow rates, and water outlet positions is simulated by using the water field irrigation hydrodynamic model, the irrigation performance of each combination is calculated, and a model simulation data set is formed. A prediction model about the water field irrigation parameters and the irrigation performance is fitted by using an ANN or other black box model.

[0028] In some embodiments, an agricultural machine operation process monitoring test is carried out, the constructed agricultural machine operation efficiency model is calibrated and verified, the agricultural machine operation efficiency model is used to simulate different water field specifications (length and width), agricultural machine widths, and operation times, the agricultural operation efficiency of each combination is calculated, and a model simulation data set is formed. A prediction model about the water field irrigation parameters and the irrigation performance is fitted by using an ANN or other black box model.

[0029] In some embodiments, paddy field irrigation and agricultural machinery operation experiments need to be conducted on paddy fields of different sizes, ranging from 50m×60m, 80m×120m, and 120m×200m. These fields require a large area, making the experiments time-consuming and labor-intensive, thus limiting the scope of experiments. Other combinations require numerical simulation, necessitating the construction of corresponding models. While the simulation time for each set of conditions is significantly reduced compared to experiments after model construction, computational limitations mean that different paddy field sizes still require 1-4 hours of simulation. Therefore, 5-10 levels for each influencing factor are initially set, requiring approximately 150 simulation scenarios in total. To obtain more refined combinations of irrigation and agricultural machinery operation efficiencies, and to conduct faster simulations, the irrigation and agricultural machinery operation efficiencies of the 150 simulated scenarios are calculated, forming simulation datasets. Black-box models such as ANNs are then used to develop corresponding prediction models.

[0030] S2: Determine the actual constraints. If there are no actual constraints, construct a paddy field specification prediction model and proceed to S3; if there are actual constraints, construct a paddy field specification optimization model and proceed to S5.

[0031] The actual constraints are those related to the layout of the road network and canal system.

[0032] A paddy field specification prediction model is a mathematical model used to predict the specifications of paddy fields. For example, the expression for a paddy field specification prediction model can be: ; in, This represents the system performance prediction index value. This represents the system performance prediction index function.

[0033] In some embodiments, the processor can establish a relationship between the length and width of the paddy field using irrigation performance and agricultural machinery operation efficiency as common parameters. An objective function is constructed based on this inherent relationship, and constraints are set according to actual conditions to solve for suitable paddy field specifications with high irrigation performance and agricultural machinery operation efficiency.

[0034] Taking the actual situation in the Heping Irrigation District of Heilongjiang Province as an example. The constraints on the length and width of farm plots are the field road network and the intervals between irrigation canals. Rural farm roads generally connect with main roads and are laid out as main roads and branch roads; branch roads are generally perpendicular to main roads and located on the boundaries of the area; within the area, field roads and production roads are set up. For plots under 200 mu, production roads are generally set up, and for plots over 200 mu, both field roads and production roads are set up. Production roads are generally perpendicular to field roads, and field roads are located in the middle, with intervals generally less than 400 meters, arranged in a cross or grid pattern, connecting with main roads and branch roads on the outside and connecting with production roads on the inside; production roads are generally no more than 200 meters long, providing access for agricultural machinery to move in the field and for people and animals to walk in the field. Each plot should be equipped with a downhill access road, about 3 meters wide; in relatively flat and open areas, field roads should have passing lanes, with a roadbed width of not less than 5.5 meters and an effective length of not less than 10 meters. Rural road networks are largely consistent with irrigation canal systems, mostly laid out along irrigation ditches. The aspect ratio of paddy fields is constrained by the efficiency of rice transplanters. Under common planting spacing (9, 12, and 15 cm) and transplanter specifications (4 rows and 6 rows), the optimal aspect ratio range for transplanter efficiency is 1-3.2. For a specific paddy field area, the aspect ratio has a corresponding range. Based on the actual situation, the constraints of local paddy field length, width, irrigation flow, and aspect ratio are determined. The paddy field specification optimization design model can be expressed as the following equation. The paddy field specification optimization model is a mathematical model used to optimize paddy field specifications.

[0035] ; in, This represents the predicted system performance index value; n is the number of fields between the two field paths, which is an integer; m is the number of fields between the two production paths, which is an integer.

[0036] S3: Normalize the irrigation performance indicators in the irrigation performance prediction data sample and the agricultural machinery operation efficiency indicators in the agricultural machinery operation efficiency prediction data sample, respectively, and set weight coefficients. Then, use the paddy field specification prediction model to simulate and obtain the paddy field specification prediction data sample with paddy field system performance.

[0037] The paddy field specification prediction data sample is a sample containing paddy field specification prediction data and corresponding indicator values.

[0038] In some embodiments, the importance of irrigation performance and agricultural machinery operation efficiency in actual agricultural production often varies with different regional precipitation differences and agricultural machinery availability. Therefore, by setting weight coefficients for both, the data samples of the paddy field specification prediction model can be made more consistent with local conditions.

[0039] In some embodiments, the expression for the performance of the paddy field system can be: ; in, Indicates the performance of the paddy field system. The weights of the normalized irrigation performance indicators, This represents a normalized irrigation performance index. The weights representing the normalized agricultural machinery operation efficiency index. This represents a normalized indicator of agricultural machinery operation efficiency.

[0040] In some embodiments, since the numerical values ​​of the irrigation performance and agricultural machinery operation efficiency indicators differ significantly, in order to unify the two, when constructing the data sample for the paddy field specification prediction model, it is necessary to normalize the irrigation performance indicators in the irrigation performance prediction data sample and the agricultural machinery operation efficiency indicators in the agricultural machinery operation efficiency prediction data sample respectively.

[0041] Different regions have different priorities regarding irrigation performance and agricultural machinery operation efficiency. For example, in areas with abundant water resources or at the head of canals, the focus is on agricultural machinery operation efficiency.

[0042] In some embodiments, the expressions for the normalized irrigation performance index and the normalized agricultural machinery operation efficiency index can be: ; ; in, This represents a normalized irrigation performance index. This represents the maximum value of the irrigation performance index. This represents the irrigation performance index value under typical operating conditions. This represents the minimum value of the irrigation performance index. An indicator representing the normalized efficiency of agricultural machinery operations. This represents the efficiency index value of agricultural machinery operations under typical working conditions. This represents the minimum value of the agricultural machinery operation efficiency index. This represents the maximum value of the agricultural machinery operation efficiency index.

[0043] Normalizing the two different evaluation metrics separately helps in the subsequent construction and solution of the objective function.

[0044] S4: Select data exceeding the index threshold from the paddy field specification prediction data sample as non-dominated solutions, and combine the non-dominated solutions to obtain the suitable range of paddy field specifications.

[0045] A non-dominated solution is a sample in the paddy field specification prediction data where the indicator value exceeds the indicator threshold. For example, a non-dominated solution could be one of the top 20% of the indicator values ​​in the paddy field specification prediction data.

[0046] The appropriate range of paddy field specifications is the range of paddy fields with high irrigation performance and high efficiency of agricultural machinery operation.

[0047] In some embodiments, such as Figure 3 As shown, for functions that are difficult to derive analytical solutions, a non-dominated solution is first selected to obtain a better threshold range, and then the paddy field specifications that are easy to implement are selected in combination with the actual conditions (such as selecting paddy field length and width as integers).

[0048] S5: Based on irrigation performance prediction data samples and agricultural machinery operation efficiency prediction data samples, the paddy field specification optimization model is used to perform multi-objective optimization and global random search to obtain suitable paddy field specifications.

[0049] The appropriate paddy field specifications are those that offer high irrigation performance and efficient agricultural machinery operation.

[0050] In some embodiments, for specific design areas where the road network and canal system layout remain unchanged, a paddy field specification optimization model is established. Using the actual road network and canal system layout as constraints, relevant intelligent algorithms determine the appropriate paddy field specifications. Paddy field specification optimization design involves numerous field technical factors and is a multi-objective optimization problem.

[0051] In some embodiments, a multi-objective optimization problem with n decision variables and m objective variables can be expressed as: ; in, Represents an n-dimensional decision vector. Let n represent the n-dimensional decision space; Represents an m-dimensional target vector. Let m be the target space; where the objective function is... m mapping functions from the decision space to the target space are defined. q inequality constraints are defined. p equality constraints are defined.

[0052] In some embodiments, the processor can combine the multi-objective optimization problem with the constraints of field length, field width and irrigation flow in actual production to obtain a multi-objective optimization that determines the appropriate paddy field specifications.

[0053] In some embodiments, the expression for multi-objective optimization can be: ; in, This represents the function that takes the maximum value. Indicates the performance of the paddy field system. Indicates irrigation efficiency. Indicates the length of the field. Indicates the width of the field. Indicates irrigation flow rate, Indicates the efficiency of agricultural machinery operations. Indicates the width of agricultural machinery. express, Indicates the maximum field length. Indicates the maximum field width. This indicates the maximum irrigation flow rate.

[0054] Based on the conventional formulation of multi-objective optimization problems, a multi-objective optimization model is constructed by combining a paddy field specification prediction model that considers irrigation performance and agricultural machinery operation efficiency, and the constraint range of each condition is set according to actual conditions.

[0055] In some embodiments, such as Figure 4 As shown, the processor can solve the problem using a global random search algorithm. Let's assume a population of m particles exists in an n-dimensional search space. The position of the i-th particle is Its speed is Its individual extreme value is The global extremum of the population is The particle will update its velocity and position according to formulas (6) and (7).

[0056] In some embodiments, the expressions for the particle's velocity and position can be: ; ; in, This represents the velocity of the id-th particle in the (t+1)th iteration. Indicates the inertia factor. This represents the velocity of the id-th particle in the t-th iteration. Indicates the first learning factor. Denotes the first random constant. Represents the extreme value of an individual in a population. This represents the position of the id-th particle in the t-th iteration. Indicates the second learning factor. Denotes the second random constant. Represents the global extremum of the population. This represents the position of the gd-th particle in the t-th iteration.

[0057] In some embodiments, and It is a random number within [0,1]; speed Based on the structure of the multi-objective optimization model, a suitable solution algorithm is determined.

[0058] In some embodiments, the expressions for inertia weight and learning factor can be: ; ; in, Indicates inertia weight, Indicates the maximum weight. Indicates the minimum weight. Indicates the current iteration number. Indicates the maximum number of iterations. Let j represent the j-th learning factor. Let j represent the j-th maximum learning factor. Let j represent the j-th smallest learning factor.

[0059] To prevent overfitting of the calculation results, some parameters in the solution algorithm are optimized to improve the accuracy of the simulation results.

[0060] S6: By analyzing the suitable range and suitable paddy field specifications, the farmland scale design results are obtained, and the design of the scale of large-scale farmland is completed.

[0061] The farmland scale design result is the farmland scale design result that optimizes irrigation performance and agricultural machinery operation efficiency.

[0062] In some embodiments of this specification, a method for designing large-scale farmland based on dual-objective constraints is provided. This method combines irrigation performance prediction models and agricultural machinery operation efficiency prediction models to determine optimal design variables. For scenarios without constraints, a paddy field specification prediction model is constructed to propose a suitable range of paddy field specifications with high system performance. This result can be widely applied to rice-growing areas in cold regions. For specific regions, the existing local irrigation canal systems and road networks are used as constraints to construct an optimal paddy field specification model. Based on intelligent optimization algorithms, a paddy field specification with high system performance suitable for the local area is proposed. This approach allows farmland scale design to better reflect actual conditions.

Claims

1. A method for designing a large-scale farmland scale based on double-target constraints, characterized in that, The method comprises the following steps: S1: fitting the actual irrigation samples and the actual agricultural operation samples by using an artificial neural network model to obtain an irrigation performance prediction model and an agricultural operation efficiency prediction model, respectively; wherein the irrigation performance prediction model is used to obtain irrigation performance prediction data samples by simulation, and the agricultural operation efficiency prediction model is used to obtain agricultural operation efficiency prediction data samples by simulation; S2: judging the actual constraint conditions, if there is no actual constraint condition, constructing a paddy field specification prediction model and entering S3; if there is an actual constraint condition, constructing a paddy field specification optimization model and entering S5; S3: normalizing and setting weight coefficients for the irrigation performance indexes in the irrigation performance prediction data samples and the agricultural operation efficiency indexes in the agricultural operation efficiency prediction data samples, respectively, simulating by using the paddy field specification prediction model to obtain paddy field specification prediction data samples with paddy field system performance; S4: selecting data exceeding the index threshold in the paddy field specification prediction data samples as non-inferior solutions, combining the non-inferior solutions to obtain a suitable range of paddy field specifications; S5: based on the irrigation performance prediction data samples and the agricultural operation efficiency prediction data samples, performing multi-objective optimization and global random search by using the paddy field specification optimization model to obtain suitable paddy field specifications; S6: obtaining a paddy field scale design result by analyzing the suitable range of paddy field specifications and the suitable paddy field specifications, and completing the design of the large-scale paddy field scale.

2. The method of claim 1, wherein, The expression of the paddy field system performance is: ； wherein, represents a paddy field system performance, represents a weight of a normalized irrigation performance index, represents a normalized irrigation performance index, represents a weight of a normalized farm work efficiency index, represents a normalized farm work efficiency index.

3. The method of claim 2, wherein, The expressions of the normalized irrigation performance indexes and the normalized agricultural operation efficiency indexes are: ； ； wherein denotes a normalized irrigation performance indicator, denotes a maximum value of the irrigation performance indicator, denotes an irrigation performance indicator value at a typical working condition, denotes a minimum value of the irrigation performance indicator, denotes a normalized farm machinery operation efficiency indicator, denotes a farm machinery operation efficiency indicator value at a typical working condition, denotes a minimum value of the farm machinery operation efficiency indicator, denotes a maximum value of the farm machinery operation efficiency indicator.

4. The method of claim 1, wherein, The expression of the multi-objective optimization is: ； wherein, denotes a max function, denotes a paddy field system performance, denotes an irrigation efficiency, denotes a field length, denotes a field width, denotes an irrigation flow rate, denotes a farm machine operation efficiency, denotes a farm machine width, denotes a maximum field length, denotes a maximum field width, denotes a maximum irrigation flow rate.

5. The method of claim 1, wherein, The suitable paddy field specifications include the velocity and position of particles, wherein the expressions of the velocity and position of particles are: ； ； wherein, represents the velocity of the idth particle at the t+1th iteration, represents an inertia factor, represents the velocity of the idth particle at the tth iteration, represents a first learning factor, represents a first random constant, represents a population individual extremum, represents the position of the idth particle at the tth iteration, represents a second learning factor, represents a second random constant, represents a population global extremum, represents the position of the gdth particle at the tth iteration.

6. The method of claim 5, wherein, The expressions of the inertia weight and the learning factor are: ； ； wherein, represents an inertia weight, represents a maximum weight, represents a minimum weight, represents a current iteration number, represents a maximum iteration number, represents a jth learning factor, represents a jth maximum learning factor, represents a jth minimum learning factor.

Images