Field operation scheduling method based on fusion of simulated annealing algorithm and genetic algorithm

Abstract

The invention discloses a field operation scheduling method based on fusion of a simulated annealing algorithm and a genetic algorithm, and relates to the technical field of field management, and the method comprises the following steps: a) starting the algorithm, and setting simulated annealing algorithm control parameters; b) generating an initial solution of a simulated annealing algorithm; c) the counter count is equal to 0; d), k = 1; e) adopting three modes of solution transformation, variation and reversion to obtain three new solutions, and taking the minimum operation sequence path length of the three new solutions as a new solution S2; f) judging whether to accept the new solution or not according to the Metropolis criterion; g) obtaining a new solution S1, and enabling k to be equal to k + 1; h) judging kgt; l; i) executing count = count + 1 and T = qT; cooling at the cooling rate q; j) judging whether T is smaller than Tend, if so, executing k), and otherwise, executing d); k) outputting an operation sequence path solution set; l) setting the population size, crossover probability and mutation probability of the genetic algorithm; m) initializing a population Chrom2 by using a genetic algorithm, and adopting a binary coding mode; n) calculating a fitness value of a second layer algorithm; o) carrying out selection, crossover and mutation operation on the population Chrom2; then, after reinsertion, an updated population Chrom2 is obtained; p) detecting the number of iterations; and q) outputting the shortest scheduling path between the global regions.

Description

Technical Field

[0001] The invention relates to the technical field of field management, and in particular to a scheduling method based on the fusion of a simulated annealing algorithm and a genetic algorithm. Background Art

[0002] my country's farmland is small and scattered. Plant protection operations require the gradual completion of pesticide application work on each piece of farmland, and finally return to the aircraft take-off and landing point. Different operation sequences and entry and exit point selections for the area will lead to significant differences in the aircraft's non-pest application range. Reducing the aircraft's non-pest application range not only means that it can effectively reduce aviation fuel consumption and significantly reduce operation time, but also significantly improve the efficiency of aircraft plant protection operations. However, the planning research on the dispatching path of aerial pesticide application operations of plant protection aircraft is relatively scarce, and it is difficult to meet the actual use needs of aerial pesticide application on multiple farmlands. Summary of the invention

[0003] The object of the present invention is to provide a field operation scheduling method based on the fusion of simulated annealing algorithm and genetic algorithm to solve at least any one of the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solution: a field operation scheduling method based on the fusion of simulated annealing algorithm and genetic algorithm, comprising the following steps:

[0005] a) Initialize the simulated annealing algorithm parameters, including cooling rate q, initial temperature T0, end temperature Tend, and chain length L;

[0006] b) generating an initial job path sequence S1 as the initial solution of the simulated annealing algorithm;

[0007] c) Set a counter count and initialize it to 0;

[0008] d) Set loop variable k to 1;

[0009] e) generating three new solutions through solution transformation, mutation, and reversal operations, and selecting the solution with the shortest operation path length from the three new solutions as the new solution S2;

[0010] f) Determine whether to accept the new solution according to the Metropolis criterion; the operation sequence path is calculated using the distance neighborhood table between the areas, and the midpoint of the line connecting the entry and exit points of each area is used as the calculation reference point. The current operation sequence path length is f(S1), the new solution operation sequence path length is f(S2), and the path difference is df = f(S2)-f(S1), then the Metropolis criterion is

[0011]

[0012] If df<0, the new job sequence path is accepted with probability 1; otherwise, the new path is accepted with probability exp(1-df / T);

[0013] g) Update solution S1 to new solution S2, and increase k by 1;

[0014] h) Determine whether k is greater than L, if so, proceed to step i), otherwise return to step e);

[0015] i) executing count=count+1, T=qT; cooling is performed using a cooling rate q;

[0016] j) Determine whether T is less than T end , if yes, execute step k, otherwise execute step d;

[0017] k) Output the job sequence path solution set;

[0018] l) Set the second-layer genetic algorithm population size, crossover probability, mutation probability, the second-layer algorithm maximum number of iterations GENMAX, and set the initial value of the number of iterations to 1, that is, gen = 1;

[0019] m) The second-level genetic algorithm initializes the population Chrom2 using binary coding;

[0020] n) Calculate the fitness value of the second-layer algorithm, taking four-piece areas as an example, and assuming the operation sequence is ABCD;

[0021] The operation starting points of area B can be B1 and B2 or B2 and B1, and the two states are represented by 0 and 1 respectively. Then the state code of A1, A2, B1, B2, C1, C2, D1 and D2 is 0000;

[0022] It indicates that the operation sequence is starting from A1, entering area A, exiting from A2, entering B1, exiting B2, entering C1, exiting C2, entering D1, exiting D2, and then returning to A1. The state codes of A1, A2, B2, B1, C1, C2, D2, and D1 are 0101. The processing of aircraft take-off and landing points is the key point. The aircraft take-off and landing points are regarded as two points, but the coordinates are set to the same value to facilitate the rapid calculation of fitness values ​​using the neighborhood table.

[0023] This can represent the global inter-regional scheduling path starting from the aircraft take-off and landing point, taking into account the entry and exit points of each area; the fitness f2 is designed to be the inverse of the shortest distance of all global inter-regional scheduling paths obtained by the first-level algorithm under each entry and exit state;

[0024] Taking state code 0101 as an example, assuming that the first layer of the algorithm obtains n types of global inter-region scheduling paths, the fitness calculation formula is:

[0025] o) Perform selection, crossover, and mutation operations on population Chrom2; then reinsert to obtain updated population Chrom2;

[0026] P) detecting whether the number of iterations exceeds the maximum number of iterations GENMAX, and when the number of iterations does not exceed the maximum number of iterations, returning to step n, otherwise proceeding to step q;

[0027] q) Output the shortest global inter-regional scheduling path, which is the optimal scheduling solution for multi-farmland path scheduling planning.

[0028] Preferably, step ak represents the first layer of the fusion algorithm, which uses a simulated annealing algorithm to solve the job sequence path. The first layer of the algorithm uses a simulated annealing algorithm to solve the job sequence path. The algorithm can solve different nonlinear problems and has strong robustness, global convergence, implicit parallelism and wide applicability. In step e, mutation and reversal operations in the genetic algorithm are added to generate new solutions. Three new solutions are obtained by solution transformation, mutation and reversal. The shortest job path sequence calculated by the three new solutions is used as the current new solution S 2 , and then use the Metropolis criterion to make a judgment. This method combines the advantages of the genetic algorithm's strong global search ability and retains the strong global convergence of the simulated annealing algorithm, making it very competitive in solving the TSP problem.

[0029] Preferably, step lq represents the second layer of the fusion algorithm, which is a practical genetic algorithm used to further consider the entry point mechanism based on the job sequence path; in the step k, relying on the solution result of the first layer simulated annealing algorithm, the fitness value calculation of the second layer genetic algorithm adopts an expansion operation for each chromosome in Chrom2, and all job path sequences obtained by the first layer algorithm, and uses the extended domain table to calculate the length of the global inter-regional scheduling distance, which does not include the distance within the area, and the inverse of the shortest distance obtained is used as the fitness value of the chromosome.

[0030] Preferably, the population size of the genetic algorithm population in step i is set to 4 to 6 times the sum of the number of farmland and the number of aircraft take-off and landing points, the generation gap is set to a constant, and the generation gap is set to 0.9 to 0.95; the crossover probability Pc and the mutation probability Pm are both designed to be dynamically changing probabilities. The crossover probability decreases as the number of iterations increases, and the mutation probability decreases as the number of iterations increases. Its calculation formula is: Pm: Among them, Pcmin=0.6, Pcmax=0.9, Pm0=0.1.

[0031] The beneficial effects of the present invention are as follows:

[0032] In the present invention, the present invention innovatively adopts an algorithm that integrates simulated annealing algorithm and genetic algorithm to carefully plan the multi-farm path scheduling path, especially focusing on the aerial pesticide application field of manned large-scale plant protection aircraft;

[0033] At the first level, the simulated annealing algorithm is used to solve a relatively good job sequence path. This algorithm innovatively uses three methods to obtain new solutions: two-neighborhood solution changes, self-mutation, and self-reversal. It successfully integrates the powerful ability of genetic algorithms in global search and significantly improves the efficiency of global search compared to traditional simulated annealing algorithms.

[0034] The second-layer algorithm cleverly uses genetic algorithms, closely combines the solution set generated by the first layer, and fully considers the entry and exit points of each area, thereby successfully obtaining the shortest global inter-regional scheduling path. This design cleverly avoids the dilemma of the local optimal solution of the global inter-regional scheduling path that may be encountered when only the shortest operation sequence path is obtained in the first-layer program, showing high computational efficiency and greatly improving the path planning efficiency of multi-field pesticide application;

[0035] In the genetic algorithm, the crossover probability and mutation probability are innovatively adopted. It is worth noting that both the crossover probability and the mutation probability will gradually decrease with the increase of the number of iterations. This unique design not only takes into account the global search capability, but also significantly improves the convergence speed of the algorithm, greatly saves the search time, and effectively improves the overall efficiency of the algorithm;

[0036] The present invention brings many significant advantages: it can significantly reduce the number of turns of the plant protection aircraft in the single-area farmland, effectively reducing the difficulty of the pilot's driving operation; it can greatly shorten the global regional dispatching path distance in the multi-farmland spraying operation, greatly saving the working time of multi-farmland aerial spraying. In addition, it can also effectively reduce the use of aviation fuel, significantly save the economic cost of farmland spraying operations, bring great convenience to multi-farmland spraying operations, and comprehensively improve the efficiency of farmland aerial spraying. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flow chart of the fusion algorithm of simulated annealing algorithm and genetic algorithm in the present invention;

[0038] Figure 2 Graph for multi-farm path planning tasks;

[0039] Figure 3 This is a schematic diagram of path planning within a single area;

[0040] Figure 4 Schematic diagram of the global path scheduling planning results for multiple farmlands. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0042] The present invention provides Figure 1-4 The field operation scheduling method shown is based on the fusion of simulated annealing algorithm and genetic algorithm, including the path planning task as follows:

[0043] Helicopter from Figure 2 The aircraft starts from the take-off and landing point H, then passes through the entry and exit points of each area, traverses each area, and then returns to H. The path for scheduling between areas is required to be the shortest. Figure 3 The full coverage path planning algorithm shown in the figure is to draw a parallel line from the longest side of the boundary according to the width of the aircraft jet. When it reaches the intersection of other boundaries, the aircraft turns around and flies until all areas of the area are traversed. Therefore, each area must have an entry point and an exit point. When scheduling, it must enter and exit from two points, but it is not specific which point. For example, from Figure 2 After entering from A1, you must exit from A2, or after entering from A2, you must exit from A1. Figure 3 The path planning shown in Figure 1 The simulated annealing algorithm and genetic algorithm shown in the figure are combined to perform route planning, and the following is obtained: Figure 4 The better results shown.

[0044] The simulated annealing algorithm and genetic algorithm fusion algorithm process is as follows Figure 1 As shown:

[0045] The following is the algorithm implementation process:

[0046] Step 1: The algorithm starts, setting the cooling rate q and initial temperature T of the simulated annealing algorithm 0 , End temperature T end and the chain length L;

[0047] Step 2: Generate an initial operation path sequence S by using a random arrangement method based on the number of farmlands to be sprayed 1 , and use this order as an initial solution for the simulated annealing algorithm; for example, if there are 5 farmlands that need to be sprayed, then S 1 The length of is set to 5+1=6. 1 represents the take-off and landing point of the aircraft and is placed at the last digit of the population length. Then 123456, 213465, and 345216 can all be new solutions, provided that the numbers are not repeated.

[0048] Step 3, counter count=0;

[0049] Step 4: Set k=1;

[0050] Step 5: Use solution transformation, mutation, and reversal to obtain three new solutions. The one with the smallest operation sequence path length among the three new solutions is taken as the new solution S. 2 Solution transformation method, if the number of farmland is 5, then a random integer r is generated in

[16] 1 and r 2 , assuming r 1 =2, r 2 =4, then 123456 is transformed into 143256. Mutation method, if the number of farmland is 5, then a random integer r is generated in

[16] 3 , assuming that r 3 =3, then the 3 position will be mutated to a random integer between

[16] , and the repeated farmland number will be set to r 3 The original number of the position. For example, 123456, in r 3 = 3 mutates to 6, then 123456 is transformed into 126453. Reversal method, assuming that the number of farmland is 5, then a random integer r is generated in

[16] 1 and r 2 , assuming r 1 =2, r 2 =4, then 123456 is transformed into 124356. The shortest operation sequence path length of the three solutions generated by the three methods is used as the new solution S 2 .

[0051] Step 6: Determine whether to accept the new solution according to the Metropolis criterion; the operation sequence path is calculated using the distance neighborhood table between the areas, and the midpoint of the line connecting the entry and exit points of each area is used as the calculation reference point. The length of the current operation sequence path is recorded as f(S 1 ), the length of the new solution's job sequence path is f(S 2 ), the path difference is df=f(S 2 )-f(S 1 ), then the Metropolis criterion is

[0052]

[0053] If d f <0, the new job sequence path is accepted with probability 1; otherwise, the new path is accepted with probability exp(1-df / T).

[0054] Calculation method of operation sequence path length: Figure 2Taking the scheduling task in as an example, the coordinates of its 18 areas and aircraft take-off and landing points are shown in Table 1. The neighborhood table generated by the midpoints of its entry and exit points is shown in Table 2. According to Table 2, the path length of the solution in the simulated annealing algorithm of this layer can be quickly calculated.

[0055] Table 1 Coordinates of farmland Unit: m

[0056]

[0057]

[0058] The midpoint coordinates of the entry and exit points of each area are used as reference points to calculate the distance of each farmland operation sequence path. The distance neighborhood table is generated using the midpoint coordinates of the 18 areas and the coordinates of the aircraft take-off and landing points.

[0059] Table 2 Distance neighborhood table generated by the midpoints of each area's entry and exit points

[0060]

[0061]

[0062]

[0063] Step 7: Get the new solution S 1 , and set k=k+1;

[0064] Step 8: Determine whether k>L. If yes, proceed to S9, otherwise return to S5;

[0065] Step 9, execute count=count+1, T=qT; cool down using cooling rate q.

[0066] Step 10: Determine whether T is less than T end , if yes, execute S11, otherwise execute S4;

[0067] Step 11, output the job sequence path solution set;

[0068] Step 12, set the genetic algorithm population size, crossover probability, and mutation probability. If the population size is set to 4 to 6 times the sum of the number of farmland and aircraft take-off and landing points, such as 18 plots, then set the population size to 100; the generation gap is set to a constant; the generation gap is set to 0.9 to 0.95, such as 0.9. The crossover probability Pc and the mutation probability Pm are both designed to be dynamically changing probabilities. The crossover probability decreases with the increase of the number of iterations, and the mutation probability decreases with the increase of the number of iterations. Its calculation formula is: Pm: Where P cmin =0.6, P cmax =0.9, initial mutation probability P m0= 0.1. The maximum number of iterations of the genetic algorithm is GENMAX, such as 2000. And the initial value of the number of iterations is set to 1, that is, gen = 1;

[0069] Step 13, the second-layer genetic algorithm initializes the population Chrom2, using binary coding. The coding method uses binary coding. Taking four areas as an example, the operation sequence is assumed to be ABCD, where A represents the aircraft take-off and landing point. The operation starting points of area B can be B1 and B2 or B2 and B1, and the two states are represented by 0 and 1 respectively. Then the state code of A1A2B1B2C1C2D1D2 is 0000. In particular, the coordinates of the aircraft take-off and landing points A1 and A2 are exactly the same, indicating that the operation sequence is starting from A, B1 in, B2 out, C1 in, C2 out, D1 in, D2 out, and then back to A. Then the state code of A1A2B2B1C1C2D2D1 is 0101. In this way, the operation scheduling path between each area considering the entry and exit points of each area starting from the aircraft take-off and landing point can be represented.

[0070] Step 14, calculate the fitness value of the second-layer algorithm. To facilitate the calculation of the fitness of the second-layer genetic algorithm, the neighborhood table is also used. The neighborhood table is generated by the horizontal and vertical coordinates of the entry and exit points of each area in Table 1, as shown in Table 3. Since the neighborhood table considering the entry and exit points of 18 areas is too large, the domain distance matrix of 4 areas and 1 aircraft take-off and landing point is displayed here. The four areas are A, B, C, and D, and the entry and exit points of each area are represented by two codes, such as the entry and exit points of area A are represented as A. 1 and A 2 The coordinates of the aircraft take-off and landing points are represented by H. The domain distance matrix of the 18-area area considering the entry and exit points and the aircraft take-off and landing points can refer to this method. Fitness f 2 It is designed as the inverse of the shortest distance of all global inter-regional scheduling paths obtained by the first-layer algorithm under each entry and exit state. Taking state code 10010 as an example, the optimal path solution set obtained by the first-layer algorithm is corrected according to the 10010 state. The correction method is shown in Table 4. Assuming that there are n types of global inter-regional scheduling paths obtained by the first layer of the algorithm, the fitness calculation formula is formula (8).

[0071]

[0072] Table 3. Domain distance matrix considering entry and exit points and aircraft take-off and landing points of the area

[0073]

[0074]

[0075] Table 4 An example of modifying the first-layer algorithm job sequence path based on the state table Chrom2

[0076]

[0077]

[0078] It should be noted that 7 in Table 4 corresponds to point A2 in Table 3, and 1 corresponds to point A1 in the table, and so on. For details, please refer to step 16.

[0079] Step 15, perform selection, crossover, and mutation operations on population Chrom2; then reinsert to obtain updated population Chrom2;

[0080] Step 16, check whether the number of iterations exceeds the maximum number of iterations GENMAX, when the number of iterations does not exceed the maximum number of iterations, return to step 11, otherwise proceed to step 14;

[0081] Step 17: output the shortest global inter-regional scheduling path, which is the optimal scheduling solution for multi-farmland path scheduling planning.

[0082] The present invention plans the multi-farmland path scheduling path by adopting the simulated annealing algorithm and the genetic algorithm fusion algorithm, especially for the aerial spraying of manned large-scale plant protection aircraft. Considering the safety and the non-spraying flight distance between the farmlands, the plant protection spray operation endurance range, drug loading, and plant protection spray operation endurance time parameters are set to be smaller than the aircraft's own performance parameters.

[0083] The simulated annealing algorithm and the genetic algorithm fusion algorithm of the present invention adopt a serial fusion method, and the number of iterations is the sum of the number of iterations of the two algorithms, which is faster than the nested fusion. Both layers of algorithms converge. The first layer adopts the simulated annealing algorithm, and adds the mutation operation and reversal operation in the genetic algorithm to generate new solutions. The three new solutions obtained by the three methods of solution transformation, mutation and reversal are used. The shortest operation path sequence calculated by the three new solutions is used as the current new solution S2, and then judged by the Metropolis criterion. This method combines the advantages of the strong global search ability of the genetic algorithm and retains the strong global convergence of the simulated annealing algorithm. It has strong competitiveness in solving the TSP problem and can quickly obtain a better operation sequence path solution set; the second layer algorithm adopts the genetic algorithm, combined with the operation sequence path solution set obtained by the first layer, considering the entry and exit points of each area, so as to obtain the shortest global inter-regional scheduling path, which avoids the local optimal solution of the global inter-regional scheduling path that only obtains the shortest operation sequence path in the first layer program, and has high calculation efficiency, thereby improving the path planning efficiency of multi-field pesticide application.

[0084] The present invention can shorten the number of turns for spraying pesticides by plant protection aircraft in a single-area farmland, shorten the inter-regional dispatching path range for spraying pesticides in multiple farmlands, save the working time of spraying pesticides in multiple farmlands, and improve the spraying efficiency of pesticides in farmlands. It can also effectively reduce the use of aviation fuel, save the economic cost of spraying pesticides in farmlands, and bring convenience to spraying pesticides in multiple farmlands.

[0085] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A field operation scheduling method based on the fusion of simulated annealing algorithm and genetic algorithm, characterized in that: The following steps are involved: a) Initialize the simulated annealing algorithm parameters, including cooling rate q, initial temperature T0, end temperature Tend, and chain length L; b) generating an initial job path sequence S1 as the initial solution of the simulated annealing algorithm; c) Set a counter count and initialize it to 0; d) Set loop variable k to 1; e) generating three new solutions through solution transformation, mutation, and reversal operations, and selecting the solution with the shortest operation path length from the three new solutions as the new solution S2; f) Determine whether to accept the new solution according to the Metropolis criterion; the operation sequence path is calculated using the distance neighborhood table between the areas, and the midpoint of the line connecting the entry and exit points of each area is used as the calculation reference point. The current operation sequence path length is f(S1), the new solution operation sequence path length is f(S2), and the path difference is df = f(S2)-f(S1), then the Metropolis criterion is If df<0, the new job sequence path is accepted with probability 1; otherwise, the new path is accepted with probability exp(1-df / T); g) Update solution S1 to new solution S2, and increase k by 1; h) Determine whether k is greater than L, if so, proceed to step i), otherwise return to step e); i) Execute count=count+1, T=qT; Cooling is performed using a cooling rate q; j) Determine whether T is less than T end , if yes, execute step k, otherwise execute step d; k) Output the job sequence path solution set; l) Set the second-layer genetic algorithm population size, crossover probability, mutation probability, the second-layer algorithm maximum number of iterations GENMAX, and set the initial value of the number of iterations to 1, that is, gen = 1; m) The second-level genetic algorithm initializes the population Chrom2 using binary coding; n) Calculate the fitness value of the second-layer algorithm, taking four-piece areas as an example, and assuming the operation sequence is ABCD; The operation starting points of area B can be B1 and B2 or B2 and B1, and the two states are represented by 0 and 1 respectively. Then the state code of A1, A2, B1, B2, C1, C2, D1 and D2 is 0000; It indicates that the operation sequence is starting from A1, entering area A, exiting from A2, entering B1, exiting B2, entering C1, exiting C2, entering D1, exiting D2, and then returning to A1. The state codes of A1, A2, B2, B1, C1, C2, D2, and D1 are 0101. The processing of aircraft take-off and landing points is the key point. The aircraft take-off and landing points are regarded as two points, but the coordinates are set to the same value to facilitate the rapid calculation of fitness values ​​using the neighborhood table. This can represent the global inter-regional scheduling path starting from the aircraft take-off and landing point, taking into account the entry and exit points of each area; the fitness f2 is designed to be the inverse of the shortest distance of all inter-regional scheduling paths obtained by the first-level algorithm under each entry and exit state; Taking state code 0101 as an example, assuming that the first layer of the algorithm obtains n types of global inter-region scheduling paths, the fitness calculation formula is: o) Perform selection, crossover, and mutation operations on population Chrom2; then reinsert to obtain updated population Chrom2; P) detecting whether the number of iterations exceeds the maximum number of iterations GENMAX, and when the number of iterations does not exceed the maximum number of iterations, returning to step n, otherwise proceeding to step q; q) Output the shortest global inter-regional scheduling path, which is the optimal scheduling solution for field path scheduling planning.

2. The field operation scheduling method based on the fusion of simulated annealing algorithm and genetic algorithm according to claim 1 is characterized in that: The first layer of the algorithm uses simulated annealing algorithm to solve the job sequence path, which can solve different nonlinear problems; In step e, mutation and reversal operations are added to the genetic algorithm to generate new solutions. The three new solutions obtained by solution transformation, mutation, and reversal are used. The shortest operation path sequence calculated by the three new solutions is used as the current new solution S2, and then the Metropolis criterion is used for judgment.

3. The field operation scheduling method based on the fusion of simulated annealing algorithm and genetic algorithm according to claim 2 is characterized in that: The second layer uses a practical genetic algorithm to consider the entry point mechanism; relying on the solution results of the first layer simulated annealing algorithm, the second layer genetic algorithm fitness value calculation adopts an expansion operation on all job path sequences obtained by the first layer algorithm for each chromosome in Chrom2, and uses the extended domain table to calculate the length of the global inter-regional scheduling distance, which does not include the distance within the area, and the inverse of the shortest distance obtained is used as the fitness value of the chromosome.

4. The field operation scheduling method based on the fusion of simulated annealing algorithm and genetic algorithm according to claim 3 is characterized in that: The population size of the genetic algorithm population in step a is set to 4 to 6 times the sum of the number of farmland and the number of aircraft take-off and landing points, the generation gap is set to a constant, and the generation gap is set to 0.9 to 0.95; The crossover probability Pc and mutation probability Pm are designed to be dynamically changing probabilities. The crossover probability decreases as the number of iterations increases, and the mutation probability decreases as the number of iterations increases. The calculation formula is: Where P cmin =0.6, P cmax =0.9, P m0 =0.1.

Images