Coal machine equipment production plan optimization method based on variable neighborhood search algorithm

Abstract

The invention provides a coal machine equipment production plan optimization method based on a variable neighborhood search algorithm, and relates to the technical field of engineering scheduling. The method comprises the following steps: 1, initializing algorithm parameters; 2, constructing a priority ranking set and a total cost function; 3, setting and optimizing an initial solution; 4, generating a neighborhood solution set and updating the neighborhood solution set to obtain a candidate solution set; 5, selecting a neighborhood structure; 6, searching a neighborhood to obtain a local optimal solution; 7, updating the optimal solution; 8, optimizing the optimal solution; and 9, outputting a globally optimal solution after the execution of the number of iterations is completed. According to the method, the initial solution and the neighborhood structure of the variable neighborhood search algorithm are improved, the quality of the initial solution is optimized, premature falling into local optimum is avoided, the approximate optimal solution can be obtained for the coal machine equipment production plan optimization problem, more effective decision support is provided for decision makers, and therefore the efficiency and benefits of equipment batch production are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering scheduling, and particularly relates to a coal machine equipment production plan optimization method based on a variable neighborhood search algorithm. BACKGROUND

[0002] With the rapid development of intelligent manufacturing and industrial internet technology, coal machine complete equipment manufacturing industry is facing unprecedented opportunities and challenges. Batch production and operation and maintenance of coal machine complete equipment play a crucial role, efficient batch production can significantly reduce production cost, shorten production cycle, improve product quality and market competitiveness.

[0003] Traditional coal machine equipment production plan optimization is often based on fixed parameters and relies heavily on manual experience. The accuracy of such manually prepared plans cannot be guaranteed, and due to real-time changes in multi-source data such as order demand, after-sales demand, traditional production plan preparation cannot be updated in time.

[0004] Therefore, it is necessary to design a production plan optimization method that can accurately prepare and optimize production plans according to actual market demand and production conditions. SUMMARY

[0005] (I) Technical problems to be solved In view of the shortcomings of the prior art, the present application provides a coal machine equipment production plan optimization method based on a variable neighborhood search algorithm, which solves the problem that traditional production plans cannot accurately respond to actual market demand and production conditions.

[0006] (II) Technical solutions To achieve the above purpose, the present application is realized by the following technical solutions: In the first aspect, the present application provides a coal machine equipment production plan optimization method based on an improved variable neighborhood search algorithm, comprising: S1, initializing each parameter in the algorithm, including production parameters and execution parameters; the production parameters include complete equipment and spare parts production information, customer information, delay penalty cost, production capacity constraint, order demand, production cost, production efficiency and delivery time; the execution parameters include maximum iteration number , current iteration number , crossover probability , local search algorithm population size ; S2, prioritizing complete equipment and spare parts according to the difference between processing time and expected delivery time; constructing a total cost function with the goal of minimizing production cost and minimizing delivery period penalty cost; S3, generating an initial solution according to the result of the priority sorting, optimizing the initial solution for the purpose of minimizing the total cost function, and setting the optimized initial solution as a global optimal solution; S4, generating a field solution set and updating according to the global optimal solution, to obtain a candidate solution set; S5, selecting a local search neighborhood structure based on a roulette probability; S6, obtaining a neighborhood and searching according to the candidate solution set and the local search neighborhood structure, to obtain a local optimal solution; S7, judging the local optimal solution and the candidate solution set according to the total cost function, if the local optimal solution is better, updating the local optimal solution as the global optimal solution, and increasing the weight of the neighborhood structure for obtaining the local optimal solution; if the candidate solution set is better, reducing the weight of the neighborhood structure for obtaining the local optimal solution, and returning to step S5; S8, based on a greedy algorithm, destructing and reconstructing the global optimal solution to obtain a new solution, and comparing the new solution with the global optimal solution, if the global optimal solution is better, judging whether to accept the new solution based on an improved Boltzmann function, and returning to step S4; otherwise, setting the new solution as the global optimal solution; S9, setting to , judging whether is true, if true, returning to step S4; otherwise, the algorithm ends, and the global optimal solution is output as the optimal coal machine equipment production plan scheme.

[0007] Preferably, the roulette probability selection includes: presetting the initial weights of each local search neighborhood structure to be equal; calculating the cumulative probability of each local search neighborhood structure according to the weight of each local search neighborhood structure and the roulette probability; selecting a neighborhood structure for local search based on the cumulative probability.

[0008] Preferably, the new solution obtained by destructing and reconstructing the global optimal solution based on the greedy algorithm includes: selecting a number of months in all months corresponding to the production plan of the global optimal solution based on epsilon-greedy rules; randomly extracting a complete set of equipment or spare parts in the months, and setting the extracted product set as ; ; inserting the products in the set one by one into the all possible positions except the initial position, and calculates a total production cost based on the total cost function; sets the insertion position with the minimum total production cost as a new solution.

[0009] Preferably, the local search neighborhood structure comprises ~ , represents the index number of the equipment, represents the index number of the spare parts; the defines a variable x, randomly generates an integer in the range of [1, P] to assign to the variable x, processes all the codes to the left of the position x in the reverse order, and then randomly selects the resource selection of all positions with a probability of ; the defines a variable x, randomly generates an integer in the range of [1, Q] to assign to the variable x, processes all the codes to the right of the position x in the reverse order, and then randomly selects the resource selection of all positions with a probability of ; the defines variables x, y, respectively generates two integers in the range of [1, P] and [P+1, P+Q] to assign to the variables x, y, wherein the xth position in the neighborhood solution set corresponds to the allocation month code of the complete equipment, and the yth position corresponds to the allocation month code of the spare parts, and exchanges the month numbers corresponding to the xth and yth positions in the neighborhood solution set; the defines variables x, y, P+Q=n, respectively generates two integers in the range of [1, P] and [P+1, P+Q] to assign to the variables x, y, wherein the xth position in the neighborhood solution set corresponds to the allocation month code of the complete equipment, and the yth position corresponds to the allocation month code of the spare parts, processes all the codes to the left of the position x and all the codes to the right of the position y in the reverse order, and then randomly selects the resource selection of all positions with a probability of (x+y) / 2n; the defines variables x, y, P+Q=n, randomly generates two integers in the range of [1, n] without replacement to assign to the variables x, y, wherein x≤y, exchanges the codes located on the left and right of the positions x and y in the neighborhood solution set, and then randomly selects the resource selection of all positions with a probability of (x+y) / 2n; the To define variables x, y, z, P+Q=n, randomly obtain three integers in the interval [1, n] without replacement, assign them to variables x, y, z, wherein x<y<z, exchange the codes located on both sides of x, z positions in the neighborhood solution set, reverse the codes inside x, y and y, z positions, and then randomly select the resource selection of all positions with a probability of (x+y+z) / 3n; The To define variables x, y, z, P+Q=n, randomly obtain three integers in the interval [1, n] without replacement, assign them to variables x, y, z, wherein x<y<z, exchange the codes located on both sides of x, z positions in the neighborhood solution set, reverse the codes inside x, y and y, z positions, and then randomly select the resource selection of all positions with a probability of (x+y+z) / 3n; The To define variables x, y, z, P+Q=n, randomly obtain three integers in the interval [1, n] without replacement, assign them to variables x, y, z, wherein x<y<z, exchange the codes located on both sides of x, z positions in the neighborhood solution set, reverse the codes inside x, y and y, z positions, and then randomly select the resource selection of all positions with a probability of (x+y+z) / 3n; The To define variables , , randomly obtain four integers in the interval [1, n] without replacement, assign them to variables , wherein , reverse the codes inside the positions of and in the neighborhood solution set, exchange the codes inside the positions between and , and then randomly select the resource selection of all positions with a probability of .

[0010] Preferably, the obtained candidate solution set specifically comprises: S41, based on the initial solution, generate a neighborhood solution set, and the solution set is denoted as ; wherein represents that a total of individuals in the neighborhood solution are considered, represents the individual in the neighborhood solution, which is obtained by randomly exchanging times from the initial solution ; S42, define variables , which have the same meaning as , and the individual in the set is assigned a value ; S43, defining a variable and , let the variable , let the variable , let the variable be assigned a random number generated in the interval ; wherein and have the same dimension as the individual ; S44, updating using the formula , wherein represents the th element in the global optimal solution, , are random numbers generated in the interval ; S45, generating a random number in the interval and determining whether is true, if true, assigning to ; otherwise, assigning to ; S46, let , determining whether is true, if true, returning to step S44, otherwise executing step S47; S47, calculating the fitness values and of the individual and the intermediate respectively according to the total cost function, and comparing and , if , assigning to the individual ; S48, let , determining whether is true, if true, returning to step S43; otherwise executing step S49; S49, screening out different individuals with better quality in the neighborhood solution set according to the total cost function, and assigning them to the candidate solution set.

[0011] Preferably, the total cost function is: In the formula, represents the total cost; represents the minimization of production cost; This indicates minimizing the cost of delivery time penalties; Indicates a complete set of equipment index; Indicates spare parts index; Indicates the customer index; Indicates the month index; Indicates complete set of equipment Production time; Indicates complete set of equipment Production completion time; Indicates spare parts Production time; Indicates spare parts Production completion time; Indicates manufacturer Monthly production complete set of equipment The cost; Indicates manufacturer Monthly production spare parts The cost; Indicates customer The complete set of equipment ordered; Indicates customer The collection of ordered spare parts; Indicates customer Ordered complete sets of equipment Quantity; Indicates customer Ordered spare parts Quantity; Indicates customer Ordered complete sets of equipment The unit delay penalty cost; Indicates customer Ordered complete sets of equipment Expected delivery time; Indicates customer Ordered spare parts The unit delay penalty cost; Indicates customer Ordered spare parts Expected delivery time; Indicates if complete set of equipment Allocated to months If production is initiated, the value is 1; otherwise, it is 0. Indicates if spare parts Allocated to months If production is initiated, the value is 1; otherwise, it is 0. Indicates if the customer Ordered complete set of equipment If the value is 1, then it equals 1; otherwise, it equals 0. If the customer spare parts are ordered , then equal to 1, otherwise equal to 0.

[0012] Preferably, the optimization of the initial solution with the minimum total cost function comprises: In the sorted production sequence, a new production task is inserted into the production sequence with the minimum total cost function.

[0013] In a second aspect, the present application also provides a coal machine equipment production planning optimization system based on an improved variable neighborhood search algorithm, comprising: A calculation module is configured to perform the following steps: S1, initializing various parameters in the algorithm, including production parameters and execution parameters; the production parameters include complete equipment and spare parts production information, customer information, delay penalty cost, production capacity constraint, order demand, production cost, production efficiency and delivery time; the execution parameters include the maximum number of iterations , the current number of iterations , the crossover probability , the population size of the local search algorithm ; S2, prioritizing the complete equipment and spare parts according to the difference between the processing time and the expected delivery time; constructing a total cost function with the minimum production cost and the minimum delivery period penalty cost as the target; S3, generating an initial solution according to the priority, optimizing the initial solution with the minimum total cost function, and setting the optimized initial solution as the optimal solution; S4, generating a neighborhood solution set according to the initial solution and updating to obtain a candidate solution set; S5, selecting and obtaining a local search neighborhood structure based on the roulette probability; S6, obtaining a neighborhood and searching based on the candidate solution set and the local search neighborhood structure to obtain a local optimal solution; S7, judging the advantages and disadvantages of the local optimal solution and the candidate solution set according to the total cost function, if the local optimal solution is better, updating the local optimal solution as the global optimal solution, and increasing the weight of the neighborhood structure from which the local optimal solution is obtained; if the candidate solution set is better, reducing the weight of the neighborhood structure of the local optimal solution, and returning to step S5; S8, based on the greedy algorithm, performing greedy destruction and greedy reconstruction on the global optimal solution to obtain a new solution, and comparing the new solution with the global optimal solution, if the global optimal solution is better, judging whether to accept the new solution based on the improved Boltzmann function, and returning to step S4; otherwise, assigning the new solution as the global optimal solution; S9, assigning to , judging Check if the condition is true; if true, return to step S4; otherwise, the algorithm execution ends. The output module is used to output the global optimal solution obtained by the calculation module as the optimal coal mining equipment production plan.

[0014] Thirdly, the present invention also provides a computer-readable storage medium storing a computer program for a coal mining equipment production planning optimization method based on an improved variable neighborhood search algorithm, wherein the computer program causes a computer to execute a coal mining equipment production planning optimization method based on an improved variable neighborhood search algorithm as described above.

[0015] Fourthly, the present invention also provides an electronic device, comprising: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including methods for performing coal mining equipment production planning optimization methods based on improved variable neighborhood search algorithms as described above.

[0016] (III) Beneficial Effects This invention provides a method for optimizing coal mining equipment production planning based on an improved variable neighborhood search algorithm. Compared with existing technologies, it has the following advantages: (1) This invention optimizes the initial solution by designing a total cost function, thereby increasing the quality of the initial solution. At the same time, it designs a greedy algorithm to further optimize the optimal solution, which can avoid the variable neighborhood search algorithm from getting trapped in local optima early and obtain a better scheduling scheme.

[0017] (2) The present invention is an optimization method for formulating equipment production plans with the goal of minimizing production costs and penalty costs, which improves the production efficiency and service quality of enterprises.

[0018] (3) By combining multiple objectives such as historical production, order demand, after-sales demand, production cost, production cycle, production efficiency, product quality and delivery time, this invention can more comprehensively and accurately reflect the complexity of the actual production environment, provide decision-makers with more effective decision support, and thus improve the efficiency and effectiveness of equipment mass production. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a coal mining equipment production planning optimization method based on an improved variable neighborhood search algorithm, provided by an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] This application provides a coal mining equipment production planning optimization method based on a variable neighborhood algorithm, which solves the problem that traditional production planning cannot accurately respond to actual market demand and production conditions, and realizes the accurate formulation and optimization of production plans.

[0023] The technical solution in this application is to solve the above-mentioned technical problems, and the general idea is as follows: The mass production process of equipment often involves multiple interrelated factors with complex nonlinear relationships, making the optimization problem exceptionally complex. Furthermore, traditional production planning optimization is often based on fixed parameters and heavily reliant on human experience, which not only fails to guarantee accuracy but also makes it impossible to update in real time according to actual market demand and labor conditions.

[0024] This study addresses the technical problems existing in the current mass production process of equipment. On the one hand, we need to construct a real-time optimization plan by combining multi-source data such as historical production, order demand, and after-sales demand. On the other hand, we need to consider the production plan of complete sets of equipment and spare parts, especially with the optimization objective of minimizing production costs and penalty costs. In terms of research methods, the main factors affecting the performance of the variable neighborhood search algorithm are: the quality of the initial solution, the neighborhood structure, and the search strategy within the neighborhood structure. In some cases, the limitations of the initial solution and the neighborhood structure can cause the variable neighborhood search algorithm to get stuck in local optima early on, failing to obtain a good scheduling scheme. Therefore, this invention makes corresponding improvements to the variable neighborhood search algorithm to solve this problem.

[0025] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0026] like Figure 1 As shown, a method for optimizing coal mining equipment production planning based on an improved variable neighborhood search algorithm includes: S1 initializes the various parameters in the algorithm, including production parameters and execution parameters. Production parameters include production information for complete sets of equipment and spare parts, customer information, delay penalty costs, production capacity constraints, order requirements, production costs, production efficiency, and delivery time. Execution parameters include the maximum number of iterations. Current iteration number Crossover probability Local search algorithm population size .

[0027] S2 prioritizes complete sets of equipment and spare parts according to the difference between processing time and expected delivery time; constructs a total cost function with the goal of minimizing production costs and delivery time penalty costs.

[0028] S3 generates an initial solution based on the priority sorting result, optimizes the initial solution with the goal of minimizing the total cost function, and sets the optimized initial solution as the global optimal solution.

[0029] S4. Generate and update the neighborhood solution set based on the global optimal solution to obtain the candidate solution set.

[0030] S5 uses roulette wheel probability selection to obtain the local search neighborhood structure.

[0031] S6. Obtain the neighborhood based on the candidate solution set and the local search neighborhood structure, and search for the local optimal solution.

[0032] S7. Based on the total cost function, determine the merits of the local optimal solution and the candidate solution set. If the local optimal solution is better, update the local optimal solution to the global optimal solution and increase the weight of the neighborhood structure that obtained the local optimal solution. If the candidate solution set is better, decrease the weight of the neighborhood structure that obtained the local optimal solution and return to step S5.

[0033] S8. Based on the greedy algorithm, the global optimal solution is greedily destroyed and greedily reconstructed to obtain a new solution. The new solution is compared with the global optimal solution. If the global optimal solution is better, the improved Boltzmann function is used to determine whether to accept the new solution and return to step S4; otherwise, the new solution is assigned as the global optimal solution.

[0034] S9, Assign to ,judge If the condition is met, return to step S4; otherwise, the algorithm ends and outputs the global optimal solution as the optimal coal mining equipment production plan.

[0035] In this embodiment, the initial solution is optimized by designing a total cost function, which increases the quality of the initial solution and avoids the variable neighborhood search algorithm from getting trapped in local optima early on. At the same time, the variable neighborhood search algorithm is used to optimize the coal mining equipment production planning method, which can obtain a better scheduling scheme.

[0036] It should be noted that each bit of the code in the solution of this application embodiment corresponds to a resource used in the production of a complete set of equipment or spare parts, and its position corresponds to its relative order in the production plan.

[0037] Step S1: Initialize the various parameters in the algorithm, including production parameters and execution parameters. Production parameters include complete equipment and spare parts production information, customer information, delay penalty costs, production capacity constraints, order requirements, production costs, production efficiency, and delivery time. Execution parameters include the maximum number of iterations. Current iteration number Crossover probability Local search algorithm population size .

[0038] Specifically, the production parameters include: Complete Equipment Index ; Spare parts index ; Customer Index ; Month Index ; Complete set of equipment Production time; Complete set of equipment Production completion time; :spare parts Production time; :spare parts Production completion time; Manufacturer Monthly production complete set of equipment The cost; Manufacturer Monthly production spare parts The cost; :client The ordered complete set of equipment, :client The ordered spare parts collection : customer : number of complete sets of equipment ordered by the customer : number of spare parts ordered by the customer : customer : unit delay penalty cost of the complete sets of equipment ordered by the customer : unit delay penalty cost of the spare parts ordered by the customer : customer : expected delivery time of the complete sets of equipment ordered by the customer : expected delivery time of the spare parts ordered by the customer : if the complete sets of equipment are allocated to the month : if the spare parts are allocated to the month : if the customer : if the customer : number of production equipment and spare parts in the : maximum production capacity in the : constraint requirement : if the complete sets of equipment : if the spare parts : if the customer : if the customer : if the complete sets of equipment : if the spare parts : if the customer : if the customer : number of production equipment and spare parts in the : maximum production capacity in the : constraint requirement : if the complete sets of equipment : if the spare parts : if the customer : if the customer : if the customer : if the customer : number of production equipment and spare parts in the : maximum production capacity in the : constraint requirement : if the complete sets of equipment : if the spare parts : if the customer : if the customer : if the customer : if the customer : if the customer : if the customer The embodiment can more comprehensively and accurately reflect the complexity of the actual production environment by considering various production parameters, provide more effective decision support for decision makers, and thus improve the efficiency and benefit of batch production of equipment.

[0039] Step S2, prioritize the sets of equipment and spare parts according to the difference between the processing time and the expected delivery time. Build a total cost function with the objective of minimizing the production cost and the delivery deadline penalty cost.

[0040] First, calculate the difference between the expected delivery time of the set of equipment and the set of spare parts: The processing time of all sets of equipment produced by the manufacturer and the set of spare parts and the difference between its corresponding expected delivery time is: Arrange the set of equipment and spare parts in non-decreasing order according to to form a priority order , where represents the spare parts, represents the set of equipment, and if there are the same , the priority is determined randomly.

[0041] Build a total cost function with the objective of minimizing the production cost and the delivery deadline penalty cost.

[0042] The total cost function is: In the formula, represents the total cost; represents the minimization of production cost; represents the minimization of delivery deadline penalty cost; represents the index of the set of equipment; represents the index of the spare parts; represents the index of the customer; represents the index of the month; represents the production time of the set of equipment ; represents the production completion time of the set of equipment ; represents the production time of the spare parts ; represents the production completion time of the spare parts ; represents the cost of the manufacturer producing the set of equipment in the month ; represents the cost of the manufacturer producing the set of spare parts in the month ; represents the set of equipment ordered by the customer ; represents the customer ordered spare parts set; representing a customer ordered equipment set ; representing a customer ordered spare parts ; representing a customer ordered equipment set ; representing a customer ordered equipment set ; representing a customer ordered spare parts ; representing a customer ordered spare parts ; representing if the equipment set is assigned to the month for production, it equals 1, otherwise it equals 0; representing if the spare parts is assigned to the month for production, it equals 1, otherwise it equals 0; representing if the customer ordered the equipment set , it equals 1, otherwise it equals 0; : if the customer ordered the spare parts , it equals 1, otherwise it equals 0.

[0043] Step S3, generating an initial solution according to the result of the priority ranking, optimizing the initial solution for the purpose of minimizing the total cost function, and setting the optimized initial solution as the global optimal solution. In the specific implementation of this step, the following steps are further included: Step S31, inserting a new production task into the production sequence for the purpose of minimizing the total cost function in the sorted production sequence.

[0044] allocating the equipment set and the spare parts in the priority order to establish an allocation list of the equipment set and the spare parts for each production machine of the manufacturer.

[0045] wherein, represents assigning the equipment set to the month production, represent spare parts are allocated to months production, in allocating the complete equipment and spare parts to months, all candidate positions in the current partial permutation list are inserted, and the position that makes the total cost minimum is selected.

[0046] That is the initial solution. Then the initial solution is optimized with the purpose of minimizing the total cost function, that is, in allocating the complete equipment to months, all candidate positions in the current partial permutation list are inserted, and the position that makes the total cost minimum is selected. The initial solution is optimized by multiple insertions.

[0047] The optimized initial solution is set as the global optimal solution, that is, .

[0048] Step S4, a neighborhood solution set is generated according to the global optimal solution and is updated to obtain a candidate solution set. The specific implementation of this step includes the following steps: S41, a neighborhood solution set is generated based on the global optimal solution, and the solution set is denoted as ; wherein, indicates that individuals in the neighborhood solution are considered, indicates the th individual in the neighborhood solution, which is obtained by randomly exchanging times from the initial solution ; S42, a variable is defined, which has the same meaning as , and the individuals in are assigned to ; S43, variables and are defined, the variable is set, the variable is set, and the variable is assigned to a random number generated in the interval ; wherein, and have the same dimension as the individual ; S44, the is updated by using the formula , wherein indicates the th element in the global optimal solution, , are respectively in the interval​​ a random number generated within the range; S45, generating a random number within the range a random number within the range and judging whether the condition is met, and if so, assigning to ; otherwise, assigning to ; S46, setting , judging whether the condition is met, and if so, returning to step S44, otherwise executing step S47; S47, calculating the fitness values of the individual and the intermediate according to the total cost function respectively, and comparing and , and if , assigning to the individual ; S48, setting , judging whether the condition is met, and if so, returning to step S43; otherwise executing step S49; S49, selecting different individuals with better quality from the neighborhood solution set and assigning them to the candidate solution set. Step S5, selecting the local search neighborhood structure based on the roulette probability. The specific implementation of this step includes the following steps: S51, presetting the initial weights of each local search neighborhood structure to be equal.

[0049] In the process of initially setting the algorithm parameters, for the local search neighborhood structure set of the variable neighborhood search algorithm , the initial weight of each neighborhood structure is

[0050] , and by default, the initial neighborhood structure weight is equal under the premise of unknown effectiveness of the local search neighborhood structure. S52, calculating the cumulative probability of each local search neighborhood structure according to the weight of each local search neighborhood structure and the roulette probability.

[0051] From the weight of each local search neighborhood structure, combined with the probability formula of roulette , the cumulative probability

[0052] of each neighborhood structure can be calculated, which is the sum of the selection probabilities of all previous individuals corresponding to each individual. In the formula, represents the cumulative probability of the neighborhood structure ; ; and ​Representing neighborhood structure The corresponding initial weights; This represents the sum of the weights of all neighborhood structures.

[0053] S53, select the neighborhood structure based on the cumulative probability to perform a local search.

[0054] Randomly generated ,if Then choose neighborhood structure Perform a local search.

[0055] The local search neighborhood structure in this embodiment includes ~ Specifically: Domain Structure Define a variable x, randomly generate an integer in the range [1, P] and assign it to variable x. Reverse the order of all codes to the left of x in the solution set of the neighborhood. Then select resources for all positions based on probability. Make a random selection.

[0056] Neighborhood structure Define a variable x, randomly generate an integer in the range [1, Q] and assign it to variable x. Reverse the encoding of all positions to the right of x in the neighborhood solution set. Then select resources for all positions based on probability. Make a random selection.

[0057] Neighborhood structure Define variables x and y, generate two integers in the range [1, P] and [P+1, P+Q] respectively, and assign them to variables x and y. The x-th position in the neighborhood solution set corresponds to the allocation month code of the complete set of equipment, and the y-th position corresponds to the allocation month code of the spare parts. Swap the month numbers corresponding to the x-th and y-th positions in the neighborhood solution set.

[0058] Neighborhood structure Define variables x, y, and P+Q=n. Generate two integers in the range [1, P] and [P+1, P+Q] respectively, and assign them to variables x and y. The x-th position in the neighborhood solution set corresponds to the allocation month code of the complete set of equipment, and the y-th position corresponds to the allocation month code of the spare parts. Reverse the order of all codes to the left of position x and all codes to the right of position y. Then, randomly select resources for all positions with probability (x+y) / 2n. Neighborhood structure : define variables x, y, P+Q=n, randomly generate two integers in the interval [1, n] without replacement, assign to variables x, y, where x

[0059] neighborhood structure : define variables x, y, z, P+Q=n, randomly generate three integers in the interval [1, n] without replacement, assign to variables x, y, z, where x

[0060] neighborhood structure : define variables x, y, z, P+Q=n, randomly generate three integers in the interval [1, n] without replacement, assign to variables x, y, z, where x

[0061] neighborhood structure : define variables x, y, z, P+Q=n, randomly generate three integers in the interval [1, n] without replacement, assign to variables x, y, z, where x

[0062] neighborhood structure : define variables , , randomly generate four integers in the interval [1, n] without replacement, assign to variables , where , reverse the code in the neighborhood solution set located inside the positions and , exchange the code inside the positions between and , then randomly select the resource selection of all positions with a probability of .

[0063] Step S6, obtain the neighborhood and search according to the candidate solution set and the local search neighborhood structure, and obtain the local optimal solution.

[0064] Step S7: Determine the relative merits of the local optimal solution and the candidate solution set based on the total cost function. If the local optimal solution is superior, update it to the global optimal solution and increase the weight of the neighborhood structure that obtained the local optimal solution. If the candidate solution set is superior, decrease the weight of the neighborhood structure that obtained the local optimal solution and return to step S5.

[0065] Step S8: Based on a greedy algorithm, the global optimal solution is greedily destroyed and greedily reconstructed to obtain a new solution. The new solution is compared with the global optimal solution. If the global optimal solution is better, the improved Boltzmann function is used to determine whether to accept the new solution, and the process returns to step S4; otherwise, the new solution is assigned as the global optimal solution. The specific implementation steps of this embodiment also include the following steps: S81, based on the production plan for all months corresponding to the global optimal solution. epsilon-greedy The rules select several months.

[0066] Specifically, in terms of probability Select the month with the highest total production cost (TC) in the current production stage; then, based on probability... Use a roulette wheel selection method to choose other months; if the maximum total production cost is the same for multiple months, then randomly select one of them.

[0067] S82, randomly select a complete set of equipment or spare parts from the month, and record the selected product set as S82. Let the remaining product set be denoted as .

[0068] Specifically, for each selected production month, one complete set of equipment or spare part is randomly selected as the product to be removed. The set of products extracted from the production month is denoted as […]. The remaining portion after removing these products is denoted as... .

[0069] S83, The products in the collection are inserted one by one. The set is used to calculate the total production cost based on the total cost function, taking into account all possible positions other than the initial position.

[0070] Specifically, products previously removed from the production month ( Insert one by one into At all possible locations (excluding their initial locations). For each insertion location, calculate the maximum total production cost (TC) of the system after insertion; when calculating the maximum total production cost (TC), products that are not inserted are ignored. The remaining part (of the text).

[0071] S84 sets the insertion position with the minimum total production cost as the new solution.

[0072] Specifically, compare all possible insertion positions and select the insertion position that minimizes the total production cost (TC) as the new solution. ).

[0073] In this embodiment, firstly, according to steps S81 to S84, the current global optimal solution ( Greedy insertion and greedy reconstruction are used to obtain a new solution. Then, the total generation cost of the global optimal solution is calculated based on the total cost function. ( ) and the new solution for total production cost ( Comparing the advantages and disadvantages of the two, if That is, when the global optimal solution is relatively good, generate ,if If so, accept the worse solution and set... = The new solution is then assigned as the global optimum. The improved Boltzmann function is: This function aims to facilitate escape from local optima during the search process; where, , indicating the solution and The relative percentage difference between the total production cost (TC). Then return to step S4.

[0074] like If the new solution is better at this point, then the new solution is assigned to the globally optimal solution. = Then proceed to the next step.

[0075] The coal mining equipment production plan optimization method in this embodiment breaks down and reconstructs the current global optimal solution to obtain a new solution. It compares the merits of the current solution and the new solution, accepts the better new solution or selectively accepts the worse new solution, and then returns to step S4 to iterate again. This avoids possible local optima and further optimizes the algorithm's ability to obtain the optimal solution.

[0076] Step S9, will Assign to ,judge If the condition is met, return to step S4; otherwise, the algorithm ends and outputs the global optimal solution as the optimal coal mining equipment production plan.

[0077] Thus, the entire process of this embodiment of the invention is completed.

[0078] This invention also provides a coal mining equipment production planning optimization system based on an improved variable neighborhood search algorithm, specifically including: A computing module is configured to perform the following steps: S1, initializing parameters in the algorithm, including production parameters and execution parameters; the production parameters include complete equipment and spare parts production information, customer information, delay penalty cost, production capacity constraint, order demand, production cost, production efficiency and delivery time; the execution parameters include maximum iteration number , current iteration number , crossover probability , population size of local search algorithm .

[0079] S2, prioritizing complete equipment and spare parts according to the difference between processing time and expected delivery time; constructing a total cost function with the goal of minimizing production cost and minimizing delivery period penalty cost.

[0080] S3, generating an initial solution according to the priority, optimizing the initial solution with the goal of minimizing the total cost function, and setting the optimized initial solution as the optimal solution.

[0081] S4, generating a neighborhood solution set from the global optimal solution and updating to obtain a candidate solution set.

[0082] S5, selecting a local search neighborhood structure based on roulette probability.

[0083] S6, obtaining a neighborhood and searching based on the candidate solution set and the local search neighborhood structure to obtain a local optimal solution.

[0084] S7, judging the advantages and disadvantages of the local optimal solution and the candidate solution set according to the total cost function; if the local optimal solution is better, updating the local optimal solution as the global optimal solution and increasing the weight of the neighborhood structure from which the local optimal solution is obtained; if the candidate solution set is better, decreasing the weight of the neighborhood structure from which the local optimal solution is obtained and returning to step S5.

[0085] S8, obtaining a new solution by greedily destroying and reconstructing the global optimal solution based on a greedy algorithm, comparing the new solution with the global optimal solution, judging whether to accept the new solution based on an improved Boltzmann function if the global optimal solution is better, and returning to step S4; otherwise, assigning the new solution as the global optimal solution.

[0086] S9, assigning to , judging whether is true, and returning to step S4 if true; otherwise, ending the algorithm and outputting the global optimal solution as the optimal coal machine equipment production plan scheme.

[0087] An output module is configured to output the global optimal solution obtained by the computing module as the optimal coal machine equipment production plan scheme.

[0088] It can be understood that the coal machine equipment production plan optimization system based on the improved variable neighborhood search algorithm provided by the embodiments of the present application corresponds to the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm, and the explanation, examples, beneficial effects and other parts of the related content can refer to the corresponding content in the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm, which will not be repeated here.

[0089] The embodiments of the present application also provide a computer readable storage medium storing a computer program for the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm, wherein the computer program enables a computer to execute the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm as described above.

[0090] The embodiments of the present application also provide an electronic device, comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs comprise a program for executing the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm as described above.

[0091] In summary, compared with the prior art, the present application has the following beneficial effects: 1、The embodiments of the present application optimize the initial solution by designing a total cost function, increase the quality of the initial solution, optimize the search strategy of the neighborhood search by the self-designed neighborhood structure, and further optimize the optimal solution by designing a greedy algorithm, so that the variable neighborhood search algorithm can avoid falling into a local optimum early, and an approximately optimal solution can be obtained for the coal machine equipment production plan optimization problem, so that the enterprise can fully utilize its production resources to the maximum extent, reduce production costs, and improve the enterprise service level and customer satisfaction level.

[0092] 2、The embodiments of the present application optimize the equipment production plan by taking the minimization of production cost and penalty cost as the optimization target, and improve the production efficiency and service quality of the enterprise.

[0093] 3、The embodiments of the present application can more comprehensively and accurately reflect the complexity of the actual production environment by combining historical production, order demand, after-sales demand, production cost, production cycle, production efficiency, product quality and delivery time, and provide more effective decision support for decision makers, so as to improve the efficiency and benefit of equipment batch production.

[0094] 4、The embodiments of the present application first encode the months, allocate the workpieces to each month through priority sorting, obtain the corresponding production plan scheme, and then continuously optimize the production plan scheme by using the improved variable neighborhood search algorithm, which performs excellently in speed and quality, and can greatly improve the production efficiency of the enterprise.

[0095] It is to be noted that the relational terms herein, such as first and second, and the like, are used solely to distinguish one from another entity or action without necessarily requiring or implying any actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0096] The above embodiments are only used to illustrate the technical solutions of the present application, not to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still make modifications to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A coal machine equipment production planning optimization method based on an improved variable neighborhood search algorithm, characterized in that, The method comprises the following steps: S1, initializing parameters in the algorithm, including production parameters and execution parameters; the production parameters include complete equipment and spare parts production information, customer information, delay penalty cost, production capacity constraint, order demand, production cost, production efficiency and delivery time; the execution parameters include maximum iteration number , current iteration number , crossover probability , local search algorithm population size ; S2, prioritizing the complete sets of equipment and spare parts according to the difference between the processing time and the expected delivery time; constructing a total cost function aiming at minimizing the production cost and the delivery period penalty cost; S3, encoding and generating an initial solution according to the result of the prioritization, and setting the initial solution as a global optimal solution; S4, generating a neighborhood solution set according to the global optimal solution and updating to obtain a candidate solution set; S5, selecting a local search neighborhood structure based on a roulette probability; S6, obtaining a neighborhood and searching according to the candidate solution set and the local search neighborhood structure to obtain a local optimal solution; S7, judging the local optimal solution and the candidate solution set according to the total cost function, if the local optimal solution is better, updating the local optimal solution as the global optimal solution, and increasing the weight of the neighborhood structure for obtaining the local optimal solution; if the candidate solution set is better, decreasing the weight of the neighborhood structure for obtaining the local optimal solution, and returning to step S5; S8, obtaining a new solution by greedily destroying and reconstructing the global optimal solution based on a greedy algorithm, comparing the new solution with the global optimal solution, if the global optimal solution is better, judging whether to accept the new solution based on an improved Boltzmann function, and returning to step S4; otherwise, setting the new solution as the global optimal solution; S9, the is assigned to , determine if it is true, if it is true, return to step S4; otherwise, output the global optimal solution as the optimal coal machine equipment production plan scheme.

2. The coal handling plant production planning optimization method based on improved variable neighborhood search algorithm according to claim 1, characterized in that, The roulette probability selection comprises: presetting the initial weights of each local search neighborhood structure as equal; calculating the cumulative probability of each local search neighborhood structure according to the weight of each local search neighborhood structure and the roulette probability; selecting a neighborhood structure for local search based on the cumulative probability.

3. The coal handling plant production planning optimization method based on improved variable neighborhood search algorithm according to claim 1, characterized in that, The obtaining of the new solution by greedily destroying and reconstructing the global optimal solution based on the greedy algorithm comprises: in all months of the production plan based on epsilon-greedy the global optimal solution; A complete set of equipment or spare parts is randomly extracted in the month, and the extracted product set is recorded as The remaining product set is recorded as ; inserting the products in the set one by one into all possible positions of the set except the initial position and calculating the total production cost based on the total cost function; inserting the products in the set one by one into all possible positions of the set except the initial position and calculating the total production cost based on the total cost function; inserting the products in the set one by one into all possible positions of the set except the initial position and setting the insertion position with the minimum total production cost as the new solution.

4. The coal handling plant production planning optimization method based on improved variable neighborhood search algorithm according to claim 2, characterized in that, The local search neighborhood structure comprises wherein, represents the index number of the devices, represents the index number of the spare parts;​ The To define the variable x, an integer in the interval [1, P] is randomly generated and assigned to the variable x, all the codes to the left of the position x in the solution set of the field are processed in reverse order, and then the resource selection of all positions is performed with a probability random selection; The To define the variable x, an integer in the interval [1, Q] is randomly generated and assigned to the variable x, and all the codes to the right of the position x in the neighborhood solution set are processed in reverse order, and then the resource selection of all positions is performed with a probability randomly selected; The To define the variables x, y, two integers in the interval [1, P] and [P+1, P+Q] respectively are generated, and are assigned to the variables x, y, wherein the xth position of the neighborhood solution set corresponds to the allocation month code of the complete set of equipment, and the yth position corresponds to the allocation month code of the spare parts, and the month numbers corresponding to the xth and yth positions of the neighborhood solution set are exchanged; The To define the variables x, y, P+Q=n, two integers in the range of [1, P] and [P+1, P+Q] are generated respectively, and assigned to the variables x, y, wherein the xth position in the neighborhood solution corresponds to the allocation month code of the complete set of equipment, and the yth position corresponds to the allocation month code of the spare parts. All codes to the left of the position x and all codes to the right of the position y are processed in reverse order, and then the resource selection of all positions is randomly selected with a probability of (x+y) / 2n. The To define the variables x, y, P+Q=n, without replacement, randomly generate two integers in the range [1, n] and assign them to the variables x, y, where x≤y, exchange the codes in the neighborhood solution set located on both sides of the position x, y, and then randomly select the resource selection at all positions with a probability of (x+y) / 2n; The To define the variables x, y, z, P+Q=n, three integers in the range [1, n] are randomly obtained without replacement, and are assigned to the variables x, y, z, wherein x<y<z, the codes on both sides of the positions x, z in the neighborhood solution are reversed, the codes inside the positions x, y and y, z are exchanged, and then the resource selection of all positions is randomly selected with a probability of (x+y+z) / 3n. The To define the variables x, y, z, P+Q=n, three integers in the range [1, n] are randomly obtained without replacement, and are assigned to the variables x, y, z, wherein x<y<z, the codes on both sides of the positions x, z in the neighborhood solution are exchanged, the codes inside the positions x, y and y, z are reversed in order, and then the resource selection of all positions is randomly selected with a probability of (x+y+z) / 3n. The To define variables x, y, z, P+Q=n, three integers in the range of [1, n] are randomly obtained without replacement, and are assigned to variables x, y, z, wherein x<y<z, the codes located on both sides of x, z in the neighborhood solution and the codes inside x, y and y, z are reversed in order, and then the resource selection of all positions is randomly selected with a probability of (x+y+z) / 3n. The To define the variable , , four integers in the range [1, n] are randomly taken without replacement and assigned to the variable , where , the codes in the neighborhood solution and are reversed, and the codes between and are exchanged, and then the resource selection for all positions is randomly selected with a probability of .

5. The coal handling plant production planning optimization method based on improved variable neighborhood search algorithm according to claim 1, characterized in that, The obtaining of the candidate solution set specifically comprises: S41, a neighborhood solution set is generated based on the initial solution, and the solution set is denoted as... ;in, The CCP is considering a solution to the neighboring region issue. Individual, The first solution in the neighborhood is represented by the second solution. There are individuals, and each individual is determined by the initial solution. Random exchange This was the first time it was obtained; S42, define variable with the same meaning as and assign the individual in to ; S43, Define variables and Let the variable Let the variable , will variables Assigned in the interval Random numbers generated within a range; where, and With individuals Same dimensions; S44, using the formula update where denotes the i-th element in the global optimal solution, , are random numbers generated in the interval ;​ S45, in the interval Generate random numbers within the range and judge Is it true? If it is true, then... Assign to Otherwise Assign to ; S46, let , determine if it is true, if it is true, return to step S44, otherwise execute step S47; S47, calculating the individual fitness value according to the total cost function and intermediates fitness value and and comparing and if then assigning to the individual ; S48, let , determine if it is true, if it is true, return to step S43; otherwise, execute step S49; S49, screening different individuals with better quality in the neighborhood solution set according to the total cost function, and assigning the individuals to the candidate solution set.

6. The coal handling plant production planning optimization method based on improved variable neighborhood search algorithm according to claim 1, characterized in that, The total cost function is: In the formula, Indicates the total cost; This indicates minimizing production costs; This indicates minimizing the cost of delivery time penalties; Indicates a complete set of equipment index; Indicates spare parts index; Indicates the customer index; Indicates the month index; Indicates complete set of equipment Production time; Indicates complete set of equipment Production completion time; Indicates spare parts Production time; Indicates spare parts Production completion time; Indicates manufacturer Monthly production complete set of equipment The cost; Indicates manufacturer Monthly production spare parts The cost; Indicates customer The ordered complete set of equipment; Indicates customer The collection of ordered spare parts; Indicates customer Ordered complete sets of equipment Quantity; Indicates customer Ordered spare parts Quantity; Indicates customer Ordered complete sets of equipment The unit delay penalty cost; Indicates customer Ordered complete sets of equipment Expected delivery time; Indicates customer Ordered spare parts The unit delay penalty cost; Indicates customer Ordered spare parts Expected delivery time; Indicates if complete set of equipment Allocated to months If production is initiated, the value is 1; otherwise, it is 0. Indicates if spare parts Allocated to months If production is initiated, the value is 1; otherwise, it is 0. Indicates if the customer ordered a complete plant equals 1, otherwise 0. : if the customer ordered spare parts equals 1, otherwise 0.

7. The coal handling plant production planning optimization method based on improved variable neighborhood search algorithm according to claim 1, characterized in that, The method further comprises: inserting a new production task into the production sequence of the initial solution in the sorted production sequence aiming at minimizing the total cost function.

8. A coal machine equipment production planning optimization system based on an improved variable neighborhood search algorithm, characterized in that, The method comprises the following steps: S2, prioritizing the complete sets of equipment and spare parts according to the difference between the processing time and the expected delivery time; constructing a total cost function aiming at minimizing the production cost and the delivery period penalty cost; S1, initializing parameters in the algorithm, including production parameters and execution parameters; the production parameters include complete equipment and spare parts production information, customer information, delay penalty cost, production capacity constraint, order demand, production cost, production efficiency and delivery time; the execution parameters include maximum iteration number , current iteration number , crossover probability , population size of local search algorithm ; S3, generating an initial solution according to the prioritization, and optimizing the initial solution aiming at minimizing the total cost function, and setting the optimized initial solution as an optimal solution; S4, generating a neighborhood solution set according to the initial solution and updating to obtain a candidate solution set; S5, selecting a local search neighborhood structure based on a roulette probability; S6, obtaining a neighborhood and searching according to the candidate solution set and the local search neighborhood structure to obtain a local optimal solution; ​ S7, judging the local optimal solution and the candidate solution set according to the total cost function, if the local optimal solution is better, updating the local optimal solution as the global optimal solution, and increasing the weight of the neighborhood structure of obtaining the local optimal solution; if the candidate solution set is better, reducing the weight of the neighborhood structure of the local optimal solution, and returning to step S5; S8, based on the greedy algorithm, greedily destroying and greedily reconstructing the global optimal solution to obtain a new solution, and comparing the new solution with the global optimal solution, if the global optimal solution is better, judging whether to accept the new solution based on the improved Boltzmann function, and returning to step S4; otherwise, assigning the new solution as the global optimal solution; S9, the assignment is given to , determine whether it is true, if true, return to step S4; otherwise, the algorithm ends; An output module is configured to output the global optimal solution obtained by the calculation module as the optimal coal machine equipment production plan scheme.

9. A computer-readable storage medium, characterized in that, The computer program for the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm is stored, wherein the computer program enables the computer to execute the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm according to any one of claims 1-7.

10. An electronic device, comprising: The computer program for the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm is stored, wherein the computer program enables the computer to execute the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm according to any one of claims 1-7. One or more processors, memories, and one or more programs, wherein the one or more programs are stored in the memories and configured to be executed by the one or more processors, and the programs include the coal machine equipment production plan optimization method based on the improved variable neighborhood search algorithm according to any one of claims 1-7.

Images