Method for generating coal distribution strategy, storage medium and electronic equipment

Abstract

The invention provides a method for generating a coal distribution strategy, a storage medium and electronic equipment, after coal quality data of a plurality of coal types are acquired in real time, when distribution needs to be realized, coal is blended according to a plurality of coal blending schemes in response to a distribution signal to obtain a plurality of pre-blended coal materials, each coal blending scheme needs to meet the condition that the proportion of each coal type is any value between [0, 1], and the proportion of each coal type is any value between [0, 1]; the sum of the proportions of all coal types in the pre-blended coal material obtained after coal blending is completed is 1. Searching an optimal solution in multiple coal blending schemes through a probability evolution population optimization algorithm to serve as a target coal blending strategy; in the probability evolution population optimization algorithm, the profit maximization is taken as a target, and the coal quality demand of the pre-blended coal material is taken as a constraint condition. According to the scheme, the maximum income is taken as the target, the coal quality demand is taken as the constraint condition, and the probability evolution population optimization algorithm is combined to realize automatic and accurate calculation of the coal quality and the income, so that the coal blending decision process is more accurate, quicker and more automatic, and the coal blending scheme with the optimal global price is quickly found.

Description

Technical Field

[0001] The present application relates to the technical field related to coal proportioning and sales, and in particular, to a method, storage medium and electronic device for generating a coal distribution strategy. Background Art

[0002] With the increasing demand for coal quality, a single type of coal often cannot meet specific coal quality requirements. Therefore, different coal types need to be blended to meet the dual requirements of specific coal quality indicators and economic benefits. Furthermore, coal blending allows for flexible response to price fluctuations and supply changes among different coal types in the market, thereby improving business flexibility and economic efficiency. Existing coal blending plans involve establishing a target coal quality range within which the target coal type meets the requirements. The coal quality data, benchmark prices, and corresponding price reward and penalty schemes for the two coal types participating in the blending plan are then manually retrieved. The target coal quality data for the resulting blend is then weighted based on the proportions of the two coal types involved in the blend. The target coal type's compliance with the target coal quality range is then used to determine whether the plan should be retained. For any retained plans, the final sales price is calculated based on the estimated target coal quality and the price reward and penalty scheme.

[0003] Existing technologies target the simple scenario of blending two types of coal. However, in real-world production processes, multiple different types of coal often need to be blended to achieve the required quality and maximize post-blending profits. Due to the large number of parameters involved and their flexible values, it's impossible to traverse all blending options. For this scenario, existing solutions often rely on experience, trying several blending parameter combinations before selecting a feasible option that empirically yields the highest blending profit. This solution requires significant computational effort and struggles to meet real-time requirements. Summary of the Invention

[0004] The technical problem to be solved by this application is that the traditional coal blending method relies on manual experience judgment and simple linear calculation. Since a large amount of data query and manual calculation are involved in the process, it is difficult to accurately control the coal quality, and the time cost of generating a distribution plan is high. Therefore, a method, storage medium and electronic equipment for generating a coal distribution strategy are provided.

[0005] In a first aspect, the technical solution of the present application provides a method for generating a coal distribution strategy, comprising:

[0006] Obtain coal quality data of multiple coal types in real time;

[0007] In response to the distribution signal, multiple coal blending schemes are used to blend coal to obtain multiple pre-blended coal materials. Each coal blending scheme must meet the following requirements: the proportion of each coal type is any value between [0, 1], and the sum of the proportions of all coal types in the pre-blended coal materials obtained after the coal blending is completed is 1;

[0008] The optimal solution among multiple coal blending schemes is searched based on a probabilistic evolutionary population optimization algorithm as a target coal blending strategy. In the probabilistic evolutionary population optimization algorithm, profit maximization is taken as the goal, and the coal quality requirement of pre-blended coal is taken as a constraint condition.

[0009] Preferably, in some embodiments, the method for generating a coal distribution strategy, wherein the method searches for the optimal solution among multiple coal distribution schemes based on a probabilistic evolutionary population optimization algorithm as the final coal distribution strategy, wherein the probabilistic evolutionary population optimization algorithm takes maximizing profit as the goal and takes the coal quality requirements of the pre-blended coal as the constraint condition, includes:

[0010] Construct the objective function and calculate the fitness of each coal blending scheme:

[0011]

[0012] Penalty term = ∑ 指标数 a j ·(index j (w)-standard value j ) 2 ;

[0013] S target is the theoretical sales price of pre-blended coal obtained from the coal blending plan, n represents the number of coal types involved in coal blending, S i is the theoretical sales price of the i-th coal type involved in coal blending, w i is the proportion of coal blending of the i-th coal type, λ is the regularization coefficient; a j is the penalty coefficient of the jth coal quality index of the pre-mixed coal; j (w) is the actual value of the jth indicator, the standard value j is the standard value of the jth indicator, and maximizing the benefit is to maximize the calculation result of the objective function F(w);

[0014] Initialization constraints: Filter according to the set coal quality requirements and only retain the coal blending plans that meet the coal quality requirements.

[0015] Preferably, in some embodiments of the method for generating a coal distribution strategy, the method of searching for the optimal solution among multiple coal distribution schemes based on a probabilistic evolutionary population optimization algorithm as the final coal distribution strategy further includes:

[0016] Initialization of the population: Screen all types of coal with inventory and generate a random set of initialization coal blending schemes, which are stored in different areas of the search space;

[0017] Probabilistic evolutionary update: In each iteration cycle, the search probability distribution of different areas in the search space is determined based on the fitness calculation results in the previous iteration cycle;

[0018] Probabilistic mutation: Based on the search probability distribution obtained in the probabilistic evolution update step, some coal blending schemes are selected for mutation to obtain the mutated coal blending scheme;

[0019] Convergence and termination: When the average fitness of each coal blending scheme in the coal blending scheme collection obtained from multiple consecutive iteration cycles is less than the set threshold or the number of iterations reaches the upper limit, the iteration is stopped;

[0020] Output the final coal blending strategy: Select the coal blending scheme that satisfies the constraints and maximizes the calculation result of the objective function F(w) as the final coal blending strategy, and output the proportion vector of each coal type w=(w1,w2,…,w n ).

[0021] Preferably, the method of generating a coal distribution strategy in some embodiments further comprises:

[0022] Calculate the profit of the target pre-provisioned coal obtained by the final coal blending strategy: the profit is obtained by subtracting the sum of the current theoretical sales prices of each type of coal in the target pre-provisioned coal from the current actual sales price of the target pre-provisioned coal.

[0023] Preferably, in some embodiments of the method for generating a coal distribution strategy, the initialization constraint condition is as follows: screening is performed according to the set coal quality requirements, and only coal distribution plans that meet the coal quality requirements are retained, and the coal distribution plans are screened in the following manner:

[0024] Set screening thresholds;

[0025] a screening step, performing a difference calculation on the coal quality of the pre-blended coal obtained from the coal quality requirement and the coal blending plan, and filtering out the coal blending plan if the absolute value of the difference calculation result is greater than the screening threshold;

[0026] The number of coal blending schemes obtained after screening is determined. If the number of coal blending schemes is less than a set number, the screening threshold is lowered and the process returns to the screening step.

[0027] Preferably, in the method for generating a coal distribution strategy described in some solutions, in the probability evolution update, the search probability distribution of different regions in the search space is determined by the following probability update formula:

[0028] p (t+1) (w) = p (t) (w) + β · [fitness (t) - average fitness (t)] · random adjustment term;

[0029] Among them, p (t) (w) represents the search probability of the area where the coal blending scheme is located in the previous iteration cycle, p (t+1)(w) represents the search probability of the area where the coal blending scheme is located in the current iteration cycle, β represents the evolutionary step factor, fitness (t) represents the fitness calculation result of the coal blending scheme in the previous iteration cycle, and average fitness (t) represents the average fitness calculation result of all coal blending schemes searched in the previous iteration cycle; the random adjustment item is a randomly generated number between [-1,1].

[0030] Preferably, in the method for generating a coal distribution strategy described in some solutions, in the probability variation step, the coal distribution plan is mutated in the following manner:

[0031] w i ′ =w i +γ·rand(-∈,∈);

[0032] Among them, γ is the variation factor, rand is the function used to generate random numbers, ∈ is the variation range constant, and w i ′ It represents the proportion of coal blending after the mutation of the i-th coal type.

[0033] In a second aspect, the technical solution of the present application provides a computer-readable storage medium, in which program information is stored. After a computer reads the program information, the computer executes the steps of the method for generating a coal distribution strategy as described in any one of the first aspects.

[0034] In a third aspect, the technical solution of the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for generating a coal distribution strategy as described in any one of the first aspects.

[0035] In a fourth aspect, the technical solution of the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method for generating a coal distribution strategy as described in any one of the first aspects.

[0036] Compared with the existing technology, the above technical solution provided by this application has the following technical effects:

[0037] The method, storage medium and electronic device for generating coal distribution strategies provided by the present application obtain the coal quality data of multiple types of coal in real time. When distribution needs to be realized, it responds to the distribution signal and distributes coal according to multiple coal distribution schemes to obtain multiple pre-distributed coal materials. Each coal distribution scheme must meet the following requirements: the proportion of each type of coal is any value between [0,1], and the sum of the proportions of all types of coal in the pre-distributed coal materials obtained after the coal distribution is completed is 1. The optimal solution among multiple coal distribution schemes is searched as the target coal distribution strategy through the probability evolutionary population optimization algorithm; in the probability evolutionary population optimization algorithm, the goal is to maximize the profit, and the coal quality requirements of the pre-distributed coal materials are used as constraints. The above-mentioned scheme of the present application obtains the optimal coal distribution scheme based on the intelligent optimization algorithm of the probability evolutionary population, which can significantly reduce the system search time and computing cost. With the maximum profit as the goal and the coal quality requirements as the constraints, the probability evolutionary population optimization algorithm is combined to realize the automatic and accurate calculation of coal quality and profit, making the coal distribution decision-making process more accurate, fast and automated, and quickly finding the coal distribution scheme with the best global price, thereby improving the economic benefits of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A flowchart of a method for generating a coal distribution strategy according to an embodiment of the present application;

[0039] Figure 2 A flowchart of a method for generating a coal distribution strategy according to another embodiment of the present application;

[0040] Figure 3 A schematic diagram of the hardware connection relationship of electronic devices for executing the method for generating coal distribution strategy described in this application. DETAILED DESCRIPTION

[0041] The specific implementation of this application is further described below with reference to the accompanying drawings.

[0042] It is easy to understand that according to the technical solution of this application, a variety of structural methods and implementation methods can be replaced with each other by those skilled in the art without changing the essential spirit of this application. Therefore, the following specific embodiments and drawings are only exemplary descriptions of the technical solution of this application and should not be regarded as the entire application or as a limitation or restriction of the technical solution of the application.

[0043] This embodiment provides a method for generating a coal distribution strategy, which can be applied to the control system of a coal enterprise, such as Figure 1 As shown, the method includes:

[0044] S110: Acquire coal quality data of multiple coal types in real time.

[0045] In this step, the coal quality data of multiple coal types can be obtained and automatically updated in real time to form a real-time coal quality database. The coal quality data includes key indicators such as ash melting point, sulfur content, calorific value, and ash content.

[0046] S120: In response to the distribution signal, coal is blended according to multiple coal blending schemes to obtain multiple pre-blended coal materials. Each coal blending scheme must meet the following requirements: the proportion of each coal type is any value between [0, 1], and the sum of the proportions of all coal types in the pre-blended coal materials obtained after coal blending is completed is 1.

[0047] Assuming that there are currently n types of coal available for distribution, there can be an infinite number of distribution combination plans due to the different proportions of each type of coal.

[0048] When distributing, the coal quality of each type of coal is weighted and summed, and then the coal quality of the coal obtained after distribution is calculated. For any coal quality P, when n types of coal are distributed, the formula for weighted summation is:

[0049]

[0050] Where X is the normalized weight of the coal types participating in the distribution, and i is the serial number of all coal types participating in the distribution. For the case of multi-coal blending, the ash melting point cannot be obtained by linear weighting, and its calculation complexity is extremely high. This application scheme constructs a nonlinear ash melting point prediction model, which is defined as:

[0051] A target =f(Q1,Q2,…,Q n ,w1,w2,…,w n );

[0052] Among them, Q i is the ash melting point and other ash components of the i-th coal type, w i The function f is nonlinear and can be determined using machine learning model predictions or historical data fitting. The model's predictions can be used to determine whether the coal blending plan meets the constraints. This step uses automated data processing to determine the coal quality for different coal blending plans, reducing manual repetitive calculations and improving real-time decision-making efficiency.

[0053] S130: searching for the optimal solution among multiple coal blending schemes as the target coal blending strategy based on the probabilistic evolutionary population optimization algorithm; in the probabilistic evolutionary population optimization algorithm, the goal is to maximize the profit, and the coal quality requirement of the pre-blended coal is used as a constraint condition.

[0054] The above-mentioned solution of the present application, after obtaining the coal quality data of multiple types of coal in real time, when it is necessary to realize distribution, responds to the distribution signal and distributes the coal according to multiple coal distribution schemes to obtain multiple pre-distributed coal materials. Each coal distribution scheme must meet the following requirements: the proportion of each type of coal is any value between [0,1], and the sum of the proportions of all types of coal in the pre-distributed coal materials obtained after the coal distribution is completed is 1. The optimal solution among the multiple coal distribution schemes is searched as the target coal distribution strategy through the probabilistic evolutionary population optimization algorithm; in the probabilistic evolutionary population optimization algorithm, the profit maximization is taken as the goal, and the coal quality requirements of the pre-distributed coal materials are taken as the constraints. The above-mentioned solution of the present application obtains the optimal coal distribution scheme based on the intelligent optimization algorithm of the probabilistic evolutionary population, which can significantly reduce the system search time and computing cost. With the maximum profit as the goal and the coal quality requirements as the constraints, the probabilistic evolutionary population optimization algorithm is combined to realize the automatic and accurate calculation of coal quality and profit, making the coal distribution decision-making process more accurate, fast and automated, and quickly finding the coal distribution scheme with the best global price, thereby improving the economic benefits of the enterprise.

[0055] Further preferably, in the above-mentioned method for generating a coal distribution strategy, the step S130 of searching for the optimal solution among multiple coal blending schemes based on the probabilistic evolutionary population optimization algorithm as the final coal blending strategy, wherein the probabilistic evolutionary population optimization algorithm takes maximizing profit as the goal and takes the coal quality requirement of the pre-blended coal as the constraint condition, includes:

[0056] Construct the objective function and calculate the fitness of each coal blending scheme:

[0057]

[0058] Penalty term = ∑ 指标数 a j ·(index j (w)-standard value j ) 2 ;

[0059] S target is the theoretical sales price of pre-blended coal obtained from the coal blending plan, n represents the number of coal types involved in coal blending, S i is the theoretical sales price of the i-th coal type involved in coal blending, w i is the proportion of coal blending of the i-th coal type, λ is the regularization coefficient; a j is the penalty coefficient of the jth coal quality index of the pre-mixed coal; j (w) is the actual value of the jth indicator, the standard value jis the standard value of the jth indicator. Maximizing the benefit is to maximize the calculated result of the objective function F(w). The required coal quality indicators are used as constraints to calculate the fitness function for each coal blending scheme. Regarding the penalty term, for example, when the actual ash melting point is lower than the standard value, which may lead to severe furnace coking, the penalty term increases significantly. In this case, the coal quality results obtained by this coal blending scheme are approximately non-compliant.

[0060] Initialization constraints: Filter according to the set coal quality requirements, retaining only coal blending solutions that meet these requirements. That is, the distributed coal must meet the coal quality requirements, otherwise sales will be affected. Specifically, this includes: filtering according to the set coal quality requirements, retaining only coal blending solutions that meet these requirements, and filtering coal blending solutions through the following methods: setting a screening threshold; performing a difference calculation between the coal quality requirements and the coal quality of the pre-blended coal obtained from the coal blending solution, and filtering out the coal blending solution if the absolute value of the difference calculation result is greater than the screening threshold; determining the number of coal blending solutions obtained after screening, and if the number of coal blending solutions is less than the set number, lowering the screening threshold and returning to the screening step. That is, if too few coal type combinations remain after screening, the filtering conditions can be appropriately relaxed to generate more alternative coal blending solutions.

[0061] like Figure 2 As shown, the search for the optimal solution among multiple coal blending schemes as the final coal blending strategy based on the probabilistic evolutionary population optimization algorithm in step S130 includes:

[0062] Initialize the population: Filter all available coal types to generate a random set of initial blending solutions, which are stored in different regions of the search space. In this step, all coal types with a stock greater than 0 are filtered out to generate a random set of initial blending solutions. In each blending solution, the percentage of each coal type is a random number between 0 and 1, and the sum of the percentages of all coal types is 1. The initial blending solution set is represented using a probability distribution, so that each initial blending solution is distributed in different regions of the search space.

[0063] Probabilistic evolutionary update: In each iteration, the search probability distribution of different regions in the search space is determined based on the fitness calculation results of the previous iteration. That is, the probability distribution of the coal blending solution in the search space is adjusted based on the fitness performance of the previous generation of solutions. Preferably, in the probabilistic evolutionary update, the search probability distribution of different regions in the search space is determined using the following probability update formula:

[0064] p (t+1) (w) = p (t) (w) + β · [fitness (t) - average fitness (t)] · random adjustment term;

[0065] Among them, p (t)(w) represents the search probability of the area where the coal blending scheme is located in the previous iteration cycle, p (t+1) (w) represents the search probability for the region containing the coal blending solution in the current iteration, β represents the evolutionary step size factor, fitness (t) represents the calculated fitness of the coal blending solution in the previous iteration, and average fitness (t) represents the average fitness calculated for all coal blending solutions found in the previous iteration. The random adjustment term is a randomly generated number between -1 and 1. The setting of β depends on the total number of coal types and the time requirements of the algorithm. In practice, a larger β results in a faster search, but also a coarser granularity in the search for the optimal solution. The random adjustment term is used to ensure search diversity.

[0066] Probabilistic mutation: Based on the search probability distribution obtained in the probability evolution update step, select some coal blending schemes for mutation to obtain the mutated coal blending scheme. That is, after each probability update, randomly select some schemes for mutation according to a certain probability (10-20% is found to be more appropriate in practice). Preferably, the coal blending scheme is mutated in the following way: w i ′ =w i +γ·rand(-∈,∈); where γ is the variation factor (usually selected from 0.1 to 0.5), rand is the function used to generate random numbers, and ∈ is the variation range constant (usually selected from 0.05 to 0.15). It is affected by the actual field stock fluctuations, w i ′ It represents the proportion of coal blending after the mutation of the i-th coal type.

[0067] Convergence and termination: The iteration is stopped when the average fitness of each coal blending scheme in the coal blending scheme set obtained in multiple consecutive iterative cycles is less than the set threshold or the number of iterations reaches the upper limit.

[0068] Output the final coal blending strategy: Select the coal blending scheme that satisfies the constraints and maximizes the calculation result of the objective function F(w) as the final coal blending strategy, and output the proportion vector of each coal type w=(w1,w2,…,w n ).

[0069] In the above scheme, the intelligent optimization algorithm based on probabilistic evolving population and the coal type pre-filtering mechanism can significantly reduce the system search time and computing cost.

[0070] Furthermore, the method further includes calculating the revenue of the target pre-blended coal obtained from the final coal blending strategy by subtracting the sum of the current theoretical sales prices of each type of coal in the target pre-blended coal from the current actual sales price of the target pre-blended coal. This allows for rapid identification of the globally optimal coal blending solution, thereby improving the company's economic benefits.

[0071] In specific implementation, the sales price corresponding to the coal distribution plan is automatically calculated based on the benchmark price of the coal type and the corresponding price reward and penalty plan. At the same time, based on the selling prices and distribution ratios of the two coal types participating in each distribution plan, the average sales price before distribution is calculated, and then this price is subtracted from the aforementioned target coal sales price to obtain the price change after distribution. Price penalties refer to the definition of a benchmark price for each type of coal when it is sold, as well as a range of several major coal qualities for that type of coal. During the actual distribution process, some of the major coal qualities of the target coal obtained may not fall within the preset range. For values ​​outside the range, the actual price will be adjusted based on the preset price reward and penalty rules when pricing.

[0072] An embodiment of the present application also provides a computer-readable storage medium, in which program information is stored. After the computer reads the program information, the computer executes the steps of the method for generating a coal distribution strategy described in any of the above solutions.

[0073] An embodiment of the present application also provides a computer program product, including a computer program / instruction, characterized in that when the computer program / instruction is executed by a processor, the steps of the method for generating a coal distribution strategy described in any of the above solutions are implemented.

[0074] The present application also provides an electronic device, such as Figure 3As shown, the electronic device includes at least one processor 31 and at least one memory 32. At least one memory 32 stores program information. After reading the program information, the at least one processor 31 executes the method for generating a coal distribution strategy described in any of the above method embodiments. The device may also include an input device 33 and an output device 34. The processor 31, memory 32, input device 33, and output device 34 are communicatively connected. Memory 32, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. By running the non-volatile software programs, instructions, and modules stored in memory 32, the processor 31 executes various functional applications and processes data, thereby implementing the method for generating a coal distribution strategy described in any of the above methods. Memory 32 may include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for the function; the data storage area may store data generated by the method for generating a coal distribution strategy. In addition, the memory 32 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 32 may optionally include a memory remotely located relative to the processor 31, and these remote memories may be connected to a device for executing the method for generating a coal distribution strategy via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The input device 33 may receive input user clicks, and generate signal inputs related to user settings and function controls of the method for generating a coal distribution strategy. The output device 34 may include a display device such as a display screen. When the one or more modules are stored in the memory 32 and are executed by the one or more processors 31, the method for generating a coal distribution strategy in any of the above-mentioned method embodiments is executed.

[0075] As needed, the above technical solutions can be combined to achieve the best technical effect.

[0076] The above are only the principles and preferred embodiments of the present application. It should be noted that, for those skilled in the art, on the basis of the principles of the present application, several other modifications can be made, which should also be considered as the scope of protection of the present application.

Claims

1. A method for generating a coal distribution strategy, characterized in that: include: Obtain coal quality data of multiple coal types in real time; In response to the distribution signal, multiple coal blending schemes are used to blend coal to obtain multiple pre-blended coal materials. Each coal blending scheme must meet the following requirements: the proportion of each coal type is any value between [0, 1], and the sum of the proportions of all coal types in the pre-blended coal materials obtained after the coal blending is completed is 1; The optimal solution among multiple coal blending schemes is searched based on a probabilistic evolutionary population optimization algorithm as a target coal blending strategy. In the probabilistic evolutionary population optimization algorithm, profit maximization is taken as the goal, and the coal quality requirement of pre-blended coal is taken as a constraint condition.

2. The method for generating a coal distribution strategy according to claim 1, characterized in that: The probabilistic evolutionary population optimization algorithm is used to search for the optimal solution among multiple coal blending schemes as the final coal blending strategy. In the probabilistic evolutionary population optimization algorithm, the goal is to maximize the benefits and the coal quality requirements of the pre-blended coal are used as constraints, including: Construct the objective function and calculate the fitness of each coal blending scheme: Penalty term = ∑ 指标数 a j ·(index j (w)-standard value j ) 2 ; S target is the theoretical sales price of pre-blended coal obtained from the coal blending plan, n represents the number of coal types involved in coal blending, S i is the theoretical sales price of the i-th coal type involved in coal blending, w i is the proportion of coal blending of the i-th coal type, λ is the regularization coefficient; a j is the penalty coefficient of the jth coal quality index of the pre-mixed coal; j (w) is the actual value of the jth indicator, the standard value j is the standard value of the jth indicator, and maximizing the benefit is to maximize the calculation result of the objective function F(w); Initialization constraints: Filter according to the set coal quality requirements and only retain the coal blending plans that meet the coal quality requirements.

3. The method for generating a coal distribution strategy according to claim 2, characterized in that: The method of searching for the optimal solution among multiple coal blending schemes based on the probabilistic evolutionary population optimization algorithm as the final coal blending strategy further includes: Initialization of the population: Screen all types of coal with inventory and generate a random set of initialization coal blending schemes, which are stored in different areas of the search space; Probabilistic evolutionary update: In each iteration cycle, the search probability distribution of different areas in the search space is determined based on the fitness calculation results in the previous iteration cycle; Probabilistic mutation: Based on the search probability distribution obtained in the probabilistic evolution update step, some coal blending schemes are selected for mutation to obtain the mutated coal blending scheme; Convergence and termination: When the average fitness of each coal blending scheme in the coal blending scheme collection obtained from multiple consecutive iteration cycles is less than the set threshold or the number of iterations reaches the upper limit, the iteration is stopped; Output the final coal blending strategy: Select the coal blending scheme that satisfies the constraints and maximizes the calculation result of the objective function F(w) as the final coal blending strategy, and output the proportion vector of each coal type w=(w1,w2,…,w n ).

4. The method for generating a coal distribution strategy according to claim 3, characterized in that: Also includes: Calculate the profit of the target pre-provisioned coal obtained by the final coal blending strategy: the profit is obtained by subtracting the sum of the current theoretical sales prices of each type of coal in the target pre-provisioned coal from the current actual sales price of the target pre-provisioned coal.

5. The method for generating a coal distribution strategy according to any one of claims 2 to 4, characterized in that: Initialization constraints: Filter according to the set coal quality requirements, and only retain the coal blending schemes that meet the coal quality requirements. The coal blending schemes are filtered in the following way: Set screening thresholds; a screening step, performing a difference calculation on the coal quality of the pre-blended coal obtained from the coal quality requirement and the coal blending plan, and filtering out the coal blending plan if the absolute value of the difference calculation result is greater than the screening threshold; The number of coal blending schemes obtained after screening is determined. If the number of coal blending schemes is less than a set number, the screening threshold is lowered and the process returns to the screening step.

6. The method for generating a coal distribution strategy according to any one of claims 2 to 4, characterized in that: In the probability evolution update, the search probability distribution of different areas in the search space is determined by the following probability update formula: p (t+1) (w) = p (t) (w) + β · [fitness (t) - average fitness (t)] · random adjustment term; Among them, p (t) (w) represents the search probability of the area where the coal blending scheme is located in the previous iteration cycle, p (t+1) (w) represents the search probability of the area where the coal blending scheme is located in the current iteration cycle, β represents the evolutionary step factor, fitness (t) represents the fitness calculation result of the coal blending scheme in the previous iteration cycle, and average fitness (t) represents the average fitness calculation result of all coal blending schemes searched in the previous iteration cycle; the random adjustment item is a randomly generated number between [-1,1].

7. The method for generating a coal distribution strategy according to any one of claims 2 to 4, characterized in that: In the probability variation step, the coal blending scheme is varied in the following ways: In' i =in i +γ rand(-∈,∈); Among them, γ is the variation factor, rand is the function used to generate random numbers, ∈ is the variation range constant, and w′ i Represents the proportion of coal blending after the mutation of the i-th coal type.

8. A computer-readable storage medium, characterized in that The storage medium stores program information, and after the computer reads the program information, it executes the steps of the method for generating a coal distribution strategy according to any one of claims 1 to 7.

9. A computer program product, characterized in that The method comprises a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the method implements the steps of the method for generating a coal distribution strategy as described in any one of claims 1 to 7.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method for generating a coal distribution strategy according to any one of claims 1 to 7.

Images