A coal railway outward transportation car request plan intelligent compilation method and system under multi-mineral coordination

Abstract

The application discloses a coal railway outward transportation car request plan intelligent compilation method and system under multi-mine cooperation, obtains railway outward transportation customer next-day demand plan and information such as monthly plan execution of each coal mine, daily production plan and capacity, current inventory condition, loading capacity and the like; then, taking the maximum outward transportation demand satisfaction rate and the balance of each mine production plan execution as optimization targets, considering the production and loading capacity of each mine, safety inventory and the like constraints, a next-day pre-day car request plan is formulated; then, in the daily execution process, the possible execution deviation of the daily plan is predicted according to the loading progress of each mine at regular time. Finally, according to the daily plan execution deviation prediction or the execution risk reported by each mine, taking the minimum daily plan adjustment amount as the optimization target, a daily car request plan adjustment optimization model is formulated, the daily car request plan is adjusted, and the intelligent compilation of the coal railway outward transportation car request plan under the multi-mine cooperation is realized.

Description

Technical Field

[0001] This invention belongs to the field of coal resource railway transportation scheduling technology, and particularly relates to an intelligent method and system for compiling coal railway transportation request plans under multi-mine collaboration. Background Technology

[0002] Currently, the railway transportation of coal is subject to a railway wagon request approval system. Coal sales companies need to prepare the train request plan for the next day in advance, and adjust the daily wagon request plan during the day's execution based on information on production delays at various coal mines, safety stock levels, and unforeseen circumstances such as loading equipment failures. This is a core business link that connects the coal production, transportation and sales chain and ensures the smooth operation of the coal transportation supply chain.

[0003] Under the traditional business model, the preparation and execution of coal resource transportation plans rely on manual offline coordination. The group's sales company needs to communicate with each mine by phone about the monthly production plan execution, daily production plan, inventory and loading line status, etc., based on the user's coal demand. Based on business experience, the company will break down and match the coal transportation volume of each mine and prepare and fill out the railway wagon request plan.

[0004] The existing techniques for preparing manual foreign transport plans have the following problems:

[0005] (1) Production data from various mines is scattered and lacks real-time communication, making it difficult to grasp the overall production situation of multiple mines in real time and to coordinate and allocate resources accordingly.

[0006] (2) The planning relies on the experience of business personnel, making it difficult to simultaneously take into account the supply and demand for external transportation and the balance of monthly production plan execution of each mine;

[0007] (3) Unable to effectively predict sudden events during the day, and after a sudden event occurs in a certain mine, it often relies on manual communication with other mines to determine whether production can be increased;

[0008] (4) The lack of an automated intelligent programming system means that the vehicle request plan is subject to frequent manual operation and modification. Summary of the Invention

[0009] Purpose of the invention: This invention provides an intelligent method and system for compiling coal railway transport request plans under multi-mine collaboration, aiming to solve the problems of existing technologies such as the difficulty of collaborative multi-mine data, reliance on manual experience in plan compilation, lack of prediction and automatic response mechanisms for intraday emergencies, and the lack of multi-mine overall planning and intraday dynamic adjustment capabilities in existing optimization studies.

[0010] Technical solution: The present invention provides an intelligent method for compiling coal railway transport request plans under multi-mine collaboration, comprising:

[0011] S1. Obtain multi-source data, including the next-day demand plan of railway transportation customers and information on various coal mines;

[0012] S2. Based on the multi-source data, construct a day-ahead vehicle request plan optimization model, and use a preset genetic algorithm to solve the day-ahead vehicle request plan optimization model to obtain the day-ahead vehicle request plan for the next day;

[0013] S3. During the execution of the vehicle request plan for the next day, predict the execution deviation of the vehicle request plan for the next day;

[0014] S4. Based on the execution deviation, construct an intraday vehicle request plan adjustment and optimization model, and use the preset genetic algorithm to solve the intraday vehicle request plan adjustment and optimization model to obtain an intraday vehicle request plan adjustment scheme.

[0015] Furthermore, the next-day demand plan for railway transport customers includes the total number of trains required by the aggregated transport customers, the number of trains required by each railway line, and information on the relationship between customers and railway lines; the information on each coal mine includes the next-day production plan, current inventory status and safety stock range, next-day coal mining plan, daily coal production capacity, and loading station loading capacity information for each coal mine.

[0016] Furthermore, the step of constructing a day-ahead train request planning optimization model based on the multi-source data includes: based on the multi-source data, taking the number of trains planned to be loaded on each railway line for the next day by each mine as the first decision variable, taking the maximization of the external transportation demand satisfaction rate and the balance of production plan execution of each mine as the first objective function, and taking the production and external transportation balance constraint of the first coal mine, the production safety stock constraint of the first coal mine, the loading capacity of the loading line of the first coal mine, and the train request transportation capacity constraint of the first railway as the first constraint conditions, and constructing a day-ahead train request planning optimization model;

[0017] The formula for the first objective function is:

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] in, This indicates that the vehicle manufacturer recently requested the development of an objective function for an optimization model. This indicates the daily demand for coal sales and transportation being met. This indicates the uniformity of monthly production plan execution across mines, and represents the mean square error of the monthly plan execution rate for each mine. These are the weighting coefficients; This indicates the number of trains required for outbound transportation. Indicates the actual number of trains transporting goods out of the country; This represents the monthly plan completion rate of the m-th coal mine. This represents the average monthly plan completion rate for all coal mines. This represents the planned monthly output of the m-th coal mine. This represents the monthly planned completion amount for the m-th coal mine. This indicates the total amount of coal loaded on a single train. This represents the number of trains scheduled to be loaded onto the m-th coal mine the following day.

[0024] The production and transportation balance constraint formula for the first coal mine is:

[0025] ;

[0026] ;

[0027] in, This represents the correlation matrix between coal mines and railway lines. This represents the planned number of coal trains to be dispatched along the l-th railway line;

[0028] The formula for the safety stock constraint of the first coal mine is:

[0029] ;

[0030] in, This represents the lower limit of the safety stock for the m-th coal mine. This represents the upper limit of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the production plan for the m-th coal mine for the next day;

[0031] The constraint formulas for the loading capacity of the first coal mine loading line and the transportation capacity of the first railway are as follows:

[0032] ;

[0033] ;

[0034] in, This represents the maximum transport capacity of the l-th railway line, i.e., the maximum number of trains that can be transported on that railway line per day. This indicates the loading capacity of a loading station, that is, the maximum number of trains that can be loaded at that loading station per day.

[0035] Furthermore, the process of using a preset genetic algorithm to solve the day-ahead vehicle request plan and compile an optimization model to obtain the day-ahead vehicle request plan for the next day includes the following solution process:

[0036] Step (1) only encodes real numbers for decision variables that are related to coal mines and railway lines;

[0037] Step (2) Generate the initial population within the first constraint condition;

[0038] Step (3) The reciprocal of the optimization objective of the optimization model of the current day vehicle request plan is used as the fitness function. The fitness value of each individual in the population is calculated and sorted from largest to smallest fitness value. Individuals with fitness values ​​ranked above a preset proportion are selected as elite individuals and directly retained to the next generation. The remaining individuals are selected as parents using the roulette wheel selection method.

[0039] Step (4) Perform crossover operation on the parent individuals according to the preset crossover probability, and perform mutation operation on the individuals according to the preset mutation probability. The mutation operation is performed according to the preset mutation range.

[0040] After step (5) crossover mutation, individual constraint correction and overall constraint feasibility check are performed. Elite individuals are merged with crossover mutated individuals to form a new generation population. Steps (3) to (5) are repeated until the preset maximum number of iterations is reached.

[0041] Step (6) Select the individual with the highest fitness value in the last generation population as the optimal solution and generate the vehicle request plan for the next day.

[0042] Furthermore, the predicted deviation in the vehicle request plan for the following day includes:

[0043] Based on the inventory status data, completed loading data, remaining operation time data, and loading equipment malfunction data of each mine, the feasibility of each mine's coal transportation plan is predicted, and the existence of execution risks is assessed. The daily plan execution deviation of the m-th coal mine is also considered. If so, then the m-th coal mine is at risk of under-payment;

[0044] Daily plan execution deviation The formula is:

[0045] ;

[0046] ;

[0047] ;

[0048] in, This indicates the total amount of coal that can be loaded onto trucks at the baseline of the m-th coal mine on that day. This represents the cumulative amount of coal that has been loaded onto trucks at the m-th coal mine. This represents the average loading rate of the m-th coal mine over the past 2 hours. This represents the remaining effective loading time for the m-th coal mine on that day. This represents the current available loading inventory of the m-th coal mine. This represents the current inventory of the m-th coal mine. This represents the lower limit of the safety stock for the m-th coal mine. This indicates the total amount of coal loaded on a single train. This represents the daily departure schedule for the m-th coal mine. This represents the daily plan execution deviation of the m-th coal mine.

[0049] Furthermore, the step of constructing an intraday vehicle request plan adjustment and optimization model based on the execution deviation includes:

[0050] Once the risk of execution is determined and confirmed by the mine, the daily plan adjustment is minimized as the second objective function, and the production and transportation balance constraints of the second coal mine, the production safety stock constraints of the second coal mine, the loading capacity of the second coal mine loading line and the transportation capacity of the second railway, and the fixed loading volume constraints of the risky coal mine are used as the second constraints to construct an intraday train request plan adjustment optimization model.

[0051] The formula for the second objective function is:

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] in, This indicates the objective function of the intraday car request plan adjustment and optimization model. This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the average daily adjustment to the outbound transport plan for each railway line. This represents the average daily adjustment to the loading and transportation plan for each coal mine. , , Indicates the weighting coefficient; This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the daily planned train volume for the l-th railway line. This represents the daily loading and transportation plan number for the m-th coal mine;

[0057] The formula for the safety stock constraint in the second coal mine is:

[0058] ;

[0059] ;

[0060] in, This represents the lower limit of the safety stock for the m-th coal mine. This represents the upper limit of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the remaining amount of the daily production plan to be completed for the m-th coal mine. This indicates the status of loading at the m-th coal mine;

[0061] The fixed loading volume constraint for high-risk coal mines means that for coal mines with risky loading warnings and after confirmation, their daily loading volume shall not exceed the loading volume reported by the coal mine. The formula is as follows:

[0062] ;

[0063] in, This represents the amount of coal that has been loaded into wagons at the m-th coal mine;

[0064] The formulas constraining the loading capacity of the second coal mine loading line and the transportation capacity of the second railway are as follows:

[0065] ;

[0066] ;

[0067] in, This represents the maximum transport capacity of the l-th railway line, i.e., the maximum number of trains that can be transported on that railway line per day. This represents the loading capacity of the m-th coal mine, i.e., the maximum number of trains that can be loaded per day at that loading station. This indicates the total effective duration of loading or rail transport. Indicates the remaining valid time for the day;

[0068] The production and transportation balance constraint formula for the second coal mine is:

[0069] ;

[0070] ;

[0071] in, This indicates the total number of coal shipments that have been completed on that day.

[0072] Furthermore, the step of using the preset genetic algorithm to solve the daily train request plan adjustment optimization model to obtain the daily train request plan adjustment scheme includes: taking the number of trains to be loaded on each railway line after the daily adjustment of each mine as the second decision variable, taking the reciprocal of the second objective function as the fitness function, and solving the problem according to the solution process of the preset genetic algorithm within the range of the second constraint conditions to generate the daily train request plan adjustment scheme.

[0073] The present invention discloses an intelligent system for generating coal railway transport request plans under multi-mine collaboration, comprising:

[0074] The data acquisition module is used to acquire multi-source data, including the next-day demand plans of railway transportation customers and information on various coal mines;

[0075] The planning and optimization module is used to construct a day-ahead vehicle request planning and optimization model based on the multi-source data, and to solve the day-ahead vehicle request planning and optimization model using a preset genetic algorithm to obtain the day-ahead vehicle request plan for the next day.

[0076] The execution deviation prediction module is used to predict the execution deviation of the next day's vehicle request plan during the execution process.

[0077] The planning adjustment module is used to construct an intraday vehicle request plan adjustment optimization model based on the execution deviation, and use the preset genetic algorithm to solve the intraday vehicle request plan adjustment optimization model to obtain an intraday vehicle request plan adjustment scheme.

[0078] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0079] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0080] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0081] 1. By constructing a multi-mine collaborative day-ahead vehicle request planning optimization model, the optimization objectives are simultaneously taken as maximizing the satisfaction rate of external transportation demand and the balance of monthly production plan execution of each mine, thus solving the technical problem that manual planning in the existing technology cannot take into account both external transportation demand supply and production balance.

[0082] 2. During the daily execution process, the daily plan execution deviation is predicted in a timely manner based on the loading progress of each mine, and an intraday vehicle request plan adjustment and optimization model is constructed to realize proactive prediction of emergencies and automatic adjustment of the plan, overcoming the shortcomings of existing technologies that rely on manual communication with other mines regarding the production increase situation after an emergency occurs and have a delayed response.

[0083] 3. With the goal of minimizing daily plan adjustments, different weight coefficients are set for three levels: total demand, railway lines, and coal mines. Under the premise of ensuring that the total daily external transportation demand is met, the plan is fine-tuned to reduce the interference of plan adjustments on normal production organization.

[0084] 4. To automate the entire process of coal railway transport request planning under multi-mine collaboration, from day-ahead preparation to intraday adjustment, thereby reducing the frequency of manual operations and improving the efficiency and accuracy of planning. Attached Figure Description

[0085] Figure 1 A flowchart for an intelligent method for generating coal railway transport request plans under multi-mine collaboration.

[0086] Figure 2 A detailed flowchart of the intelligent compilation method for coal railway transport requesting plans under multi-mine collaboration.

[0087] Figure 3 A simplified schematic diagram of the railway transportation route for coal resources of a certain energy group. Detailed Implementation

[0088] like Figure 1 As shown, the intelligent method for compiling coal railway transport request plans under multi-mine collaboration described in this invention includes:

[0089] S1. Obtain multi-source data, including the next-day demand plan of railway transportation customers and information on various coal mines;

[0090] Specifically, the next-day demand plan for railway transport customers is obtained, including the total number of trains required by the aggregated transport customers, the number of trains required by each railway line, and information on the relationship between customers and railway lines. Information on each coal mine is obtained, including the next-day production plan, current inventory status and safety stock range, next-day coal mining plan, daily coal production capacity, and loading station loading capacity information.

[0091] S2. Based on the multi-source data, construct a day-ahead vehicle request plan optimization model, and use a preset genetic algorithm to solve the day-ahead vehicle request plan optimization model to obtain the day-ahead vehicle request plan for the next day.

[0092] Specifically, based on the multi-source data, the number of trains planned to be loaded on each railway line for the next day by each mine is taken as the first decision variable, the maximum satisfaction rate of transportation demand and the balance of production plan execution of each mine are taken as the first objective function, and the production and transportation balance constraints of the first coal mine, the production safety stock constraints of the first coal mine, the loading capacity of the loading line of the first coal mine and the transportation capacity of the first railway are taken as the first constraints, and a day-ahead train requesting plan preparation optimization model is constructed.

[0093] The formula for the first objective function is:

[0094] ;

[0095] ;

[0096] ;

[0097] ;

[0098] ;

[0099] in, This indicates that the vehicle manufacturer recently requested the development of an objective function for an optimization model. This indicates the daily demand for coal sales and transportation being met. This indicates the uniformity of monthly production plan execution across mines, and represents the mean square error of the monthly plan execution rate for each mine. This is a weighting coefficient, representing the group's emphasis on the balance of plan execution across mines. Sales groups generally pay more attention to the total daily coal sales volume, so 0.1 can be used here. This indicates the number of trains required for outbound transportation. Indicates the actual number of trains transporting goods out of the country; This represents the monthly plan completion rate of the m-th coal mine. This represents the average monthly plan completion rate for all coal mines. This represents the planned monthly output of the m-th coal mine. This represents the monthly planned completion amount for the m-th coal mine. This indicates the total amount of coal loaded on a single train. This represents the number of trains scheduled to be loaded onto the m-th coal mine the following day.

[0100] The production and transportation balance constraint formula for the first coal mine is:

[0101] ;

[0102] ;

[0103] in, This represents the correlation matrix between coal mines and railway lines. This represents the planned number of coal trains to be dispatched along the l-th railway line;

[0104] The formula for the safety stock constraint of the first coal mine is:

[0105] ;

[0106] in, This represents the lower limit of the safety stock for the m-th coal mine. This represents the upper limit of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the production plan for the m-th coal mine for the next day;

[0107] The constraint formulas for the loading capacity of the first coal mine loading line and the transportation capacity of the first railway are as follows:

[0108] ;

[0109] ;

[0110] in, This represents the maximum transport capacity of the l-th railway line, i.e., the maximum number of trains that can be transported on that railway line per day. This indicates the loading capacity of the loading station, that is, the maximum number of trains that can be loaded at the loading station per day;

[0111] Since the mean square calculation in the objective function is a quadratic function, it cannot be solved using linear programming. Therefore, this invention uses a genetic algorithm to solve the optimization model of the current vehicle request plan. The hyperparameters of the genetic algorithm are set as follows: maximum number of iterations 2000, population size 100, crossover probability 0.8, mutation probability 0.2, elite retention ratio 10%, and mutation range ±1 column. The solution steps are as follows:

[0112] Step (1) Encode the first decision variable (shipment plan of each mine on each railway line) using real number encoding. Only the decision variables that are related to the coal mine and the railway line are encoded, and those that are not related are fixed as 0.

[0113] Step (2) Generate the initial population. The population is generated within the first constraint condition to avoid too many invalid solutions.

[0114] Step (3) The reciprocal of the first objective function of the optimization model of the current day car request plan is used as the fitness function. The fitness value of each individual in the population is calculated. The individuals are sorted from largest to smallest fitness value. The top 10% of individuals with the largest fitness value are selected as elite individuals and directly retained to the next generation. The remaining individuals are selected as parents by roulette wheel selection.

[0115] Step (4) Perform crossover operation on the parent individuals with a crossover probability of 0.8 and perform mutation operation on the individuals with a mutation probability of 0.2. The mutation operation is performed with a mutation range of ±1 column, that is, each decision variable is added or removed by 1 column on the original basis.

[0116] After step (5) crossover mutation, individual constraint correction and overall constraint feasibility check are performed. Elite individuals are merged with crossover mutated individuals to form a new generation population. Steps (3) to (5) are repeated until the maximum number of iterations of 2000 is reached.

[0117] Step (6) Select the individual with the highest fitness value in the last generation population as the optimal solution and generate the vehicle request plan for the next day.

[0118] The next day's train request plan is sent to each mine for confirmation, and feedback is obtained from each mine. If each mine makes corrections, the corrected data from that mine is used as a fixed constraint, and the previous day's train request plan is called back to compile an optimization model for solution. If there are no corrections, the next day's train request plan is sent to the train request system and executed.

[0119] S3. During the execution of the next day's truck request plan, predict the execution deviation of the daily plan in a timely manner based on the loading progress of each mine;

[0120] Specifically, based on real-time inventory status data, completed loading data, remaining operation time data, and loading equipment malfunction status data from each mine, the feasibility of each mine's coal transportation plan is predicted, and the existence of execution risks is assessed. The execution risk prediction model uses the difference between the baseline total loading volume and the planned loading volume for the current day to determine the execution risk. When the daily plan execution deviation of the m-th coal mine... Therefore, the m-th coal mine faces the risk of under-payment. The larger the absolute value, the higher the risk of underpayment;

[0121] Daily plan execution deviation of the m-th coal mine The formula is:

[0122] ;

[0123] ;

[0124] ;

[0125] in, This indicates the total amount of coal that can be loaded onto trucks at the baseline of the m-th coal mine on that day. This represents the cumulative amount of coal that has been loaded onto trucks at the m-th coal mine. This represents the average loading rate of the m-th coal mine over the past 2 hours. This represents the remaining effective loading time for the m-th coal mine on that day. This represents the current available loading inventory of the m-th coal mine. This represents the current inventory of the m-th coal mine. This represents the lower limit of the safety stock for the m-th coal mine. This indicates the total amount of coal loaded on a single train. This represents the daily departure schedule for the m-th coal mine. This represents the daily plan execution deviation for the m-th coal mine;

[0126] Send coal mines with execution risks to the mine for risk confirmation. If malfunctions such as loading equipment are detected, directly confirm with the mine whether there is a risk that the loading plan cannot be completed.

[0127] Receive feedback from each mine. If each mine reports no risk of plan failure or no abnormalities are found during execution, the daily truck request plan will be executed according to the previous day's plan. If a mine reports execution risk, the revised data from that mine will be used as a fixed loading volume constraint, and the process will proceed to S4.

[0128] S4. Based on the execution deviation, construct an intraday vehicle request plan adjustment and optimization model, and use the preset genetic algorithm to solve the intraday vehicle request plan adjustment and optimization model to obtain an intraday vehicle request plan adjustment scheme;

[0129] Specifically, once S3 determines that there is an execution risk and the mine confirms it, the optimization objective is to minimize the daily plan adjustment. Considering constraints such as the production and loading capacity of each mine, safety stock, completed loading status, and remaining effective time, an intraday vehicle request plan adjustment optimization model is constructed.

[0130] The intraday train request plan adjustment optimization model takes the number of trains to be loaded on each railway line after the daily adjustment of each mine as the second decision variable, the minimization of the daily plan adjustment amount as the objective function, and the production and transportation balance constraints of the second coal mine, the production safety stock constraints of the second coal mine, the loading capacity of the second coal mine loading line and the train request transportation capacity of the second railway, and the fixed loading quantity constraints of the risk coal mine as the second constraints.

[0131] The second objective function is set considering three levels: total demand adjustment, railway line plan adjustment, and daily mine plan adjustment. Different weighting coefficients are set for different levels to prioritize ensuring the satisfaction of daily external transportation demand. The second objective function is as follows:

[0132] ;

[0133] ;

[0134] ;

[0135] ;

[0136] in, This indicates the objective function of the intraday car request plan adjustment and optimization model. This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the average daily adjustment to the outbound transport plan for each railway line. This represents the average daily adjustment to the loading and transportation plan for each coal mine. , , This represents the weighting coefficient, with values ​​of 0.8, 0.15, and 0.05. When adjustments are needed, changes to the total outbound volume should be avoided first. This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the daily planned train volume for the l-th railway line. This represents the daily loading and transportation plan number for the m-th coal mine;

[0137] The formula for the safety stock constraint in the second coal mine is:

[0138] ;

[0139] ;

[0140] in, This represents the lower limit of the safety stock for the m-th coal mine. This represents the upper limit of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the remaining amount of the daily production plan to be completed for the m-th coal mine. This indicates the status of loading at the m-th coal mine;

[0141] The fixed loading volume constraint for high-risk coal mines means that for coal mines with risky loading warnings and after confirmation, their daily loading volume shall not exceed the loading volume reported by the coal mine. The formula is as follows:

[0142] ;

[0143] in, This represents the amount of coal that has been loaded into wagons at the m-th coal mine;

[0144] The formulas constraining the loading capacity of the second coal mine loading line and the transportation capacity of the second railway are as follows:

[0145] ;

[0146] ;

[0147] in, This represents the maximum transport capacity of the l-th railway line, i.e., the maximum number of trains that can be transported on that railway line per day. This represents the loading capacity of the m-th coal mine, i.e., the maximum number of trains that can be loaded per day at that loading station. This indicates the total effective duration of loading or rail transport. Indicates the remaining valid time for the day;

[0148] The production and transportation balance constraint formula for the second coal mine is:

[0149] ;

[0150] ;

[0151] in, This indicates the total number of coal sales and transportation plans completed on that day.

[0152] The same genetic algorithm as S2 was used to solve the intraday vehicle request plan adjustment and optimization model. The key hyperparameter settings were the same as those of S2: maximum number of iterations 2000, population size 100, crossover probability 0.8, mutation probability 0.2, elite retention ratio 10%, and mutation range ±1 column.

[0153] The solution steps are as follows:

[0154] Step (1) The second decision variable (the number of trains to be loaded on each railway line after daily adjustment) is encoded using real number encoding. Only coal mines and railway lines that are related are encoded, while those that are not related are fixed at 0.

[0155] Step (2) generates an initial population within the second constraint range to avoid too many invalid solutions;

[0156] Step (3) The reciprocal of the second objective function of the intraday car request plan adjustment and optimization model is used as the fitness function. The fitness value of each individual in the population is calculated. The individuals are sorted from largest to smallest fitness value. The top 10% of individuals with the largest fitness value are selected as elite individuals and directly retained to the next generation. The remaining individuals are selected as parents using the roulette wheel selection method.

[0157] Step (4) Perform crossover operation on the parent individuals with a crossover probability of 0.8 and perform mutation operation on the individuals with a mutation probability of 0.2. The mutation operation is performed with a mutation range of ±1 column, that is, each decision variable is added or removed by 1 column on the original basis.

[0158] After step (5) crossover mutation, individual constraint correction and overall constraint feasibility check are performed. Elite individuals are merged with crossover mutated individuals to form a new generation population. Steps (3) to (5) are repeated until the maximum number of iterations of 2000 is reached.

[0159] Step (6) Select the individual with the highest fitness value in the last generation population as the optimal solution, generate the intraday vehicle request plan adjustment scheme, and send it to the relevant coal mines that have made adjustments for confirmation;

[0160] If each mine has no objection to the daily train request plan adjustment scheme, the adjusted scheme shall be implemented. If a mine has any objection, the revised data of that mine shall be used as a fixed constraint, and the process shall return to this step to call the daily train request plan adjustment optimization model again to solve the problem and generate a new daily train request plan adjustment scheme.

[0161] The present invention discloses an intelligent system for generating coal railway transport request plans under multi-mine collaboration, comprising:

[0162] The data acquisition module is used to acquire multi-source data, including the next-day demand plans of railway transportation customers and information on various coal mines;

[0163] The planning and optimization module is used to construct a day-ahead vehicle request planning and optimization model based on the multi-source data, and to solve the day-ahead vehicle request planning and optimization model using a preset genetic algorithm to obtain the day-ahead vehicle request plan for the next day.

[0164] The execution deviation prediction module is used to predict the execution deviation of the next day's vehicle request plan during the execution process.

[0165] The planning adjustment module is used to construct an intraday vehicle request plan adjustment optimization model based on the execution deviation, and use the preset genetic algorithm to solve the intraday vehicle request plan adjustment optimization model to obtain an intraday vehicle request plan adjustment scheme.

[0166] The computer device of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above method.

[0167] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0168] Regarding the above technical solution in this embodiment, the following is a detailed explanation. Figures 2 to 3 This will detail the experimental process of applying the technical solution to a specific coal railway transportation scheduling experiment of a certain energy group, as well as the technical effects of the solution.

[0169] The experiment selected a simplified coal rail transport scenario from an energy group, involving 5 coal mines (Mine A, Mine B, Mine C, Mine D, and Mine E), 2 rail transport lines (Railway Line 1 and Railway Line 2), and 3 transport customers. The relationships between the coal mines, railway lines, and customers are as follows: Figure 3 As shown.

[0170] First, we obtain the next-day demand plan for railway transport customers, including the total number of trains required by aggregated transport customers, the number of trains required for each railway line, and information on the relationship between customers and railway lines. Information on each mine includes the next-day production plan, current inventory status and safety stock range, next-day coal mining plan, daily coal production capacity, and loading station loading capacity information for each coal mine. The total number of trains required by aggregated transport customers and the number of trains required for each railway line are shown in Table 1. The maximum daily capacity for both Line 1 and Line 2 is 10 trains, with each train carrying 3750 tons.

[0171] Table 1: Total Number of Trains Demand by Customers for Outbound Transportation Aggregation and Number of Trains Demand on Each Railway Line

[0172]

[0173] The daily information for each mine is shown in Table 2.

[0174] Table 2 Information on each mine

[0175]

[0176] Based on the above data, a day-ahead vehicle request planning optimization model is constructed. The objective function is set to maximize the fulfillment rate of external transportation demand and ensure balanced execution of production plans across mines. The formula is:

[0177] ;

[0178] ;

[0179] ;

[0180] ;

[0181] ;

[0182] In the formula This indicates that the vehicle manufacturer recently requested the development of an objective function for an optimization model. This indicates the daily demand for coal sales and transportation being met. This represents the uniformity of monthly production plan execution across mines, expressed as the mean square error of the monthly plan execution rate for each mine. This is a weighting coefficient, representing the group's emphasis on the balanced execution of plans across mines. The sales group generally focuses more on the total daily coal sales volume; here, a value of 0.1 can be used. This indicates the number of trains required for outbound transportation. This indicates the actual number of trains transported out of the country. This represents the monthly plan completion rate of the m-th coal mine. This represents the average monthly plan completion rate for all coal mines. Let represent the monthly planned quantity and the monthly planned completed quantity of the m-th coal mine, respectively. This indicates the total amount of coal loaded on a single train. This represents the number of trains scheduled to be loaded onto the m-th coal mine the following day.

[0183] The constraints for constructing the day-ahead vehicle request plan optimization model include the coal mine's production and transportation balance constraints, and the formula is as follows:

[0184] ;

[0185] ;

[0186] In the formula, This represents the correlation matrix between coal mines and railway lines. This indicates the planned number of coal trains to be dispatched from the l-th railway line.

[0187] Coal mine production safety stock constraints:

[0188] ;

[0189] In the formula, , This represents the upper and lower limits of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the production plan for the m-th coal mine for the next day.

[0190] Constraints on coal mine loading line loading capacity and railway wagon requesting capacity:

[0191] ;

[0192] ;

[0193] In the formula, This represents the maximum transport capacity of the l-th railway line (i.e., the maximum number of trains that can be transported per day on that railway line). This indicates the loading capacity of a loading station, that is, the maximum number of trains that can be loaded at that loading station per day.

[0194] The above method solves the model. Since the mean square in the objective function is a quadratic function, it cannot be solved using linear programming. Therefore, a genetic algorithm is used to solve it, and the steps are as follows:

[0195] Step (1) Use real number coding to encode the decision variables (shipment plans of each mine on each railway line). Only the decision variables that are related to the coal mine and the railway line are coded, and those that are not related are fixed as 0.

[0196] Step (2) generates the initial population. The population is generated within the constraints to avoid too many invalid solutions.

[0197] Step (3) The reciprocal of the optimization objective of the optimization model of the current day vehicle request plan is used as the fitness function. The fitness value of each individual in the population is calculated. The individuals are sorted from largest to smallest fitness value. The top 10% of individuals with the largest fitness value are selected as elite individuals and directly retained to the next generation. The remaining individuals are selected as parents by roulette wheel selection.

[0198] Step (4) Perform crossover operation on the parent individuals with a crossover probability of 0.8 and perform mutation operation on the individuals with a mutation probability of 0.2. The mutation operation is performed with a mutation range of ±1 column, that is, each decision variable is added or removed by 1 column on the original basis.

[0199] After step (5) crossover mutation, individual constraint correction and overall constraint feasibility check are performed. Elite individuals are merged with crossover mutated individuals to form a new generation population. Steps (3) to (5) are repeated until the maximum number of iterations of 2000 is reached.

[0200] Step (6) Select the individual with the highest fitness value in the last generation population as the optimal solution and generate the vehicle request plan for the next day.

[0201] The key hyperparameter settings for the genetic algorithm are shown in Table 3.

[0202] Table 3 Hyperparameters of the genetic algorithm

[0203]

[0204] The optimized daily vehicle request plans for each mine are shown in Table 4.

[0205] Table 4. Optimized Daily Mine Truck Request Plan

[0206]

[0207] Then, the optimized daily vehicle request plan allocation scheme is sent to each mine for confirmation. If no correction is made by each mine, the plan is sent to the vehicle request system to begin execution within the day.

[0208] During the day's execution, real-time monitoring of inventory and loading progress data at each mine is conducted, and risk warnings are issued periodically to determine if there are any loading risks.

[0209] The execution data of each mine was monitored when the daily plan was half completed, as shown in Table 5.

[0210] Table 5 shows the monitoring data for each mine.

[0211]

[0212] Based on the access data on the inventory status of each mine, the data on completed loading, the remaining operational market data, and the data on the failure status of loading equipment, the feasibility of each mine's coal transportation plan can be predicted.

[0213] The formula for assessing the risk of loading plan execution is shown below. Assessing the execution risk of a coal mine, when This indicates the existence of a risk of underpayment; the larger the absolute value, the higher the risk of underpayment. The calculation is shown in the following formula:

[0214] ;

[0215] ;

[0216] ;

[0217] In the formula, This indicates that the total amount of coal that can be loaded at the current benchmark level is [value missing]. This indicates the current cumulative loading volume. This indicates the average loading rate over the past 2 hours. This indicates the remaining valid loading time for the day. This indicates the currently available loading inventory. Indicates current inventory. This indicates the day-ahead departure schedule for the m-th coal mine, which was issued yesterday.

[0218] Calculations revealed that for coal mine A This carries execution risks. The coal mine confirmed the order, and reported that the loading line underwent temporary maintenance that day, and it is estimated that it can only meet the capacity for one more train shipment in the future.

[0219] Based on the above intraday situation, an intraday train request plan adjustment and optimization model is constructed. The optimization objective is to minimize the daily plan adjustment. The optimization objective is constructed considering three levels: total demand adjustment, plan adjustment for each railway line, and daily plan adjustment for each mine. Different weight coefficients are set for different levels to prioritize ensuring the satisfaction of the total daily transportation demand.

[0220] ;

[0221] In the formula, This represents the objective function of the intraday vehicle request plan adjustment and optimization model. This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the average daily adjustment to the outbound transport plan for each railway line. This represents the average daily adjustment to the loading and transportation plan for each coal mine. This represents the weighting coefficient, with values ​​of 0.8, 0.15, and 0.05. When adjustments are needed, changes in the total volume of outbound shipments should be avoided first.

[0222] ;

[0223] ;

[0224] ;

[0225] In the formula, This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the daily planned train volume for the l-th railway line. This represents the daily loading and transportation plan number for the m-th coal mine.

[0226] The constraints of the intraday vehicle request plan adjustment and optimization model are the same as those of the previous day, including production and transportation balance constraints and coal mine production safety inventory constraints.

[0227] Coal mine production safety stock constraints:

[0228] ;

[0229] ;

[0230] In the formula, , This represents the upper and lower limits of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the remaining amount of the daily production plan to be completed for the m-th coal mine. This indicates the status of loading at the m-th coal mine.

[0231] For coal mines that have issued risk warnings for loading and have confirmed the risk, their daily loading volume shall not exceed the loading volume reported by the coal mine.

[0232] ;

[0233] In the formula, This represents the amount of coal that has been loaded into wagons at the m-th coal mine.

[0234] Constraints on coal mine loading line loading capacity and railway wagon requesting capacity:

[0235] ;

[0236] ;

[0237] In the formula, This represents the maximum transport capacity of the l-th railway line (i.e., the maximum number of trains that can be transported per day on that railway line). This indicates the loading capacity of a loading station, that is, the maximum number of trains that can be loaded at that loading station per day. This indicates the total effective time for loading, transporting, or rail freight. This indicates the remaining valid time for the day.

[0238] The production and transportation balance constraint of a coal mine is shown in the following formula:

[0239] ;

[0240] ;

[0241] In the formula, This indicates the total number of trains for which the coal sales and transportation plan has been completed on that day.

[0242] Table 6. Adjusted Vehicle Request Plan

[0243]

[0244] Then, the same genetic algorithm was used to optimize the intraday vehicle request adjustment model, generating intraday vehicle request plan adjustment information. The adjusted vehicle request plan is shown in Table 6.

Claims

1. A method for intelligently compiling coal railway freight requisition plans under multi-mine collaborative operation, characterized in that, include: S1. Obtain multi-source data, including the next-day demand plan of railway transportation customers and information on various coal mines; S2. Based on the multi-source data, construct a day-ahead vehicle request plan optimization model, and use a preset genetic algorithm to solve the day-ahead vehicle request plan optimization model to obtain the day-ahead vehicle request plan for the next day; S3. During the execution of the vehicle request plan for the next day, predict the execution deviation of the vehicle request plan for the next day; S4. Based on the execution deviation, construct an intraday vehicle request plan adjustment and optimization model, and use the preset genetic algorithm to solve the intraday vehicle request plan adjustment and optimization model to obtain an intraday vehicle request plan adjustment scheme.

2. The intelligent method for compiling coal railway transport request plans under multi-mine collaboration as described in claim 1, characterized in that, The next-day demand plan for railway transport customers includes the total number of trains required by the aggregated transport customers, the number of trains required by each railway line, and information on the relationship between customers and railway lines; the information on each coal mine includes the next-day production plan, current inventory status and safety stock range, next-day coal mining plan, daily coal production capacity, and loading station loading capacity information for each coal mine.

3. The intelligent method for compiling coal railway transport request plans under multi-mine collaboration as described in claim 2, characterized in that, The step of constructing a day-ahead train request planning optimization model based on the multi-source data includes: based on the multi-source data, taking the number of trains planned to be loaded on each railway line for the next day by each mine as the first decision variable, taking the maximization of the external transportation demand satisfaction rate and the balance of production plan execution of each mine as the first objective function, and taking the production and external transportation balance constraint of the first coal mine, the production safety stock constraint of the first coal mine, the loading capacity of the first coal mine loading line and the train request transportation capacity constraint of the first railway as the first constraint conditions, and constructing a day-ahead train request planning optimization model. The formula for the first objective function is: ； ； ； ； ； in, This indicates that the vehicle manufacturer recently requested the development of an objective function for an optimization model. This indicates the daily demand for coal sales and transportation being met. This indicates the uniformity of monthly production plan execution across mines, and represents the mean square error of the monthly plan execution rate for each mine. These are the weighting coefficients; This indicates the number of trains required for outbound transportation. Indicates the actual number of trains transporting goods out of the country; This represents the monthly plan completion rate of the m-th coal mine. This represents the average monthly plan completion rate for all coal mines. This represents the planned monthly output of the m-th coal mine. This represents the monthly planned completion amount for the m-th coal mine. This indicates the total amount of coal loaded on a single train. This represents the number of trains scheduled to be loaded onto the m-th coal mine the following day. The production and transportation balance constraint formula for the first coal mine is: ； ； in, This represents the correlation matrix between coal mines and railway lines. This represents the planned number of coal trains to be dispatched along the l-th railway line; The formula for the safety stock constraint of the first coal mine is: ； in, This represents the lower limit of the safety stock for the m-th coal mine. This represents the upper limit of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the production plan for the m-th coal mine for the next day; The constraint formulas for the loading capacity of the first coal mine loading line and the transportation capacity of the first railway are as follows: ； ； in, This represents the maximum transport capacity of the l-th railway line, i.e., the maximum number of trains that can be transported on that railway line per day. This indicates the loading capacity of a loading station, that is, the maximum number of trains that can be loaded at that loading station per day.

4. The intelligent method for compiling coal railway transport request plans under multi-mine collaboration as described in claim 3, characterized in that, The process of using a preset genetic algorithm to solve the day-ahead vehicle request plan and compile an optimization model to obtain the day-ahead vehicle request plan for the next day includes the following solution process: Step (1) only encodes real numbers for decision variables that are related to coal mines and railway lines; Step (2) Generate the initial population within the first constraint condition; Step (3) The reciprocal of the optimization objective of the optimization model of the current day vehicle request plan is used as the fitness function. The fitness value of each individual in the population is calculated and sorted from largest to smallest fitness value. Individuals with fitness values ​​ranked above a preset proportion are selected as elite individuals and directly retained to the next generation. The remaining individuals are selected as parents using the roulette wheel selection method. Step (4) Perform crossover operation on the parent individuals according to the preset crossover probability, and perform mutation operation on the individuals according to the preset mutation probability. The mutation operation is performed according to the preset mutation range. After step (5) crossover mutation, individual constraint correction and overall constraint feasibility check are performed. Elite individuals are merged with crossover mutated individuals to form a new generation population. Steps (3) to (5) are repeated until the preset maximum number of iterations is reached. Step (6) Select the individual with the highest fitness value in the last generation population as the optimal solution and generate the vehicle request plan for the next day.

5. The intelligent method for compiling coal railway transport request plans under multi-mine collaboration as described in claim 4, characterized in that, The predicted deviation from the next day's pre-arranged vehicle request plan includes: Based on the inventory status data, completed loading data, remaining operation time data, and loading equipment malfunction data of each mine, the feasibility of each mine's coal transportation plan is predicted, and the existence of execution risks is assessed. The daily plan execution deviation of the m-th coal mine is also considered. If so, then the m-th coal mine is at risk of under-payment; Daily plan execution deviation The formula is: ； ； ； in, This indicates the total amount of coal that can be loaded onto trucks at the baseline of the m-th coal mine on that day. This represents the cumulative amount of coal that has been loaded onto trucks at the m-th coal mine. This represents the average loading rate of the m-th coal mine over the past 2 hours. This represents the remaining effective loading time for the m-th coal mine on that day. This represents the current available loading inventory of the m-th coal mine. This represents the current inventory of the m-th coal mine. This represents the lower limit of the safety stock for the m-th coal mine. This indicates the total amount of coal loaded on a single train. This represents the daily departure schedule for the m-th coal mine. This represents the daily plan execution deviation of the m-th coal mine.

6. The intelligent method for compiling coal railway transport request plans under multi-mine collaboration as described in claim 5, characterized in that, The step of constructing an intraday vehicle request plan adjustment and optimization model based on the execution deviation includes: Once the risk of execution is determined and confirmed by the mine, the daily plan adjustment is minimized as the second objective function, and the production and transportation balance constraints of the second coal mine, the production safety stock constraints of the second coal mine, the loading capacity of the second coal mine loading line and the transportation capacity of the second railway, and the fixed loading volume constraints of the risky coal mine are used as the second constraints to construct an intraday train request plan adjustment optimization model. The formula for the second objective function is: ； ； ； ； in, This indicates the objective function of the intraday car request plan adjustment and optimization model. This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the average daily adjustment to the outbound transport plan for each railway line. This represents the average daily adjustment to the loading and transportation plan for each coal mine. , , Indicates the weighting coefficient; This indicates the adjustment to the total planned volume of coal sales and transportation before the due date. This indicates the daily planned train volume for the l-th railway line. This represents the daily loading and transportation plan number for the m-th coal mine; The formula for the safety stock constraint in the second coal mine is: ； ； in, This represents the lower limit of the safety stock for the m-th coal mine. This represents the upper limit of the safety stock of the m-th coal mine. This represents the current inventory status of the m-th coal mine. This represents the remaining amount of the daily production plan to be completed for the m-th coal mine. This indicates the status of loading at the m-th coal mine; The fixed loading volume constraint for high-risk coal mines means that for coal mines with risky loading warnings and after confirmation, their daily loading volume shall not exceed the loading volume reported by the coal mine. The formula is as follows: ； in, This represents the amount of coal that has been loaded into wagons at the m-th coal mine; The formulas constraining the loading capacity of the second coal mine loading line and the transportation capacity of the second railway are as follows: ； ； in, This represents the maximum transport capacity of the l-th railway line, i.e., the maximum number of trains that can be transported on that railway line per day. This represents the loading capacity of the m-th coal mine, i.e., the maximum number of trains that can be loaded per day at that loading station. This indicates the total effective duration of loading or rail transport. Indicates the remaining valid time for the day; The production and transportation balance constraint formula for the second coal mine is: ； ； in, This indicates the total number of trains for which the coal sales and transportation plan has been completed on that day.

7. The intelligent method for compiling coal railway transport request plans under multi-mine collaboration as described in claim 6, characterized in that, The step of using the preset genetic algorithm to solve the daily train request plan adjustment optimization model to obtain the daily train request plan adjustment scheme includes: taking the number of trains to be loaded on each railway line after the daily adjustment of each mine as the second decision variable, taking the reciprocal of the second objective function as the fitness function, and solving the problem according to the preset genetic algorithm solution process within the range of the second constraints to generate the daily train request plan adjustment scheme.

8. A smart system for generating coal railway transport dispatch plans under multi-mine collaboration, characterized in that, include: The data acquisition module is used to acquire multi-source data, including the next-day demand plans of railway transportation customers and information on various coal mines; The planning and optimization module is used to construct a day-ahead vehicle request planning and optimization model based on the multi-source data, and to solve the day-ahead vehicle request planning and optimization model using a preset genetic algorithm to obtain the day-ahead vehicle request plan for the next day. The execution deviation prediction module is used to predict the execution deviation of the next day's vehicle request plan during the execution process. The planning adjustment module is used to construct an intraday vehicle request plan adjustment optimization model based on the execution deviation, and use the preset genetic algorithm to solve the intraday vehicle request plan adjustment optimization model to obtain an intraday vehicle request plan adjustment scheme.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images