Redundant path design system and method for improving stability of coal conveying system

Abstract

The invention relates to the technical field of data processing, and discloses a redundant path design system and method for improving the stability of a coal transmission system, and the method comprises the steps: obtaining the historical operation data of the coal transmission system, constructing an equipment fault prediction model and a transportation demand prediction model, and predicting the health state and transportation demand of equipment in a preset time; based on the prediction result and the network topology structure of the coal conveying system, constructing a redundant path multi-objective optimization model to obtain a redundant path set; and finally, performing intelligent flow control based on the redundant path set. The stability and the adaptability of the coal conveying system can be improved, and the continuity and the high efficiency of coal conveying can still be guaranteed under the conditions of equipment failure risks and transportation demand changes.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically, to a redundant path design system and method for improving the stability of a coal transmission system. Background Art

[0002] In the prior art, the Chinese patent application with publication number CN117666367A discloses a remote control method and system for underground machinery in coal mines, which realizes real-time monitoring and updating of equipment status through a graph neural network algorithm. This method can effectively monitor the equipment status and update node information, but the scheme mainly focuses on monitoring and path planning at the equipment level, and fails to involve complex issues such as path redundancy and flow distribution in the entire coal transportation system. At the same time, the application of graph neural networks in equipment status monitoring has improved the speed and accuracy of information processing, but in actual applications, due to the lack of unified modeling of multi-objective optimization of equipment failures and transportation needs, path redundancy design and flow management are difficult to fully play their due role. Therefore, although this scheme has certain application value in the remote control of underground mechanical equipment, it still has certain defects in redundant path optimization and flow control of coal transportation systems.

[0003] The Chinese patent with the authorization announcement number CN118036841B discloses a dynamic material loading scheduling method and system based on smart mining services. By analyzing the mine material categories, storage volume and production business data, the decision tree network is used to evaluate the timeliness and feasibility of the loading scheduling strategy. This method improves the timeliness and rationality of material loading scheduling to a certain extent, but it focuses on the loading scheduling of mine materials, and pays more attention to the feasibility evaluation and scheduling optimization of transportation tasks. It has not yet considered how to effectively design redundant paths, perform flow management, or respond to dynamic changes in transportation paths in real time during transportation. Therefore, although this technology provides a scheduling optimization solution for mine transportation, its application in redundant path design and flow control of coal transportation systems is still insufficient.

[0004] In summary, the existing technology lacks an effective multi-objective optimization model to simultaneously consider redundant path design and flow distribution, which results in the system being unable to respond flexibly under high load or equipment failure conditions. Although the existing equipment failure prediction and transportation demand prediction models have achieved certain results in their respective fields, the existing technology lacks a unified application of these two types of prediction results, fails to effectively guide path optimization and flow scheduling decisions, and limits the efficiency improvement of the entire coal transportation system. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a redundant path design system and method for improving the stability of the coal transmission system. By combining equipment status prediction and transportation demand prediction, the problem that path planning and flow management cannot be optimized synchronously in the prior art is solved, and the stability, reliability and transportation efficiency of the coal transmission system are further improved.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The redundant path design method for improving the stability of coal transmission system includes:

[0008] Acquire historical operation data of the coal transmission system, and construct a first data set based on the historical operation data; construct and train an equipment failure prediction model and a transportation demand prediction model based on the first data set; use the trained equipment failure prediction model and transportation demand prediction model to predict the equipment health status and transportation demand within a preset time, and obtain a first prediction result and a second prediction result;

[0009] The network topology structure of the coal transportation system is obtained, and a redundant path multi-objective optimization model is constructed based on the first prediction result and the second prediction result, as well as the network topology structure of the coal transportation system; a redundant path set is obtained according to the redundant path multi-objective optimization model; and intelligent flow control is performed based on the redundant path set.

[0010] Furthermore, the historical operation data includes historical equipment operating parameters and historical coal transportation volume data; the historical equipment operating parameters and historical coal transportation volume data are sorted by time to form a historical equipment operating parameter sequence and a historical coal transportation volume data sequence;

[0011] The constructing the first data set comprises:

[0012] Perform statistical analysis on the historical equipment condition parameter sequence, extract the historical health status characteristics, and form the historical health status characteristic sequence;

[0013] Conduct time series analysis on the coal transportation volume data series, extract historical transportation demand indicators, and form a historical transportation demand indicator series;

[0014] The historical health status feature sequence and the historical transportation demand index sequence are combined to form a first data set.

[0015] Furthermore, the construction and training of the equipment failure prediction model and the transportation demand prediction model includes:

[0016] Using the time series prediction algorithm, we build an equipment failure prediction model based on the historical health status feature sequence; using the regression prediction algorithm, we build a transportation demand prediction model based on the historical transportation demand indicator sequence;

[0017] The first prediction result includes the equipment failure rate FA within a preset time, and the second prediction result includes the predicted transportation volume TR within a preset time.

[0018] Furthermore, the construction of the redundant path multi-objective optimization model includes:

[0019] Acquire a network topology structure of a coal transportation system, wherein the network topology structure of the coal transportation system includes a path set and a node set;

[0020] The equipment failure rate FA and the predicted transportation volume TR within the preset time are used as the constraints of the redundant path multi-objective optimization model;

[0021] Defining decision variables of a redundant path multi-objective optimization model, wherein the decision variables include a path selection variable and a traffic distribution ratio;

[0022] Determining an objective function of a redundant path multi-objective optimization model, wherein the objective function includes minimizing coal transportation costs, minimizing equipment failure risks, and maximizing transportation efficiency;

[0023] According to the path set and node set in the network topology structure of the coal transportation system, as well as the decision variables and objective function of the redundant path multi-objective optimization model, a redundant path multi-objective optimization model is constructed.

[0024] Furthermore, obtaining the redundant path set according to the redundant path multi-objective optimization model includes:

[0025] Solve the redundant path multi-objective optimization model and obtain the Pareto efficient solution set; each solution in the Pareto efficient solution set corresponds to a redundant path solution;

[0026] Select the main path from the Pareto efficient solution set;

[0027] Using the main path plan, calculate the transportation risk and determine whether an alternative path is needed; if an alternative path is not needed, the main path forms a redundant path set; if an alternative path is needed, select the alternative path from the remaining Pareto efficient solution set after eliminating the main path plan; combine the main path and the alternative path to form a redundant path set.

[0028] Further, selecting the main path from the Pareto efficient solution set includes:

[0029] Calculate the normalized value y of each solution in the Pareto efficient solution set on each objective l,g , the objectives include coal transportation cost, equipment failure risk and transportation efficiency, y l,grepresents the normalized value of the lth solution in the Pareto efficient solution set on the gth target; 1≤g≤m, m is the total number of targets; l is the index of the solution in the Pareto efficient solution set;

[0030] Set the target weight vector w = [w1,w2,…,w m ],w m is the weight of the mth target; the weighted summation method is used, based on the target weight vector w and the normalized value y of each solution on each target l,g , calculate the comprehensive evaluation index for each solution; where w1+w2+…+w m =1;

[0031] The solution with the largest comprehensive evaluation index is taken as the optimal solution, and the redundant path solution corresponding to the optimal solution is selected as the main path solution;

[0032] According to the main path plan, the node sequence that the main path passes through is determined to form the physical topology structure of the main path.

[0033] Furthermore, the physical topology structure forming the main path includes:

[0034] Step S22241, initialize the node sequence N of the main path main is an empty collection;

[0035] Step S22242, select the starting point s' of the main path as the current node and add it to N main middle;

[0036] Step S22243: for the current node s', find the path selection variable x that satisfies the main path solution s't' =1 next node t';

[0037] Step S22244, add node t' to N main and set node t' as the new current node;

[0038] Step S22245, repeating steps S22243 and S22244 until reaching the end of the main path;

[0039] Step S22246: Set the node sequence N of the main path main The physical topology that serves as the primary path.

[0040] Further, the determining whether a backup path is needed includes:

[0041] Select variable x according to the path in the main path scheme i'j' , determine the path set E that the main path passes through main ;x i'j'Indicates whether to select the path (i', j') from node i' to node j'. If so, then x i'j' =1, otherwise x i'j' =0;

[0042] For each path (i', j')∈E on the main path main , according to the flow distribution ratio f of path (i', j') i'j' , allocate the predicted transport volume TR to the path (i', j'), and obtain the transport volume TR allocated to the path (i', j') i'j' ;

[0043] For each path (i', j')∈E on the main path main , combined with the equipment failure probability p of path (i', j') i'j' Economic loss caused by failure i'j' , and the transport volume TR allocated to the path (i', j') i'j' , calculate the transportation risk R of path (i',j') i'j' ;

[0044] Sum the transportation risks of all paths on the main path to get the total transportation risk R of the main path main ;

[0045] The total transportation risk R of the main path main Compared with the preset risk threshold R th Compare; if R main >R th , then an alternative path is required; otherwise, no alternative path is required.

[0046] Further, the selecting of a backup path from the remaining Pareto efficient solution set after eliminating the main path solution includes:

[0047] Eliminate the optimal solution corresponding to the main path from the Pareto efficient solution set to form a set of alternative paths;

[0048] Calculate the transportation risk R of each alternative path in the alternative path set b , selecting the first N1 solutions with the lowest transportation risk in the alternative path set to form a first alternative path set, wherein the first alternative path set includes N1 first alternative paths;

[0049] Calculate the similarity SI between the first backup path and the main path k , SI k Indicates the similarity between the kth first backup path and the main path; 1≤k≤N1;

[0050] According to the similarity SI kThe first backup paths are sorted in ascending order, and the first N2 first backup paths are selected as final backup paths; N2≤N1.

[0051] A redundant path design system for improving the stability of a coal transmission system, which is used to implement the redundant path design method for improving the stability of a coal transmission system, and the system includes:

[0052] Model building module: acquiring historical operation data of the coal transmission system, and building a first data set based on the historical operation data; building and training an equipment failure prediction model and a transportation demand prediction model based on the first data set;

[0053] Prediction module: uses the trained equipment failure prediction model and transportation demand prediction model to predict the equipment health status and transportation demand within a preset time to obtain the first prediction result and the second prediction result;

[0054] Multi-objective optimization module: obtaining the network topology of the coal transportation system, building a redundant path multi-objective optimization model based on the first prediction result and the second prediction result, as well as the network topology of the coal transportation system; obtaining a redundant path set according to the redundant path multi-objective optimization model;

[0055] Control module: Intelligent flow control based on a set of redundant paths.

[0056] An electronic device includes a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit. When the central processing unit executes the computer program, the redundant path design method for improving the stability of a coal transmission system is implemented.

[0057] A computer-readable storage medium stores a computer program, which, when executed, implements the redundant path design method for improving the stability of a coal transmission system.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The redundant path design method for improving the stability of the coal transmission system proposed in the present invention can accurately predict the equipment health status and transportation demand changes within a preset time by constructing an equipment failure prediction model and a transportation demand prediction model. Based on the prediction results, combined with the network topology of the coal transportation system, a redundant path multi-objective optimization model is constructed, which comprehensively considers multiple goals such as transportation cost, equipment failure risk and transportation efficiency, optimizes path selection and flow distribution, and obtains the optimal redundant path set. By evaluating the transportation risk of the main path, adding backup paths when necessary, and selecting backup paths with large differences based on path similarity, it is possible to effectively disperse transportation risks and improve the robustness of the system.

[0060] In addition, the present invention also introduces an intelligent flow control mechanism, which monitors the flow of the main path and the backup path in real time, calculates the flow balance, and dynamically adjusts the flow distribution to ensure the reasonable distribution of the path load. This dynamic flow adjustment mechanism can adapt to the fluctuation of transportation demand, while ensuring transportation efficiency, reducing the risks caused by uneven flow.

[0061] The present invention makes full use of data-driven modeling and optimization methods, combines equipment status prediction, transportation demand prediction and path optimization, and constructs a complete set of redundant path design solutions. This solution can maximize the continuity, stability and efficiency of coal transportation through reasonable planning of redundant paths and intelligent flow control in the case of equipment failure risks and changes in transportation demand. Compared with traditional static path planning, the present invention has stronger adaptability and robustness, and can cope with complex and changeable actual transportation scenarios. This has important practical significance for improving the overall performance of the coal transportation system, reducing transportation costs, and improving resource utilization. The present invention provides new ideas and methods for the intelligent and digital transformation of the coal transportation industry, and has a positive role in promoting technological progress and sustainable development in the industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 It is a principle flow chart of the redundant path design method for improving the stability of the coal transmission system in the present invention;

[0064] Figure 2 A flow chart of a method for constructing a first data set in a redundant path design method for improving the stability of a coal transmission system according to the present invention;

[0065] Figure 3 A flow chart of a method for constructing and training an equipment failure prediction model and a transportation demand prediction model in a redundant path design method for improving the stability of a coal transportation system according to the present invention;

[0066] Figure 4 A flow chart of a method for constructing a redundant path multi-objective optimization model in a redundant path design method for improving the stability of a coal transmission system according to the present invention;

[0067] Figure 5A flowchart of a method for obtaining a redundant path set according to a redundant path multi-objective optimization model in a redundant path design method for improving the stability of a coal transmission system of the present invention;

[0068] Figure 6 A flow chart of a method for forming a physical topology structure of a main path in a redundant path design method for improving the stability of a coal transmission system according to the present invention;

[0069] Figure 7 A flow chart of a method for determining whether a backup path is needed in a redundant path design method for improving the stability of a coal transmission system according to the present invention;

[0070] Figure 8 A flow chart of a method for selecting a backup path from a set of remaining Pareto efficient solutions after removing a main path solution in a redundant path design method for improving the stability of a coal transmission system according to the present invention;

[0071] Fig. 9 A flow chart of a method for intelligent flow control based on a redundant path set in a redundant path design method for improving the stability of a coal transmission system according to the present invention;

[0072] Fig.10 This is a functional module diagram of the redundant path design system for improving the stability of the coal transmission system in the present invention. DETAILED DESCRIPTION

[0073] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] Example 1

[0075] See also Figure 1 As shown, this embodiment provides a redundant path design method for improving the stability of the coal transmission system, including:

[0076] Step S1000, obtaining historical operation data of the coal transmission system, constructing a first data set based on the historical operation data; constructing and training an equipment failure prediction model and a transportation demand prediction model based on the first data set; using the trained equipment failure prediction model and transportation demand prediction model to predict the equipment health status and transportation demand within a preset time, and obtaining a first prediction result and a second prediction result;

[0077] Furthermore, step S1000 includes:

[0078] Step S1100, obtaining historical operation data of the coal transportation system, wherein the historical operation data includes historical equipment operating parameters and historical coal transportation volume data; sorting the historical equipment operating parameters and historical coal transportation volume data by time to form a historical equipment operating parameter sequence and a historical coal transportation volume data sequence;

[0079] Specifically, historical equipment operating parameters refer to a series of status parameters of coal transportation equipment during operation. These parameters reflect the working condition of the equipment and can help identify potential failure risks or performance bottlenecks. Common equipment operating parameters include:

[0080] Speed: The operating speed of the device reflects the workload of the device.

[0081] Temperature: The temperature of the equipment during operation. Excessive temperature may be a precursor to failure.

[0082] Vibration frequency and amplitude: Vibration is a common precursor to failure of mechanical equipment. Changes in vibration frequency and amplitude can reflect the degree of wear or damage to the equipment.

[0083] Angle and speed: The angle and speed of the equipment are important operating parameters, which are usually used to judge the load condition of the equipment.

[0084] Fault status: whether a device fault has occurred, as well as the time, duration, and type of fault.

[0085] Historical coal transportation data reflects the actual situation of coal transportation, including the size of transportation volume and specific parameters of the transportation process. These data are crucial for evaluating system operation efficiency, transportation demand, and whether there are bottlenecks. They mainly include:

[0086] Coal order volume: refers to the total amount of coal that needs to be transported within a certain period of time.

[0087] Actual loading volume: refers to the actual amount of coal loaded.

[0088] Number of transport vehicles: The number of vehicles used to transport coal.

[0089] Single car load: The coal load per transport car.

[0090] Daily transport frequency: The number of transports carried out in one day reflects the operating frequency of the system.

[0091] Transportation Time: The time required to complete the transportation, including the time for loading, transporting and unloading.

[0092] By acquiring and sorting these historical data by time, two sets of sequences can be formed:

[0093] Historical equipment condition parameter sequence: Equipment condition parameter data sorted by time can help analyze the operating status and failure frequency of the equipment.

[0094] Historical coal transportation volume data series: transportation volume data sorted by time, used to reveal the changing trend of coal transportation demand and system load conditions.

[0095] In the process of optimizing and improving the coal transportation system, it is crucial to accurately obtain and process historical operation data, which provides the basis for subsequent analysis, modeling and optimization decisions.

[0096] Step S1200, obtaining a historical health status feature sequence and a historical transportation demand index sequence according to historical operation data, and constructing a first data set;

[0097] Furthermore, if Figure 2 As shown, step S1200 includes:

[0098] Step S1210, statistically analyzing the historical equipment operating condition parameter sequence, extracting historical health status characteristics, and forming a historical health status characteristic sequence;

[0099] Step S1220, performing time series analysis on the coal transportation volume data sequence, extracting historical transportation demand indicators, and forming a historical transportation demand indicator sequence;

[0100] Step S1230, combining the historical health status feature sequence and the historical transportation demand index sequence to form a first data set.

[0101] Specifically, step S1200 aims to extract features and indicators that are helpful for optimizing the coal transportation system through in-depth analysis of historical operation data. This process mainly includes two aspects: on the one hand, extracting health status features from equipment operating parameters, and on the other hand, extracting transportation demand indicators from coal transportation volume data. These two indicator sequences are combined to form a complete data set for subsequent prediction and decision-making.

[0102] Through statistical analysis of historical equipment operating parameters, important features that reflect the health status of equipment can be extracted. These features can reveal the stability, reliability and maintenance needs of equipment operation, help evaluate whether the equipment is operating normally, and predict possible failures. The historical health status features include equipment failure rate, mean time between failures (MTBF), mean time to repair (MTTR), equipment availability and equipment utilization;

[0103] Equipment failure rate: indicates the frequency of failures within a certain period of time, usually calculated as the ratio of the number of failures per unit time to the total working time. A high failure rate usually means increased equipment instability, indicating that the system may face a greater risk of downtime. By monitoring the failure rate in real time, it is possible to identify and take maintenance measures in advance to reduce the frequency of system downtime.

[0104] Mean Time Between Failures (MTBF): Indicates the average time between the occurrence of one equipment failure and the next. The larger the MTBF, the higher the reliability of the equipment. Long-term trouble-free operation can provide more stable support for production and reduce maintenance and replacement costs.

[0105] Mean Time to Repair (MTTR): It indicates the average time from when equipment fails to when it is repaired. A shorter MTTR means faster equipment repair and shorter system downtime, which improves system availability and efficiency.

[0106] Equipment availability rate: refers to the ratio of equipment in normal working condition over a period of time. It is the ratio of normal working time to total time. The higher the equipment availability rate, the better the operating efficiency and stability of the system, which can reduce the risk of production interruption.

[0107] Equipment utilization rate: refers to the proportion of the time that the equipment is in effective working state in a period of time to the total working time. A higher equipment utilization rate means that the equipment is more efficient when working, reducing idle time and resource waste.

[0108] Coal transportation volume data reflects the changing trend of demand during coal transportation. Through time series analysis, important indicators reflecting transportation demand fluctuations, seasonal changes and long-term growth trends can be extracted. The historical transportation demand indicators include daily average transportation volume, monthly transportation volume change rate, annual transportation volume growth rate and transportation volume seasonality index;

[0109] Average daily transport volume: reflects the basic scale of transport demand, calculated by dividing the total transport volume in a certain period by the number of days. Average daily transport volume provides a basic perspective on transport demand and can be used to formulate reasonable transport plans and ensure the rational allocation of transport resources.

[0110] Monthly transportation volume change rate: reflects the short-term trend of transportation demand, indicating the difference between this month's transportation volume and last month's transportation volume. By analyzing the monthly transportation volume change rate, it is possible to quickly identify fluctuations and abnormalities in transportation demand and adjust transportation strategies in a timely manner.

[0111] Annual transportation volume growth rate: reflects the long-term trend of transportation demand, indicating the growth rate of this year's transportation volume compared with last year's transportation volume. The annual transportation volume growth rate can help grasp the long-term trend of transportation demand and help make future resource planning and demand forecasts.

[0112] Transport volume seasonality index: reflects the seasonal characteristics of transport demand and evaluates transport demand in different seasons by analyzing the monthly distribution of historical data. The calculation steps are as follows:

[0113] Classify the transportation volume data of many years by month to obtain the multi-year average transportation volume of each month;

[0114] Calculate the average of all monthly transport volumes (i.e. the average of the total annual transport volume), compare the average monthly transport volume with the multi-year average transport volume, and calculate the seasonal index for each month;

[0115] If the seasonal index is greater than 1, it means that the transportation demand in that month is higher than the annual average; if it is less than 1, it means that it is lower than the annual average.

[0116] Through the seasonal index, we can better understand the cyclical fluctuations in transportation demand, optimize the allocation of transportation resources, and avoid excessive or insufficient transportation.

[0117] The historical health status feature sequence and the historical transportation demand index sequence extracted in step S1210 and step S1220 are combined to form a first data set. This data set provides comprehensive input data for subsequent prediction models, optimization algorithms, and decision support systems, which can help the system predict trends in equipment and transportation demand based on historical data and make more accurate scheduling and maintenance decisions.

[0118] Step S1300, constructing and training an equipment failure prediction model and a transportation demand prediction model according to the first data set;

[0119] Furthermore, if Figure 3 As shown, step S1300 includes:

[0120] Step S1310, using a time series prediction algorithm to build an equipment failure prediction model based on a historical health status feature sequence;

[0121] Step S1320, using a regression prediction algorithm, based on the historical transportation demand indicator sequence, constructs a transportation demand prediction model.

[0122] Specifically, the core task of step S1300 is to use the first data set (composed of a historical equipment operating parameter sequence and a historical coal transportation volume data sequence) to construct and train two prediction models: an equipment failure prediction model and a transportation demand prediction model. These two models are used to predict the probability of equipment failure and the change trend of future coal transportation volume, respectively.

[0123] The goal of the equipment failure prediction model is to predict the probability of equipment failure at a certain point in the future based on the equipment's historical operating data. Using time series prediction algorithms, especially deep learning models such as long short-term memory networks (LSTM) or gated recurrent unit networks (GRU), can capture long-term dependencies from the equipment's historical operating parameter sequences and accurately predict equipment failures.

[0124] LSTM is a special type of recurrent neural network (RNN) designed to capture long-term dependencies in time series data. Compared with traditional RNN, LSTM avoids the gradient vanishing problem by introducing memory cells and gating mechanisms (input gate, forget gate, and output gate), so that it can effectively process long-term time series data. LSTM can better capture the patterns of long-term changes in equipment operation, such as temperature fluctuations and gradual increase in vibration amplitude, which may be precursors to equipment failure. GRU is another improved RNN, similar to LSTM, but its structure is simpler, with only update gates and reset gates, and higher computational efficiency. GRU may be more efficient than LSTM in processing shorter-term dependencies, and is suitable for simpler and faster equipment failure prediction scenarios.

[0125] Equipment failure prediction model training process:

[0126] Input: Historical health status feature sequence, such as equipment temperature, vibration frequency, rotation speed and other parameters in the past period of time.

[0127] Label: Whether the device fails at the next moment, usually a binary label (0: not failed, 1: failed). For example, if historical data shows that the device fails when the temperature exceeds a certain threshold, the label of the training data is "1" (failure), otherwise it is "0" (no failure).

[0128] Output: The model predicts the probability of a device failure within a preset time. The output value is a number between [0,1], indicating the probability of a failure occurring.

[0129] Through the training of equipment fault prediction models, it is possible to achieve accurate early warning of equipment health status and identify potential faults in advance. Taking coal transportation equipment as an example, if the temperature of the equipment continues to rise and exceeds the normal range, the fault prediction model can issue an alarm in advance, so as to take preventive maintenance measures, avoid equipment downtime, and reduce the losses caused by unexpected downtime.

[0130] The goal of the transportation demand forecasting model is to predict the coal transportation demand (i.e., coal transportation volume) within a preset time based on historical coal transportation volume data. In order to deal with this continuous numerical prediction problem, a regression prediction algorithm is usually used. Commonly used regression algorithms include support vector machine regression (SVR) and gradient boosting decision tree (GBDT).

[0131] SVR is the application of support vector machine (SVM) in regression problems. It performs regression prediction by finding a hyperplane that best fits the data. SVR has strong capabilities in modeling high-dimensional data and nonlinear data. It is suitable for complex regression problems such as coal transportation volume prediction. Especially when the data is noisy or fluctuates irregularly, SVR can maintain good prediction accuracy. GBDT is an integrated learning algorithm that trains multiple decision trees and optimizes the model using the gradient boosting method to achieve higher prediction accuracy in complex regression problems. GBDT can accurately model nonlinear and complex transportation demand data by gradually optimizing and combining the prediction results of multiple decision trees. It is especially suitable for situations with multiple influencing factors (such as seasonal changes, market demand fluctuations, etc.).

[0132] Transportation demand forecasting model training process:

[0133] Input: Historical transport demand indicator series, such as daily average transport volume, monthly transport volume change rate, annual transport volume growth rate, etc.

[0134] Label: The transportation volume at a certain moment in the future, usually a continuous value, such as the coal transportation volume in the next day (unit: tons), the total transportation volume in the next month (unit: tons).

[0135] Output: The model predicts the coal transportation volume within a preset time.

[0136] The transportation demand forecasting model can accurately predict the trend of future coal transportation volume changes. For example, in the case of seasonal changes in coal production, the model can accurately predict the increase in transportation volume in a certain month and provide a scientific basis for the scheduling and resource allocation of transportation vehicles. In this way, the waste of transportation resources can be effectively reduced and the transportation efficiency of the system can be improved.

[0137] In order to ensure the generalization ability and prediction accuracy of the model, step S1300 also includes data set division and model evaluation process.

[0138] Division of training set, validation set and test set:

[0139] The first data set is divided into a training set, a validation set and a test set. Generally, the training set accounts for 70%-80%, the validation set accounts for 10%-15%, and the test set accounts for 10%-15%.

[0140] The training set is used to train the model, the validation set is used to tune the model's hyperparameters, and the test set is used to evaluate the final performance of the model.

[0141] Model training:

[0142] The equipment failure prediction model and transportation demand prediction model are trained using the training set data. The model parameters are optimized through repeated iterations, so that the model can gradually improve its prediction accuracy.

[0143] Model evaluation and tuning:

[0144] Use the validation set data to evaluate the performance of the trained model. Common evaluation indicators include:

[0145] Equipment failure prediction model: accuracy, recall rate, F1 value, etc.

[0146] Transportation demand forecasting model: mean square error (MSE), mean absolute error (MAE), etc.

[0147] Adjust the model's hyperparameters based on the evaluation results to avoid overfitting or underfitting and ensure that the model has good generalization capabilities.

[0148] Cross Validation:

[0149] Cross-validation is a commonly used model evaluation method. By dividing the data set into multiple subsets and alternating between different subsets for training and validation, the stability and accuracy of the model can be effectively improved, and the deviation caused by a single data division can be avoided.

[0150] Through a reasonable training, validation and testing process, the prediction accuracy of the model can be effectively improved and the reliability of the model in practical applications can be ensured. For example, through cross-validation, the overfitting of the model on a specific data set can be prevented, thereby improving the generalization ability of the model and providing accurate prediction results stably in different time periods and different coal transportation scenarios.

[0151] Step S1400, using the trained equipment failure prediction model and transportation demand prediction model, predicting the equipment health status and transportation demand within a preset time, and obtaining a first prediction result and a second prediction result, wherein the first prediction result includes the equipment failure rate FA within the preset time, and the second prediction result includes the predicted transportation volume TR within the preset time;

[0152] Further, step S1400 includes:

[0153] Step S1410, obtaining a real-time equipment operating condition parameter sequence, obtaining a real-time health status feature sequence according to the real-time equipment operating condition parameter sequence, inputting the real-time health status feature sequence into a trained equipment failure prediction model, and obtaining a first prediction result;

[0154] Step S1420, obtaining a real-time coal transportation volume data sequence, obtaining a real-time transportation demand index sequence according to the real-time coal transportation volume data sequence, inputting the real-time transportation demand index sequence into a trained transportation demand prediction model, and obtaining a second prediction result.

[0155] Specifically, the real-time equipment condition parameter sequence refers to the equipment operating status data at the current moment and within a period of time before, which will reflect the immediate working status of the equipment. Common real-time equipment condition parameters include: equipment speed, temperature, vibration frequency, rotation speed, pressure, etc. For example, the real-time condition data of equipment A may include: temperature = 80°C, vibration frequency = 50Hz, rotation speed = 3000rpm, which constitute a set of real-time equipment condition parameter sequences. Based on the real-time equipment condition parameter sequence, the health status feature sequence of the equipment is calculated according to the same feature extraction method as in step S1210. These features include: equipment failure rate, MTBF, MTTR, equipment integrity rate and equipment utilization rate. Assuming that the real-time condition data of the equipment includes temperature and vibration frequency, it is possible to generate a feature sequence reflecting the current health status of the equipment by calculating the historical average value, standard deviation and other features of these parameters. By extracting the health status features of the equipment in real time, the immediate operation status of the equipment can be effectively tracked, and accurate input can be provided for the equipment failure prediction model.

[0156] The health status feature sequence calculated in real time is input into the trained equipment failure prediction model (such as LSTM or GRU) to obtain the probability FA of equipment failure within the preset time. The output of the equipment failure prediction model is the probability of failure within the preset time. For example, the model predicts that the probability of equipment failure within the next 24 hours is 0.15, which means that the probability of equipment failure is 15%.

[0157] By collecting and processing equipment condition data in real time, the system can monitor the health status of the equipment in real time and predict future failure risks. This predictive capability is crucial for discovering potential equipment problems in advance. For example, if the temperature and vibration amplitude of a certain device have gradually increased over the past few hours, the fault prediction model can issue an alarm in advance and recommend preventive maintenance to avoid sudden shutdown of the equipment due to failure, reducing production interruptions and repair costs caused by sudden failures.

[0158] The real-time coal transportation volume data sequence refers to the coal transportation data at the current moment and in the previous period of time. These data reflect the actual load of the transportation system, including coal order quantity, actual loading quantity, number of transportation vehicles, etc. For example, the real-time transportation volume data may include: coal order quantity = 5,000 tons, actual loading quantity = 4,800 tons, number of transportation vehicles = 120. Based on the real-time coal transportation volume data sequence, according to the same indicator calculation method as in step S1220, the historical transportation demand characteristics are extracted, including the average daily transportation volume, the monthly transportation volume change rate, the annual transportation volume growth rate, the transportation volume seasonality index, etc. For example, based on the transportation volume data of the past 7 days, the average daily transportation volume, the monthly transportation volume change rate and other indicators are calculated. These indicators reflect the basic laws and fluctuation trends of transportation demand. By extracting real-time transportation demand indicators, the system can accurately grasp the fluctuation trend of transportation demand, adjust the transportation plan in time, and avoid waste of resources or transportation bottlenecks.

[0159] The transport demand indicator sequence extracted in real time is input into the trained transport demand forecasting model (such as SVR or GBDT) to obtain the predicted transport volume TR within the preset time. The output of the transport demand forecasting model is the predicted value of coal transportation volume within the preset time. For example, the model predicts that the transportation volume in the next day will be 5,200 tons and the transportation volume in the next month will be 160,000 tons. By real-time monitoring and forecasting of transportation demand, the system can dynamically adjust transportation plans and resource allocation. For example, in seasons when transportation demand rises sharply, the forecasting model can inform the increasing trend of transportation volume in advance, help enterprises to allocate transportation vehicles and equipment in advance, and avoid insufficient transportation resources. When demand decreases, it can also reduce the input of transportation resources in time and optimize costs.

[0160] The core of step S1400 is to acquire and process equipment and transportation data in real time, and combine the trained prediction model to make predictions within a preset time, so as to provide scientific decision-making support for the coal transportation system. Specifically, through the prediction of equipment failure rate, it is possible to identify the risk of equipment failure in advance and take preventive measures, while through the prediction of transportation volume, it is possible to provide an accurate basis for the configuration and scheduling of transportation resources. Overall, these prediction models provide strong data support for the intelligent management of the coal transportation system, helping the system to improve operational efficiency, reduce risks, and respond flexibly to a changing market environment.

[0161] Step S2000, obtaining a network topology structure of the coal transportation system, and constructing a redundant path multi-objective optimization model based on the first prediction result and the second prediction result, as well as the network topology structure of the coal transportation system; and obtaining a redundant path set according to the redundant path multi-objective optimization model;

[0162] Furthermore, step S2000 includes:

[0163] Step S2100, obtaining the network topology of the coal transportation system, and constructing a redundant path multi-objective optimization model based on the equipment failure rate FA and the predicted transportation volume TR within a preset time, as well as the network topology of the coal transportation system; the network topology of the coal transportation system includes a path set and a node set;

[0164] Furthermore, if Figure 4 As shown, step S2100 includes:

[0165] Step S2110, using the equipment failure rate FA and the predicted transportation volume TR within a preset time as constraint conditions of the redundant path multi-objective optimization model;

[0166] Step S2120, defining decision variables of the redundant path multi-objective optimization model, including path selection variables and traffic distribution ratio;

[0167] The path selection variable can be represented by a binary variable, which takes a value of 1 if selected and 0 otherwise; the traffic distribution ratio is represented by a continuous variable, which takes a value range of [0,1], indicating the proportion of the transportation volume allocated to each path to the total transportation volume;

[0168] Step S2130, determining the objective function of the redundant path multi-objective optimization model, including minimizing the coal transportation cost, minimizing the risk of equipment failure, and maximizing the transportation efficiency;

[0169] Step S2140, constructing a redundant path multi-objective optimization model according to the path set and the node set in the network topology structure of the coal transportation system, and the decision variables and the objective function of the redundant path multi-objective optimization model.

[0170] The redundant path multi-objective optimization model includes:

[0171] Let G = (V, E) represent the network topology of the coal transportation system, where V is the node set and E is the path set. Define the following decision variables:

[0172] x ij : Path selection variable, a binary variable indicating whether to select the path from node i to node j. If selected, then x ij =1, otherwise x ij =0; i and j are node indices, i∈V, j∈V.

[0173] f ij : Traffic allocation ratio is a continuous variable, which represents the traffic allocation ratio (the proportion of transport volume to the total transport volume) allocated to path (i, j), and its value range is [0,1].

[0174] Objective function 1: Minimize coal transportation cost:

[0175] min∑ (i,j)∈E c ij ·x ij ·f ij ;

[0176] Objective function 2: Minimize the risk of equipment failure:

[0177] min ∑ (i,j)∈E p ij ·x ij ·L ij ·f ij ;

[0178] Objective function 3: Maximize transportation efficiency:

[0179]

[0180] Constraints:

[0181] Constraint 1: Flow conservation:

[0182]

[0183] Constraint 2: Transport volume allocation:

[0184]

[0185] Constraint 3: Satisfy forecasted transportation demand:

[0186]

[0187] Constraint 4: Allocation ratio range:

[0188]

[0189] Constraint 5: Decision variable values:

[0190]

[0191] c ij : The cost of transporting a unit of coal on path (i, j) can be estimated based on factors such as path length, transportation time, and energy consumption; (i, j) represents a path from node i to node j.

[0192] x ji : Path selection variable, indicating whether to select the path from node j to node i. If so, then x ji =1, otherwise x ji =0;

[0193] p ij: The probability of equipment failure on path (i, j) can be obtained based on the equipment failure rate FA within the preset time output by the equipment failure prediction model.

[0194] L ij : If the equipment on path (i, j) fails, the expected economic loss can be estimated based on historical data or expert experience.

[0195] T i,j :The estimated transportation time of path (i, j) can be calculated based on the path length and the average transportation speed.

[0196] Q: Average load of coal transport vehicles, available based on vehicle type and historical data.

[0197] D: The predicted total coal transportation demand, which is obtained based on the predicted transportation volume TR within the preset time output by the transportation demand prediction model.

[0198] s: The source point of the coal transportation network, that is, the production or distribution center of coal.

[0199] t: The meeting point of the coal transportation network, that is, the destination or consumption place of the coal.

[0200] j:(i,j)∈E: represents the set of the end nodes j of all paths (i,j) starting from node i, including the end nodes of all paths starting from node i.

[0201] i:(i,j)∈E: represents the set of all paths (i,j) starting from node i, that is, all paths from i to other nodes.

[0202] Constraint 1 (flow conservation) ensures that the inflow is equal to the outflow for each node in the network except the source and sink; for the source, the outflow is 1, and for the sink, the inflow is 1. Constraint 2 (transport volume allocation) ensures that the sum of the flows of all paths is 1. Constraint 3 (meeting the predicted transportation demand) ensures that the total transportation volume of the network can meet the predicted total demand for coal transportation. Constraint 4 (allocation ratio range) limits the flow allocation ratio of each path to the range of [0,1]. Constraint 5 (decision variable value) limits the path selection variable x ij A variable that is either 0 or 1.

[0203] The redundant path multi-objective optimization model improves system reliability and reduces economic losses caused by equipment failure while controlling costs through objective function 1 (minimizing coal transportation costs) and objective function 2 (minimizing equipment failure risks). Through objective function 3 (maximizing transportation efficiency), the flow distribution of redundant paths is optimized, the coal in-transit time is reduced, and the overall efficiency of the transportation system is improved. Through constraints 1 (flow conservation) and 2 (transport volume distribution), the logistics balance in the coal transportation network is ensured to avoid node blockage or idle paths. Specifically, constraint 1 ensures that the inflow of each node except the source and sink is equal to the outflow, ensuring the flow balance of each node in the network; constraint 2 ensures that the sum of the flow distribution of all paths is 1, avoiding the situation of too much or too little flow in the network. Through constraint 3 (meeting the predicted transportation demand), it can ensure that the coal transportation network can meet the changes in transportation demand within the preset time, thereby improving the adaptability and flexibility of the system. By reasonably setting 0-1 decision variables and continuous decision variables, path selection and flow distribution are integrated in the same mathematical model, thus realizing the integrated optimization solution of redundant path planning. This model can more efficiently select the optimal path and traffic allocation plan, further improving the overall performance and adaptability of the transportation system.

[0204] The goal of step S2100 is to build a redundant path multi-objective optimization model based on the network topology of the coal transportation system (including a set of paths and a set of nodes), combined with the equipment failure rate FA and the predicted transportation volume TR within a preset time. Through this model, the following goals are achieved: optimize the reliability of coal transportation, reduce transportation costs, improve the transportation efficiency of the system, and effectively avoid the potential risks caused by equipment failures.

[0205] When constructing a redundant path multi-objective optimization model, it is first necessary to introduce the equipment failure rate FA and the predicted transportation volume TR as constraints. These factors affect the decision-making of path selection and flow distribution in the model, ensuring that the optimization results can not only meet the transportation needs, but also improve the stability and reliability of the system.

[0206] The equipment failure rate FA refers to the probability of equipment failure within a preset time, which reflects the reliability of equipment operation. Paths with higher equipment failure rates may require lower flow allocations to reduce transportation interruptions caused by equipment failures. Paths with high failure rates should avoid or reduce flow allocations to ensure the stability of coal transportation. If the probability of equipment failure on a certain path is high, the optimization model should tend to select other paths with higher equipment reliability. Assuming that the equipment failure rate of path (i, j) is 0.2, it means that the probability of failure on this path within the preset time is 20%. If the transportation demand for this path is too large, it may cause frequent equipment failures and affect the overall transportation efficiency. The optimization model will take this into account in the constraints and reasonably reduce the flow allocation of this path.

[0207] The predicted transportation volume TR refers to the total amount of transportation tasks that the coal transportation system needs to complete within the preset time, usually in tons. The transportation volume forecast not only reflects the load demand of the system, but is also closely related to the flow distribution of the path. The predicted transportation volume reflects the scale of transportation tasks faced by the system. To ensure that each path can meet the transportation demand, the model needs to reasonably allocate the transportation volume under the premise of ensuring reliability. Excessive transportation volume concentrated on a certain path may cause the path to be overloaded, thereby affecting the overall transportation efficiency and the health of the equipment. If the predicted transportation volume is 10,000 tons, and the equipment failure rate of a certain path in the system is high, the system should reasonably adjust the flow distribution according to the failure rate to avoid too many transportation tasks being concentrated on the path with a high failure rate.

[0208] When constructing a redundant path multi-objective optimization model, it is necessary to define decision variables, which will determine the optimal solution for path selection and traffic allocation. The two main decision variables are path selection variables and traffic allocation ratio.

[0209] Path selection variable x ij , is a binary variable indicating whether to select the path from node i to node j. If selected, then x ij =1, otherwise x ij =0; The path selection variable determines whether a path in the coal transportation system is selected for transportation. If the probability of path selection is high, the flow distribution ratio of the path will be large, otherwise it will be small. Suppose at a certain moment, there are two paths to choose from between node i and node j: path 1 and path 2. If the equipment failure rate of path 1 is low, then x ij The corresponding value of path 1 may be 1, and the value of path 2 may be 0. The optimization model will select the most appropriate path based on the equipment failure rate and transportation demand.

[0210] Traffic distribution ratio f ijis a continuous variable, which indicates the proportion of the transport volume allocated to path (i, j) to the total transport volume, and its value range is [0,1]. The flow distribution ratio directly affects the amount of coal transported on each path. The flow distribution ratio determines the distribution of each path in the total transport volume. When multiple paths are available, the system will adjust the flow distribution based on factors such as path reliability, transportation tasks, and failure probability. If the total transport volume of the system is 10,000 tons, and the flow distribution ratio f of path (i, j) is ij If the ratio is 0.3, the transport volume allocated to this route is 3,000 tons. By optimizing the traffic distribution ratio, the load of each route can be balanced to avoid overloading of some routes.

[0211] The objective function of the redundant path multi-objective optimization model includes minimizing coal transportation costs, minimizing equipment failure risks, and maximizing transportation efficiency. By balancing these objectives, the model is able to find the optimal path selection and flow allocation solution.

[0212] The cost of coal transportation includes direct and indirect costs due to factors such as transportation route selection, route usage time, and maintenance costs. One of the optimization goals is to reduce the overall transportation cost by reasonably selecting routes and traffic distribution. By optimizing route selection and traffic distribution, the overuse of high-cost routes can be avoided, thereby reducing unnecessary expenses. For example, routes with high reliability and low transportation costs can be selected to reduce maintenance and fuel consumption. The risk of equipment failure is related to the failure rate of equipment on the path. The goal is to avoid frequent equipment failures and ensure the health of the equipment by adjusting traffic distribution. By reasonably allocating transportation tasks and avoiding overuse of high-failure rate routes, the occurrence rate of equipment failures can be reduced, and the utilization rate of equipment and the stability of the system can be improved. Transportation efficiency refers to the amount of transportation tasks completed per unit time. One of the optimization goals is to use each route as efficiently as possible to ensure that the maximum transportation volume is completed in the shortest time. By optimizing traffic distribution and route selection, the overall operating efficiency of the coal transportation system can be improved, transportation time and resource waste can be reduced, and the response speed and service level of the overall system can be improved.

[0213] The redundant path multi-objective optimization model can effectively avoid transportation interruptions caused by equipment failures and improve the overall reliability and stability of the system by considering the optimization of equipment failure rate and flow distribution ratio. By reasonably allocating transportation tasks, excessive use of high-cost paths can be avoided, thereby reducing waste in transportation and reducing overall transportation costs. By maximizing transportation efficiency, the system's transportation capacity can be maximized while ensuring safety and reliability, improving overall operational efficiency and enhancing the system's responsiveness under high load.

[0214] Step S2200, obtaining a redundant path set according to a redundant path multi-objective optimization model;

[0215] Furthermore, if Figure 5 As shown, step S2200 includes:

[0216] Step S2210, solving the redundant path multi-objective optimization model to obtain a Pareto effective solution set; each solution in the Pareto effective solution set corresponds to a redundant path solution;

[0217] Specifically, the purpose of step S2210 is to solve the redundant path multi-objective optimization model and obtain a Pareto efficient solution set. The Pareto efficient solution set contains all solutions that cannot improve one objective without reducing other objectives, and represents the preferred combination of decision variables. In multi-objective optimization, there is usually no unique optimal solution, but there are multiple solutions, and the trade-offs and balances between them represent different optimization objectives.

[0218] Choosing a suitable multi-objective optimization algorithm is the key to solving the Pareto efficient solution set. Common multi-objective optimization algorithms include:

[0219] Weighted sum method: By assigning different weights to each objective, the weighted sum maximization or minimization problem is solved. This method solves the problem by converting multiple objective functions into a weighted sum.

[0220] Constraint method: optimize one objective function as the main objective, and use other objective functions as constraints. Gradually adjust the constraint values ​​of other objectives to optimize the main objective.

[0221] Genetic algorithm: Genetic algorithm generates possible solutions from generation to generation by simulating the natural selection process, and uses fitness function to evaluate the quality of solutions. It generates new solutions through crossover and mutation, thereby optimizing multiple objectives.

[0222] In practical applications, genetic algorithms are often used to deal with complex multi-objective optimization problems and can find near-optimal solutions in a large solution space. Through the iterative process of genetic algorithms, the system can find multiple different solutions, which will constitute a Pareto efficient solution set.

[0223] A Pareto efficient solution is a solution that cannot further improve a certain objective without reducing other objectives in multi-objective optimization. In other words, if there is a solution A that cannot increase or decrease the value of a certain objective without making the values ​​of other objectives worse, then the solution is a Pareto efficient solution. The Pareto efficient solution set contains all such solutions, which represent the optimal balance between all optimization objectives.

[0224] For example, in redundant path multi-objective optimization, the path selection and flow allocation involve multiple objectives: minimizing coal transportation costs, minimizing the risk of equipment failure, and maximizing transportation efficiency. Each solution in the Pareto efficient solution set represents a balance between these three, and it is impossible to optimize one objective without affecting other objectives.

[0225] Step S2220, selecting a main path from the Pareto efficient solution set;

[0226] Further, step S2220 includes:

[0227] Step S2221, calculate the normalized value y of each solution in the Pareto efficient solution set on each target l,g , the objectives include coal transportation cost, equipment failure risk and transportation efficiency, y l,g represents the normalized value of the lth solution in the Pareto efficient solution set on the gth target; 1≤g≤m, m is the total number of targets; l is the index of the solution in the Pareto efficient solution set;

[0228] Step S2222, set the target weight vector w=[w1,w2,…,w m ],w m is the weight of the mth target; the weighted summation method is used, based on the target weight vector w and the normalized value y of each solution on each target l,g , calculate the comprehensive evaluation index for each solution; where w1+w2+…+w m =1;

[0229] Step S2223, taking the solution with the maximum comprehensive evaluation index as the optimal solution, and selecting the redundant path solution corresponding to the optimal solution as the main path solution;

[0230] Specifically, since the result of the multi-objective optimization problem is a set of multiple solutions, the system needs to select an optimal solution, namely the main path solution, through further evaluation. The main path is the path solution that best balances multiple objectives.

[0231] Before selecting the main path, each solution in the Pareto efficient solution set needs to be normalized. Normalization is to convert target values ​​of different dimensions into the same standard range so that each target value is comparable. The specific steps are as follows:

[0232] Objectives include coal transportation costs, equipment failure risk, and transportation efficiency:

[0233] Coal transportation cost: reflects the economic cost of coal transportation, and the usual goal is to minimize the coal transportation cost.

[0234] Equipment failure risk: reflects the probability or risk of equipment failure, and the goal is usually to minimize the risk of equipment failure.

[0235] Transportation efficiency: refers to the amount of coal transported per unit time. The goal is to maximize transportation efficiency.

[0236] Normalization usually uses the linear normalization method, that is, mapping the target value to the interval [0,1]. During normalization, transportation efficiency is the maximization target, while coal transportation cost and equipment failure risk are the minimization targets. Therefore, during the normalization process, the normalized value of transportation efficiency is proportional to the original value, while the normalized value of coal transportation cost and equipment failure risk is inversely proportional to the original value. The result of normalization makes the range of each target value consistent, so that subsequent comparison and weighting can be carried out fairly.

[0237] The weighted summation method assigns different weights to each goal, calculates the comprehensive evaluation index of each solution, and then selects the optimal solution. The weight reflects the importance that decision makers attach to each goal. For the three goals of coal transportation cost, equipment failure risk, and transportation efficiency, decision makers can set their weights according to actual conditions. For example, if decision makers attach more importance to reducing transportation costs, they can give transportation costs a higher weight. Assume that there are three goals, transportation cost w1 = 0.4, equipment failure risk w2 = 0.3, and transportation efficiency w3 = 0.3, and the sum of these weights is 1.

[0238] For each solution in the Pareto efficient solution set, after calculating its normalized value on each objective, the weighted summation method is used to obtain the comprehensive evaluation index z of each solution: l :

[0239] z l =w1×y l,1 +w2×y l,2 +...+w m ×y l,m ;

[0240] Among them, y l,m represents the normalized value of the lth solution on the mth target, z l Represents the comprehensive evaluation index of the lth solution.

[0241] In step S2223, by calculating the comprehensive evaluation index of each solution, the solution with the maximum comprehensive evaluation index is selected. This solution represents the best balance point between coal transportation cost, equipment failure risk and transportation efficiency. After the optimal solution is selected, its corresponding redundant path solution is determined as the main path solution;

[0242] After calculating the comprehensive evaluation index of each solution, the solution with the largest index is selected as the optimal solution by comparing the comprehensive evaluation index of each solution. This means that although some solutions may perform well in a single objective, their overall performance is poor. Finally, a solution with the best overall performance is selected through the weighted summation method.

[0243] For example, suppose the original values ​​of solution A on the three objectives are: transportation cost 1 million yuan, equipment failure risk 0.1, and transportation efficiency 5,000 tons / hour; the original values ​​of solution B are: transportation cost 2 million yuan, equipment failure risk 0.05, and transportation efficiency 2,000 tons / hour. After normalization, the normalized values ​​of A may be (0.5, 0.5, 1.0), and the normalized values ​​of B may be (1.0, 1.0, 0.4).

[0244] If the weights of the three objectives are all 1 / 3, then the comprehensive evaluation index of solution A is:

[0245] z A =(1 / 3)×0.5+(1 / 3)×0.5+(1 / 3)×1.0=0.67;

[0246] The comprehensive evaluation index of solution B is:

[0247] z B =(1 / 3)×1.0+(1 / 3)×1.0+(1 / 3)×0.4=0.80;

[0248] It can be seen that although solution A performs well in transportation efficiency, solution B has more advantages in coal transportation cost and equipment failure risk. Therefore, the comprehensive evaluation index of solution B is higher, indicating that solution B has achieved a better balance between the three objectives and is a more preferred solution.

[0249] Therefore, in step S2223, the solution with the largest comprehensive evaluation index should be selected as the optimal solution and the decision-making scheme for the main path. This can ensure the best balance between transportation cost, equipment failure risk and transportation efficiency, thereby achieving overall optimization of redundant path planning.

[0250] By using the weighted sum method to evaluate the comprehensive performance of each solution, it can be ensured that the main path solution finally selected achieves a good balance between multiple objectives. This method can fully consider the weights and importance between different objectives, thereby achieving the optimal combination of the system in terms of transportation cost, equipment failure risk and transportation efficiency, and ensuring the overall optimization of redundant path planning.

[0251] Step S2224: According to the main path plan, determine the node sequence that the main path passes through to form a physical topology structure of the main path.

[0252] Furthermore, if Figure 6 As shown, step S2224 includes:

[0253] Step S22241, initialize the node sequence N of the main path main is an empty collection;

[0254] Step S22242, select the starting point s' of the main path as the current node and add it to N main middle;

[0255] Step S22243: for the current node s', find the path selection variable x that satisfies the main path solution s't' =1 next node t';

[0256] Step S22244, add node t' to N main and set node t' as the new current node;

[0257] Step S22245, repeating steps S22243 and S22244 until reaching the end of the main path;

[0258] Step S22246: Set the node sequence N of the main path main The physical topology that serves as the primary path.

[0259] Specifically, the purpose of step S2224 is to gradually determine the node sequence passed by the main path according to the path selection rules through the given main path plan, and finally form the physical topology of the path. This process mainly relies on the path selection algorithm in graph theory to help determine the optimal or agreed path in a complex network. The physical topology of the main path shows the specific node arrangement of the path, which is crucial for many engineering applications (such as communication networks, transportation systems, coal transportation paths, etc.).

[0260] First, define an empty set N main , used to store the node sequence that the main path passes through. Set N main It will be gradually filled as the path selection process progresses, eventually forming a complete node sequence. Initially, it is empty, indicating that no nodes of the main path have been determined. By initializing it to an empty set, a clean starting point can be provided for path construction, avoiding interference from any existing data or errors, and ensuring the smooth progress of subsequent steps.

[0261] In the process of path construction, a starting point (s') is first selected. In network topology, the starting point is the starting node of the main path and is usually the initial condition for path selection. This node is selected and added to N main, which means that the path starts from this node and will start from this node in subsequent steps and move forward step by step until it reaches the end. After determining the starting point, a clear starting position can be provided for path construction, making subsequent path construction more orderly and logically clear. For example, in a coal transportation system, the starting point may be the loading area of ​​a coal mine, and the starting point of the path selection means starting to plan the transportation route of coal.

[0262] The key to step S22243 is the path selection variable x s't' This variable is a decision variable used to indicate whether there is a valid path between the current node s' and the candidate node t'. Specifically, x s't' The value of 1 indicates that there is an optional path from node s' to node t'. s't' 0 means no path is available. By judging the path selection variable, we can decide which path to take from the current node s' to the next node t'. By using the path selection variable, we can ensure that the path construction follows the given path planning rules. For example, in a complex logistics system, the path selection variable can ensure that coal is accurately transported from one warehouse to another without deviating from the predetermined route. Assume that the current node is warehouse A, and the candidate nodes include warehouses B, C, and D. According to the path selection rule, if x AB =1 and x AC =0,x AD =0, then the next node will be B.

[0263] Once the next node t' that meets the path selection criteria is found, add it to N main , indicating that node t' is part of the main path. At the same time, t' is updated to the current node so that the subsequent nodes can be continued to be found in subsequent steps. This process is iterative, and each time a new node is determined, the value of the current node is updated and it continues to move forward until the end node is reached. This step ensures that the construction of the path is continuous, and each step is pushed from one node to the next node, eventually forming a complete path from the starting point to the end point. For example, in the process of coal transportation, this step helps ensure the continuity of the transportation path so that each coal transportation can be carried out on the correct path. For example, if warehouse B is selected as the next node in step S22243, then warehouse B will be added to the node sequence N main and becomes the new current node. Then continue to search for the next feasible node from warehouse B.

[0264] Step S22245 is the core part of path construction. It gradually adds each node on the main path to N by repeatedly executing steps S22243 and S22244. mainuntil it reaches the destination node. Each time, a next node that meets the path selection criteria is selected and added to the path sequence until no new nodes can be selected. This iterative process ensures that path selection is performed step by step, and each selection is based on the result of the previous step, so as to ensure that the final path is in line with the predetermined plan. It can also dynamically adapt to different path selection situations in complex networks. For example, suppose the path starts from warehouse A, passes through nodes B, C, D, and finally reaches destination Z. Each time, according to the path selection rules, starting from the current node, the next node is selected in turn until Z is reached.

[0265] In step S22246, the node sequence N of the main path main has been completed and contains all the nodes that the path passes through. This node sequence represents the physical topology of the main path. The topology structure represents the connection relationship between each node in the network and reflects the actual form of the path. In practical applications such as coal transportation, this physical topology structure can be used for subsequent tasks such as path planning, resource scheduling, and system optimization. By obtaining the physical topology structure, support can be provided for subsequent decision-making, such as optimizing resource allocation, improving transportation efficiency, and reducing costs. Path planning is not just a theoretical issue. The actual physical topology structure can directly affect the efficiency and effectiveness of actual operations. For example, assuming that the final path selection is the transportation path from the coal mine to the warehouse, N main The node sequence can be used to construct the topological structure of the entire transportation network, thereby optimizing the logistics system of coal transportation.

[0266] Step S2224 completes the main path determination process step by step through a series of precise operations, from initializing the node sequence to determining the physical topology. s't' The application of the method, each step of the path selection has been rigorously calculated and planned. The beneficial effect of this method is that it can accurately construct a feasible and optimal path based on a given path planning solution, providing a solid foundation for subsequent path optimization, resource allocation, network management, etc.

[0267] Step S2230, using the main path solution, calculate the transportation risk and determine whether a backup path is needed; if a backup path is not needed, the main path forms a redundant path set; if a backup path is needed, select a backup path from the remaining Pareto efficient solution set after eliminating the main path solution;

[0268] Furthermore, if Figure 7 As shown, step S2230 includes:

[0269] Step S2231, select variable x according to the path in the main path plan i'j', determine the path set E that the main path passes through main ;x i'j' Indicates whether to select the path (i', j') from node i' to node j'. If so, then x i'j' =1, otherwise x i'j' =0;

[0270] Step S2232, for each path (i', j')∈E on the main path main , according to the flow distribution ratio f of path (i', j') i'j' , allocate the predicted transport volume TR to the path (i', j'), and obtain the transport volume TR allocated to the path (i', j') i'j' ;

[0271] Step S2233, for each path (i', j')∈E on the main path main , combined with the equipment failure probability p of path (i', j') i'j' Economic loss caused by failure i'j' , and the transport volume TR allocated to the path (i', j') i'j' , calculate the transportation risk R of path (i',j') i'j' ;

[0272] Step S2234: sum up the transportation risks of all paths on the main path to obtain the total transportation risk R of the main path. main ;

[0273] Step S2235: The total transportation risk R of the main path main Compared with the preset risk threshold R th Compare; if R main >R th , then a backup path needs to be added; otherwise, no backup path is needed.

[0274] Specifically, the core task of step S2230 is to calculate the transportation risk based on the main path plan and determine whether to add an alternative path. This process involves a detailed assessment of the transportation risk of the main path to ensure that the coal transportation is carried out under controllable risks. If the calculated risk exceeds the preset threshold, it is necessary to add an alternative path to disperse the risk. Otherwise, the risk of the main path is within an acceptable range and there is no need to add an alternative path.

[0275] In step S2231, firstly, variable x is selected according to the path in the main path scheme. i'j' , determine the path set E that the main path passes through main . Path selection variable x i'j' is a binary variable indicating whether a path is selected as part of the main path.

[0276] x i'j' is a binary variable used to decide whether a path belongs to the main path. i'j' = 1, then the path (i', j') is selected as part of the main path; if x i'j' =0, the path (i', j') is not selected as the main path.

[0277] For example, suppose there are two paths connecting node i' and node j', path 1 and path 2. In path selection, if x i'j' = 1 and x of path 2 i'j' = 0, the main path plan will select path 1. The path selection variable helps determine which paths are used in the main path. i'j' , can clearly define the path set E that the main path passes through main , that is, the set of all paths selected as main paths.

[0278] In step S2232, for each path (i', j')∈E on the main path main , according to the flow distribution ratio f of path (i', j') i'j' , distribute the predicted total transport volume TR to the route. Flow distribution ratio f i'j' Indicates the proportion of the transport volume allocated to each path in the total transport volume. Calculate the transport volume TR allocated to path (i', j') i'j' The formula is: TR i'j' =f i'j' ×TR;

[0279] For example: Assume that the predicted total transport volume TR = 10,000 tons, and the flow distribution ratio of path (i', j') is f i'j' =0.3, then the transport volume allocated to the path (i', j') is:

[0280] TR i'j' =0.3×10000=3000 tons;

[0281] This calculation ensures that the amount of transport allocated on each route matches the total transport demand.

[0282] In step S2233, the device failure probability p of the combined path (i', j') is i'j' , Economic losses caused by equipment failure i'j' , and the transport volume TR allocated to the path (i', j') i'j' , calculate the transportation risk R of path (i',j') i'j' . Transportation Risk R i'j'It is a metric that comprehensively evaluates the possible equipment failures on a path and their economic impact.

[0283] Equipment failure probability p i'j' It indicates the probability of equipment failure on the path (i', j') in a certain period of time in the future. This value is usually estimated through historical data, equipment status monitoring and other means. The higher the probability of equipment failure, the greater the failure risk of the path. For example: if the probability of equipment failure on the path (i', j') p i'j' = 0.1, then the probability of equipment failure on this path in the future is 10%. Economic loss caused by equipment failure L i'j' It reflects the economic impact of equipment failure on path (i', j'), including repair costs, profit loss caused by transportation delays, etc.

[0284] The transportation risk R of path (i', j') i'j' It is the comprehensive result of equipment failure probability, transportation volume and economic losses caused by failure. Its calculation formula is:

[0285] R i'j' =p i'j' ×TR i'j' ×L i'j' ;

[0286] This formula represents the failure risk of the transportation task on the path (i', j'), taking into account the probability of equipment failure, transportation volume and economic losses.

[0287] In step S2234, the transportation risk R of all paths on the main path is calculated. i'j' Sum and get the total transportation risk R of the main path main The total transportation risk of the main path is the accumulation of the path transportation risks on all main paths, indicating the failure risk of the entire main path. Risk threshold R th It is a preset standard that represents the maximum acceptable value of transportation risk. The threshold can be set based on historical data, risk preferences, or system risk management strategies. For example, the risk threshold can be set to 80% of the maximum monthly transportation risk in the past year. For example: If the maximum monthly transportation risk in the past year is 40,000,000 yuan, the risk threshold R can be set to 80% of the maximum monthly transportation risk in the past year. th Set at 32,000,000 yuan.

[0288] If R main >R th , it means that the transportation risk of the main path exceeds the acceptable range, and it is necessary to add a backup path to disperse the risk and ensure the stable operation of the system.

[0289] If R main ≤R th, the transportation risk of the main path is controllable and no backup path is needed.

[0290] By calculating and evaluating the transportation risk of the main route, the system can identify and respond to potential risks during transportation. If the risk of the main route exceeds the acceptable range, adding an alternative route can effectively disperse the risk and avoid transportation interruptions caused by equipment failure or other reasons. In addition, if the risk is controllable, avoid adding alternative routes, save resources and improve efficiency. In this way, the risk management capability of the coal transportation system has been improved, thereby ensuring the smooth completion of the transportation task.

[0291] Step S2240, selecting a backup path from the remaining Pareto efficient solution set after eliminating the main path solution;

[0292] Furthermore, if Figure 8 As shown, step S2240 includes:

[0293] Step S2241, removing the optimal solution corresponding to the main path from the Pareto efficient solution set to form a set of candidate paths;

[0294] Step S2242, calculate the transportation risk R of each alternative path in the alternative path set b , selecting the first N1 solutions with the lowest transportation risk in the alternative path set to form a first alternative path set, wherein the first alternative path set includes N1 first alternative paths;

[0295] Step S2243, calculating the similarity SI between the first backup path and the primary path k , SI k Indicates the similarity between the kth first backup path and the main path; 1≤k≤N1;

[0296] Step S2244, based on the similarity SI k The first backup paths are sorted in ascending order, and the first N2 first backup paths are selected as final backup paths; N2≤N1.

[0297] Step S2250: Combine the primary path and the backup path to form a redundant path set.

[0298] Specifically, the goal of step S2240 is to select a suitable backup path from a set of optimized paths that meet the Pareto efficiency condition. The selection of the backup path needs to ensure that the transportation risk is minimized and has a low similarity with the main path in order to disperse the risk and improve the robustness of the system and the ability to respond to emergencies.

[0299] In multi-objective optimization, the Pareto efficient solution set represents the optimal compromise solution between all optimization objectives. In this process, the optimal solution corresponding to the main path (i.e. the main path itself) is first removed from these Pareto efficient solutions, leaving other candidate paths to form a set of alternative paths. This is done to ensure that the selected backup paths do not overlap with the main path, so that redundancy can be effectively provided and risks can be dispersed. By eliminating the main path, the diversity of the backup path set is ensured, thereby improving the robustness of the system. The diversity of backup paths can help deal with various emergencies, such as situations where the main path is blocked by external factors such as equipment failure and weather effects.

[0300] In the set of alternative paths, the paths need to be screened based on transportation risks. The same method as step S2230 is used to calculate the transportation risks of the alternative paths. By calculating the transportation risks of each alternative path, the top N1 paths with the lowest risks are selected as the first set of alternative paths. The transportation risks of these paths are low, which means that they are relatively stable and reliable, and can ensure the safety of transportation. Selecting the path with the lowest transportation risk can minimize the unexpected situations that may occur during the path selection process. Compared with other paths, these paths can better ensure the safety, stability and efficiency of transportation. For example, if the main path cannot be used due to some factors, selecting a path with low transportation risk can greatly reduce the probability of accidents and ensure the smooth progress of the transportation task.

[0301] The calculation of similarity in step S2243 is to measure the degree of overlap between the backup path and the main path, and the node overlap index is usually used to measure the similarity of the paths. Node overlap refers to the ratio of the number of nodes shared between the first backup path and the main path to the total number of nodes in the path. The lower the path overlap, the higher the degree of difference between the first backup path and the main path, which is conducive to risk dispersion. The purpose of calculating the similarity is to further select backup paths that are more different from the main path. By selecting paths with lower similarity, the risks that may arise during transportation can be effectively dispersed. For example, if the main path cannot be used due to traffic congestion, selecting a backup path with low node overlap with the main path can avoid failures under the influence of the same problem and enhance the resilience of the transportation system. For example: If the main path passes through nodes A, B, C, and D, and a backup path passes through nodes B, C, E, and F, the node overlap is This means that the backup path has 33% node overlap with the main path. If the backup path has a low overlap, it means that the backup path is geographically different from the main path, so it can provide an effective alternative when the main path is blocked.

[0302] After calculating the similarity between each first backup path and the main path, the backup paths are sorted from low to high according to the similarity. The first N2 backup paths with the lowest similarity are selected as the final backup paths. These paths have low node overlap, so they can effectively avoid the impact of main path problems and reduce transportation risks. By selecting paths with low node overlap as backup paths, risks can be dispersed and the system's response capabilities can be enhanced. Selecting the path with the lowest similarity means that the backup path can provide a more effective alternative when the main path is blocked, thereby improving the flexibility and anti-interference ability of the coal transportation system.

[0303] Once the primary path and the best backup path are selected, the primary path and the selected backup path can be combined to form a redundant path set. A redundant path set refers to a complete path set that includes the primary path and all backup paths. Each path should be assigned a unique identification number and sorted from low to high according to transportation risk, with paths with lower risks being used first. The redundant path set provides multiple options for coal transportation, ensuring that if an accident occurs on any path, there are still other paths that can continue to be used. By sorting the paths by risk, it can ensure that the safest and most stable path is given priority when an abnormality occurs, greatly improving the stability and emergency response capabilities of the system. Assume that the transportation risk of the primary path is lower, the risk of the first backup path is slightly higher, and the risk of the second backup path is higher. In the actual transportation process, if a problem occurs on the primary path, the backup path with the lowest risk is given priority to ensure that the transportation task will not be interrupted.

[0304] Step S3000: performing intelligent flow control based on a redundant path set.

[0305] Furthermore, if Fig. 9 As shown, step S3000 includes:

[0306] Step S3100, collecting the coal flow of the main path and the coal flow of the backup path in real time;

[0307] Step S3200, calculating the ratio of the coal flow rate of the main path to the coal flow rate of the backup path to obtain the flow balance;

[0308] Step S3300, determining whether the flow balance exceeds the balance threshold, and if so, adjusting the coal flow distribution.

[0309] Specifically, the purpose of step S3000 is to dynamically adjust the flow of the main path and the backup path in the coal transmission system through intelligent flow control based on a redundant path set to ensure balanced distribution of flow and improve the stability and operation efficiency of the system. The key to intelligent flow control is to monitor the flow in real time and adjust the flow according to the preset flow balance threshold to ensure that the load of the main path and the backup path remains within a reasonable range.

[0310] First, the system needs to collect coal flow on the main and backup paths in real time. Coal flow here refers to the amount of coal transported per unit time, usually measured in tons / hour or other suitable units. The main path is the main transportation route, while the backup path is used to share the transportation load, especially when the flow on the main path is too large, the backup path will play an important role. In order to collect flow data in real time, it is usually necessary to deploy flow monitoring equipment or sensors, which can accurately measure the amount of coal passing through each path. For example, by monitoring the traffic flow of rail or road transportation through sensors, combined with GPS and other positioning systems, flow data can be accurately calculated. By monitoring the flow in real time, the load situation of the main and backup paths can be grasped in time. If the main path shows signs of overload, measures can be taken quickly to allocate part of the flow to the backup path to avoid problems such as transportation delays or system overload. For example, if the transportation volume of the main path suddenly increases, timely acquisition of flow data can enable the backup path to respond more quickly, share the flow, and avoid congestion or failure.

[0311] The goal of step S3200 is to calculate the flow balance, that is, the ratio of the main path flow to the backup path flow, and determine the load balance between them. The flow balance is an indicator to measure the rationality of the path load distribution. If the flow of the main path is much larger than the flow of the backup path, the flow balance will be higher, indicating that the main path is overloaded and the backup path is not fully utilized. The calculation of the flow balance provides a quantitative basis for subsequent flow adjustments. By calculating the balance in real time, it is possible to promptly detect the imbalance of path loads and ensure reasonable flow distribution. For example, when the flow balance is high, the system can automatically adjust the flow based on this information, directing part of the flow from the main path to the backup path to avoid overloading the main path, thereby improving transportation efficiency and ensuring smooth operation of the system.

[0312] Step S3300 compares the calculated flow balance with the preset balance threshold. If the flow balance exceeds the set balance threshold, it means that the load on the main path is too heavy, while the load on the backup path is light. At this time, the system needs to adjust the distribution of coal flow. If the flow balance exceeds the balance threshold, it means that the load on the main path is too heavy. The system must adjust the flow and allocate more flow to the backup path to reduce the burden on the main path. If the flow balance is within a reasonable range, it means that the flow distribution of the main path and the backup path is relatively balanced and no adjustment is required. By setting a balance threshold and making timely judgments, the system can automatically respond when the load is uneven, avoid overloading a single path, and thus improve the stability of the transportation system. Assuming that during the peak transportation period, the main path is overloaded, the system can automatically identify and guide the backup path to share part of the flow, avoiding the risk of overloading the main path and ensuring transportation efficiency.

[0313] When the flow balance exceeds the threshold, the coal flow of the main path and the backup path needs to be dynamically adjusted to make the load between them more balanced. The adjustment formula can be made according to the following method:

[0314]

[0315] Among them, F main is the coal flow of the main path, is the coal flow of the main path after adjustment, F backup is the coal flow of the alternative path, is the coal flow of the adjusted backup path; Δf is the flow adjustment ratio, which is a parameter that is dynamically adjusted according to actual needs. For example, when the main path flow is overloaded, Δf can be set to 0.1, that is, the main path flow is reduced by 10% and the backup path flow is increased by 10%. The adjusted flow distribution will make the flow of the main path and the backup path more balanced, thereby improving the stability of the coal transportation system. This adjustment formula provides a dynamic flow distribution mechanism. Through the flow adjustment ratio Δf, it is possible to flexibly respond to flow changes in different situations. The adjusted flow will make the flow of the main path and the backup path more balanced, thereby improving the stability of the system. For example, when there is traffic congestion on the main path, this formula can be used to transfer the flow from the main path to the backup path, reduce the burden on the main path, and ensure that coal transportation is not interrupted.

[0316] Once the traffic adjustment is completed, the system will continue to monitor the traffic changes of the main path and the backup path in real time. If the adjusted traffic distribution still does not reach an acceptable balance range, the system will enter a cyclic adjustment phase to continuously optimize the traffic distribution until the ideal balance state is reached. This process ensures the continuity and stability of traffic adjustment. Continuous monitoring and adjustment can ensure that the system is always in the optimal operating state and avoid overload or waste of resources on any single path. Through cyclic adjustment, the system can adaptively adjust the traffic distribution to ensure that the load of the main path and the backup path is always kept within a reasonable range, improving the overall transportation efficiency and system reliability. For example, when multiple peak periods occur alternately, the system can adjust the traffic according to real-time data at each moment to avoid transportation delays caused by uneven path loads.

[0317] Step S3000 uses intelligent flow control based on a set of redundant paths to ensure the stability and efficiency of the coal transportation system by real-time monitoring, calculating flow balance, and adjusting flow distribution when necessary. Through a reasonable flow adjustment formula and a cyclic optimization mechanism, it can dynamically respond to changes in transportation loads and ensure that the system can provide smooth and efficient services under different circumstances. The beneficial effects of real-time monitoring and intelligent adjustment not only improve transportation efficiency, but also reduce risks and failures caused by uneven flow, ensuring the smooth completion of coal transportation tasks.

[0318] Example 2

[0319] Based on Example 1, this embodiment provides a redundant path design system for improving the stability of the coal transmission system. Fig.10 As shown, including:

[0320] Model building module: obtaining historical operation data of the coal transmission system, and building a first data set based on the historical operation data; building and training an equipment failure prediction model and a transportation demand prediction model based on the first data set;

[0321] Prediction module: uses the trained equipment failure prediction model and transportation demand prediction model to predict the equipment health status and transportation demand within a preset time to obtain the first prediction result and the second prediction result;

[0322] Multi-objective optimization module: obtaining the network topology of the coal transportation system, building a redundant path multi-objective optimization model based on the first prediction result and the second prediction result, as well as the network topology of the coal transportation system; obtaining a redundant path set according to the redundant path multi-objective optimization model;

[0323] Control module: Intelligent flow control based on a set of redundant paths.

[0324] In the model building module, the historical operation data includes historical equipment operating parameters and historical coal transportation volume data; the historical equipment operating parameters and historical coal transportation volume data are sorted by time to form a historical equipment operating parameter sequence and a historical coal transportation volume data sequence; the construction of the first data set includes:

[0325] Step S1210, statistically analyzing the historical equipment operating condition parameter sequence, extracting historical health status characteristics, and forming a historical health status characteristic sequence;

[0326] Step S1220, performing time series analysis on the coal transportation volume data sequence, extracting historical transportation demand indicators, and forming a historical transportation demand indicator sequence;

[0327] Step S1230, combining the historical health status feature sequence and the historical transportation demand index sequence to form a first data set.

[0328] In the model building module, the building and training of the equipment failure prediction model and the transportation demand prediction model includes:

[0329] Step S1310, using a time series prediction algorithm to build an equipment failure prediction model based on a historical health status feature sequence;

[0330] Step S1320, using a regression prediction algorithm, based on the historical transportation demand indicator sequence, constructs a transportation demand prediction model.

[0331] In the prediction module, obtaining the first prediction result and the second prediction result includes:

[0332] Step S1410, obtaining a real-time equipment operating condition parameter sequence, obtaining a real-time health status feature sequence according to the real-time equipment operating condition parameter sequence, inputting the real-time health status feature sequence into a trained equipment failure prediction model, and obtaining a first prediction result;

[0333] Step S1420, obtaining a real-time coal transportation volume data sequence, obtaining a real-time transportation demand index sequence according to the real-time coal transportation volume data sequence, inputting the real-time transportation demand index sequence into a trained transportation demand prediction model, and obtaining a second prediction result.

[0334] The first prediction result includes the equipment failure rate FA within a preset time, and the second prediction result includes the predicted transportation volume TR within a preset time.

[0335] In the multi-objective optimization module, the construction of the redundant path multi-objective optimization model includes:

[0336] Step S2100, obtaining the network topology of the coal transportation system, and constructing a redundant path multi-objective optimization model based on the equipment failure rate FA and the predicted transportation volume TR within a preset time, as well as the network topology of the coal transportation system; the network topology of the coal transportation system includes a path set and a node set;

[0337] The step S2100 includes:

[0338] Step S2110, using the equipment failure rate FA and the predicted transportation volume TR within a preset time as constraint conditions of the redundant path multi-objective optimization model;

[0339] Step S2120, defining decision variables of the redundant path multi-objective optimization model, including path selection variables and traffic distribution ratio;

[0340] Step S2130, determining the objective function of the redundant path multi-objective optimization model, including minimizing the coal transportation cost, minimizing the risk of equipment failure, and maximizing the transportation efficiency;

[0341] Step S2140, constructing a redundant path multi-objective optimization model according to the path set and the node set in the network topology structure of the coal transportation system, and the decision variables and the objective function of the redundant path multi-objective optimization model.

[0342] In the multi-objective optimization module, the redundant path set is obtained according to the redundant path multi-objective optimization model, including:

[0343] Step S2210, solving the redundant path multi-objective optimization model to obtain a Pareto effective solution set; each solution in the Pareto effective solution set corresponds to a redundant path solution;

[0344] Step S2220, selecting a main path from the Pareto efficient solution set;

[0345] Step S2230, using the main path solution, calculate the transportation risk and determine whether a backup path is needed; if a backup path is not needed, the main path forms a redundant path set; if a backup path is needed, select a backup path from the remaining Pareto efficient solution set after eliminating the main path solution;

[0346] Step S2240, selecting a backup path from the remaining Pareto efficient solution set after eliminating the main path solution;

[0347] Step S2250: Combine the primary path and the backup path to form a redundant path set.

[0348] The step S2220 includes:

[0349] Step S2221, calculate the normalized value y of each solution in the Pareto efficient solution set on each targetl,g , the objectives include coal transportation cost, equipment failure risk and transportation efficiency, y l,g represents the normalized value of the lth solution in the Pareto efficient solution set on the gth target; 1≤g≤m, m is the total number of targets; l is the index of the solution in the Pareto efficient solution set;

[0350] Step S2222, set the target weight vector w=[w1,w2,…,w m ],w m is the weight of the mth target; the weighted summation method is used, based on the target weight vector w and the normalized value y of each solution on each target l,g , calculate the comprehensive evaluation index for each solution; where w1+w2+…+w m =1;

[0351] Step S2223, taking the solution with the maximum comprehensive evaluation index as the optimal solution, and selecting the redundant path solution corresponding to the optimal solution as the main path solution;

[0352] Step S2224: According to the main path plan, determine the node sequence that the main path passes through to form a physical topology structure of the main path.

[0353] The step S2224 includes:

[0354] Step S22241, initialize the node sequence N of the main path main is an empty collection;

[0355] Step S22242, select the starting point s' of the main path as the current node and add it to N main middle;

[0356] Step S22243: for the current node s', find the path selection variable x that satisfies the main path solution s't' =1 next node t';

[0357] Step S22244, add node t' to N main and set node t' as the new current node;

[0358] Step S22245, repeating steps S22243 and S22244 until reaching the end of the main path;

[0359] Step S22246: Set the node sequence N of the main path main The physical topology that serves as the primary path.

[0360] The step S2230 includes:

[0361] Step S2231, select variable x according to the path in the main path plani'j' , determine the path set E that the main path passes through main ;x i'j' Indicates whether to select the path (i', j') from node i' to node j'. If so, then x i'j' =1, otherwise x i'j' =0;

[0362] Step S2232, for each path (i', j')∈E on the main path main , according to the flow distribution ratio f of path (i', j') i'j' , allocate the predicted transport volume TR to the path (i', j'), and obtain the transport volume TR allocated to the path (i', j') i'j' ;

[0363] Step S2233, for each path (i', j')∈E on the main path main , combined with the equipment failure probability p of path (i', j') i'j' Economic loss caused by failure i'j' , and the transport volume TR allocated to the path (i', j') i'j' , calculate the transportation risk R of path (i',j') i'j' ;

[0364] Step S2234: sum up the transportation risks of all paths on the main path to obtain the total transportation risk R of the main path. main ;

[0365] Step S2235: The total transportation risk R of the main path main Compared with the preset risk threshold R th Compare; if R main >R th , then a backup path needs to be added; otherwise, no backup path is needed.

[0366] The step S2240 includes:

[0367] Step S2241, removing the optimal solution corresponding to the main path from the Pareto efficient solution set to form a set of candidate paths;

[0368] Step S2242, calculate the transportation risk R of each alternative path in the alternative path set b , selecting the first N1 solutions with the lowest transportation risk in the alternative path set to form a first alternative path set, wherein the first alternative path set includes N1 first alternative paths;

[0369] Step S2243, calculating the similarity SI between the first backup path and the primary path k , SI kIndicates the similarity between the kth first backup path and the main path; 1≤k≤N1;

[0370] Step S2244, based on the similarity SI k The first backup paths are sorted in ascending order, and the first N2 first backup paths are selected as final backup paths; N2≤N1.

[0371] In the control module, the intelligent flow control based on the redundant path set includes:

[0372] Step S3100, collecting the coal flow of the main path and the coal flow of the backup path in real time;

[0373] Step S3200, calculating the ratio of the coal flow rate of the main path to the coal flow rate of the backup path to obtain the flow balance;

[0374] Step S3300, determining whether the flow balance exceeds the balance threshold, and if so, adjusting the coal flow distribution.

[0375] Example 3

[0376] This embodiment discloses an electronic device, which may include one or more processors and one or more memories, wherein the memories store computer-readable codes, and when the computer-readable codes are executed by the one or more processors, the redundant path design method for improving the stability of the coal transmission system as described above may be executed.

[0377] The method or system according to the embodiment of the present application can also be implemented with the help of the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, etc. The storage device in the electronic device, such as a ROM or a hard disk, can store the redundant path design method for improving the stability of the coal transmission system provided in the present application. The redundant path design method for improving the stability of the coal transmission system may, for example, include: obtaining historical operation data of the coal transmission system, and constructing a first data set based on the historical operation data; constructing and training an equipment failure prediction model and a transportation demand prediction model based on the first data set; using the trained equipment failure prediction model and the transportation demand prediction model, predicting the equipment health status and transportation demand within a preset time, and obtaining a first prediction result and a second prediction result; obtaining the network topology of the coal transportation system, and constructing a redundant path multi-objective optimization model based on the first prediction result and the second prediction result, as well as the network topology of the coal transportation system; obtaining a redundant path set according to the redundant path multi-objective optimization model; and performing intelligent flow control based on the redundant path set.

[0378] Furthermore, the electronic device may also include a user interface. Of course, the architecture disclosed in the present invention is only exemplary, and when implementing different devices, one or more components in the electronic device disclosed in the present invention may be omitted according to actual needs.

[0379] Example 4

[0380] This embodiment discloses a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the redundant path design method for improving the stability of a coal transmission system in the embodiment of the present application can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0381] In addition, according to the implementation of the present application, the process described in the above reference flow chart can be implemented as a computer software program. For example, the present application provides a non-temporary machine-readable storage medium, the non-temporary machine-readable storage medium stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: obtaining historical operation data of the coal transmission system, and constructing a first data set based on the historical operation data; constructing and training an equipment failure prediction model and a transportation demand prediction model based on the first data set; using the trained equipment failure prediction model and transportation demand prediction model, predicting the equipment health status and transportation demand within a preset time, and obtaining a first prediction result and a second prediction result; obtaining the network topology of the coal transportation system, and constructing a redundant path multi-objective optimization model based on the first prediction result and the second prediction result, as well as the network topology of the coal transportation system; obtaining a redundant path set based on the redundant path multi-objective optimization model; and performing intelligent flow control based on the redundant path set. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0382] The methods, systems, and devices of the present application may be implemented in many ways. For example, the methods, systems, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0383] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0384] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A redundant path design method for improving the stability of a coal transmission system, characterized in that: The method comprises: Acquire historical operation data of the coal transmission system, and construct a first data set based on the historical operation data; construct and train an equipment failure prediction model and a transportation demand prediction model based on the first data set; use the trained equipment failure prediction model and transportation demand prediction model to predict the equipment health status and transportation demand within a preset time, and obtain a first prediction result and a second prediction result; The network topology structure of the coal transportation system is obtained, and a redundant path multi-objective optimization model is constructed based on the first prediction result and the second prediction result, as well as the network topology structure of the coal transportation system; a redundant path set is obtained according to the redundant path multi-objective optimization model; and intelligent flow control is performed based on the redundant path set.

2. The redundant path design method for improving the stability of the coal transmission system according to claim 1 is characterized in that: The historical operation data includes historical equipment operating parameters and historical coal transportation volume data; the historical equipment operating parameters and historical coal transportation volume data are sorted by time to form a historical equipment operating parameter sequence and a historical coal transportation volume data sequence; The constructing the first data set comprises: Perform statistical analysis on the historical equipment condition parameter sequence, extract the historical health status characteristics, and form the historical health status characteristic sequence; Conduct time series analysis on the coal transportation volume data series, extract historical transportation demand indicators, and form a historical transportation demand indicator series; The historical health status feature sequence and the historical transportation demand index sequence are combined to form a first data set.

3. The redundant path design method for improving the stability of the coal transmission system according to claim 2 is characterized in that: The construction and training of the equipment failure prediction model and the transportation demand prediction model includes: Using the time series prediction algorithm, we build an equipment failure prediction model based on the historical health status feature sequence; using the regression prediction algorithm, we build a transportation demand prediction model based on the historical transportation demand indicator sequence; The first prediction result includes the equipment failure rate FA within a preset time, and the second prediction result includes the predicted transportation volume TR within a preset time.

4. The redundant path design method for improving the stability of the coal transmission system according to claim 3 is characterized in that: The construction of the redundant path multi-objective optimization model comprises: Acquire a network topology structure of a coal transportation system, wherein the network topology structure of the coal transportation system includes a path set and a node set; The equipment failure rate FA and the predicted transportation volume TR within the preset time are used as the constraints of the redundant path multi-objective optimization model; Defining decision variables of a redundant path multi-objective optimization model, wherein the decision variables include a path selection variable and a traffic distribution ratio; Determining an objective function of a redundant path multi-objective optimization model, wherein the objective function includes minimizing coal transportation costs, minimizing equipment failure risks, and maximizing transportation efficiency; According to the path set and node set in the network topology structure of the coal transportation system, as well as the decision variables and objective function of the redundant path multi-objective optimization model, a redundant path multi-objective optimization model is constructed.

5. The redundant path design method for improving the stability of the coal transmission system according to claim 4 is characterized in that: The method of obtaining a redundant path set according to the redundant path multi-objective optimization model includes: Solve the redundant path multi-objective optimization model and obtain the Pareto efficient solution set; each solution in the Pareto efficient solution set corresponds to a redundant path solution; Select the main path from the Pareto efficient solution set; Using the main path plan, calculate the transportation risk and determine whether an alternative path is needed; if an alternative path is not needed, the main path forms a redundant path set; if an alternative path is needed, select the alternative path from the remaining Pareto efficient solution set after eliminating the main path plan; combine the main path and the alternative path to form a redundant path set.

6. The redundant path design method for improving the stability of the coal transmission system according to claim 5, characterized in that: The selecting of the main path from the Pareto efficient solution set comprises: Calculate the normalized value y of each solution in the Pareto efficient solution set on each objective l,g , the objectives include coal transportation cost, equipment failure risk and transportation efficiency, y l,g represents the normalized value of the lth solution in the Pareto efficient solution set on the gth target; 1≤g≤m, m is the total number of targets; l is the index of the solution in the Pareto efficient solution set; Set the target weight vector w = [w1,w2,…,w m ],w m is the weight of the mth target; the weighted summation method is used, based on the target weight vector w and the normalized value y of each solution on each target l,g , calculate the comprehensive evaluation index for each solution; where w1+w2+…+w m =1; The solution with the largest comprehensive evaluation index is taken as the optimal solution, and the redundant path solution corresponding to the optimal solution is selected as the main path solution; According to the main path plan, the node sequence that the main path passes through is determined to form the physical topology structure of the main path.

7. The redundant path design method for improving the stability of the coal transmission system according to claim 6 is characterized in that: The physical topology structure forming the main path includes: Step S22241, initialize the node sequence N of the main path main is an empty collection; Step S22242, select the starting point s' of the main path as the current node and add it to N main middle; Step S22243: for the current node s', find the path selection variable x that satisfies the main path solution s't' =1 next node t'; Step S22244, add node t' to N main and set node t' as the new current node; Step S22245, repeating steps S22243 and S22244 until reaching the end of the main path; Step S22246: Set the node sequence N of the main path main The physical topology that serves as the primary path.

8. The redundant path design method for improving the stability of the coal transmission system according to claim 7, characterized in that: The determining whether a backup path is needed includes: Select variable x according to the path in the main path scheme i'j' , determine the path set E that the main path passes through main ;x i'j' Indicates whether to select the path (i', j') from node i' to node j'. If so, then x i'j' =1, otherwise x i'j' =0; For each path (i', j')∈E on the main path main , according to the flow distribution ratio f of path (i', j') i'j' , allocate the predicted transport volume TR to the path (i', j'), and obtain the transport volume TR allocated to the path (i', j') i'j' ; For each path (i', j')∈E on the main path main , combined with the equipment failure probability p of path (i', j') i'j' Economic loss caused by failure i'j' , and the transport volume TR allocated to the path (i', j') i'j' , calculate the transportation risk R of path (i',j') i'j' ; Sum the transportation risks of all paths on the main path to get the total transportation risk R of the main path main ; The total transportation risk R of the main path main Compared with the preset risk threshold R th Compare; if R main >R th , then an alternative path is required; otherwise, no alternative path is required.

9. The redundant path design method for improving the stability of the coal transmission system according to claim 8, characterized in that: The step of selecting a backup path from the remaining Pareto efficient solution set after removing the main path solution comprises: Eliminate the optimal solution corresponding to the main path from the Pareto efficient solution set to form a set of alternative paths; Calculate the transportation risk R of each alternative path in the alternative path set b , selecting the first N1 solutions with the lowest transportation risk in the alternative path set to form a first alternative path set, wherein the first alternative path set includes N1 first alternative paths; Calculate the similarity SI between the first backup path and the main path k , SI k Indicates the similarity between the kth first backup path and the main path; 1≤k≤N1; According to the similarity SI k The first backup paths are sorted in ascending order, and the first N2 first backup paths are selected as final backup paths; N2≤N1.

10. A redundant path design system for improving the stability of a coal transmission system, which is used to implement the redundant path design method for improving the stability of a coal transmission system according to any one of claims 1 to 9, characterized in that: The system comprises: Model building module: acquiring historical operation data of the coal transmission system, and building a first data set based on the historical operation data; building and training an equipment failure prediction model and a transportation demand prediction model based on the first data set; Prediction module: uses the trained equipment failure prediction model and transportation demand prediction model to predict the equipment health status and transportation demand within a preset time to obtain the first prediction result and the second prediction result; Multi-objective optimization module: obtaining the network topology of the coal transportation system, building a redundant path multi-objective optimization model based on the first prediction result and the second prediction result, as well as the network topology of the coal transportation system; obtaining a redundant path set according to the redundant path multi-objective optimization model; Control module: Intelligent flow control based on a set of redundant paths.

Images