Decision-making method for joint reserve of emergency rescue materials by coal mining enterprises and sharing centers

Abstract

The present application provides a decision-making method for a coal mine enterprise and a sharing center to jointly reserve emergency rescue materials, the method comprising: for a specific accident type, based on historical accident information, using a random forest algorithm to construct a material demand model for a specific type of emergency rescue materials during the accident emergency rescue period as time passes and the accident evolves, and the demand for emergency rescue materials; obtaining basic information of the coal mine, and according to the basic information, applying the demand model to determine the curve of the change in the demand for emergency rescue materials during the accident as time passes and the accident evolves; according to the requirements of the curve of the change in the demand for emergency rescue materials, when the time and the quantity of materials meet the requirements, determining the storage information of the emergency rescue materials. The present application solves the problems of high reserve cost and low efficiency in the prior art.

Description

Technical Field

[0001] The present application relates to the field of emergency management technology, and specifically, to a decision-making method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials, a decision-making device for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials, a computer program product, and a material storage system. Background Art

[0002] At present, when my country's coal mining enterprises carry out emergency rescue material reserves, they mainly rely on the enterprise's own assessment of potential accident risks, and establish self-reserve warehouses to reserve materials accordingly. However, there are problems in the reserve process such as lack of standards, insufficient regulations and unclear regulations, which leads to differences in material reserve management among different enterprises, resulting in high reserve costs and low efficiency. This situation will greatly affect the effective deployment and use of emergency rescue materials and reduce the speed and effectiveness of accident rescue emergency response. Summary of the invention

[0003] The main purpose of this application is to provide a decision-making method for a coal mine enterprise and a sharing center to jointly reserve emergency rescue materials, a decision-making device for a coal mine enterprise and a sharing center to jointly reserve emergency rescue materials, a computer program product and a material storage system, so as to solve the problems of high reserve costs and low efficiency of coal mines in the prior art.

[0004] In order to achieve the above-mentioned purpose, according to the first aspect of the present application, a decision-making method for a coal mine enterprise and a sharing center to jointly reserve emergency rescue materials is provided, including: for a specific accident type, based on historical accident information, using a random forest algorithm to construct a material demand model for a specific type of emergency rescue materials during the accident emergency rescue period as time passes and the accident evolves, and determining the demand for emergency rescue materials through the material demand model; obtaining basic information of the coal mine, wherein the basic information at least includes one or more of the geological conditions, mine structure, mining method, accident type and characteristic parameters, accident impact range, equipment usage and personnel work conditions of the coal mine, and according to the basic information, applying the demand model to determine the change curve of the demand for emergency rescue materials during the accident as time passes and the accident evolves; according to the requirements of the change curve of the demand for emergency rescue materials, when the time and material quantity meet the requirements, determining the storage information of the emergency rescue materials, wherein the storage information includes one or more of the first reserve of emergency rescue materials of the coal mine, the site selection of the shared base and the second reserve of emergency rescue materials of the shared base, wherein the emergency rescue materials of the shared base can be used by multiple coal mines.

[0005] In order to achieve the above-mentioned purpose, according to the second aspect of the present application, a decision-making device for a coal mine enterprise and a sharing center to jointly reserve emergency rescue materials is provided, comprising: a processing unit, for a specific accident type, based on historical accident information, using a random forest algorithm to construct a material demand model for a specific type of emergency rescue materials during the accident emergency rescue period as time passes and the accident evolves, and the demand for emergency rescue materials is determined by the material demand model; an acquisition unit, for acquiring basic information of the coal mine, wherein the basic information at least includes one or more of the geological conditions, mine structure, mining method, accident type and characteristic parameters, accident impact range, equipment usage and personnel work conditions of the coal mine, and according to the basic information, the demand model is applied to determine the change curve of the demand for emergency rescue materials during the accident as time passes and the accident evolves; a determination unit, for determining the storage information of the emergency rescue materials according to the requirements of the change curve of the demand for emergency rescue materials, when the time and material quantity meet the requirements, wherein the storage information includes one or more of the first reserve of emergency rescue materials of the coal mine, the site selection of the shared base and the second reserve of emergency rescue materials of the shared base, wherein the emergency rescue materials of the shared base can be used by multiple coal mines.

[0006] According to the third aspect of the present application, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned decision-making methods for the coal mining enterprise and the sharing center to jointly reserve emergency rescue materials.

[0007] According to the fourth aspect of the present application, a material storage system is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a decision-making method for executing any one of the above-mentioned coal mining enterprises and sharing centers to jointly reserve emergency rescue materials.

[0008] By applying the technical solution of the present application, based on the random forest algorithm, the demand for specific emergency rescue materials can be accurately predicted, and then a joint storage model between coal mining enterprises and shared reserve centers (i.e., shared bases) can be established based on the predicted demand. In this way, emergency rescue materials can be stored in both coal mines and shared bases, making the decision to store emergency rescue materials more in line with the actual situation of accident rescue, thereby improving the speed and effectiveness of emergency response to accident rescue. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0010] Figure 1 It is a flowchart of a decision-making method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials according to an embodiment of the present application;

[0011] Figure 2 is a schematic diagram of a random forest algorithm according to an embodiment of the present application;

[0012] Figure 3 is a schematic diagram of a Bagging algorithm according to an embodiment of the present application;

[0013] Figure 4 is a schematic diagram of the Boosting algorithm principle according to an embodiment of the present application;

[0014] Figure 5 is a schematic diagram of changes in demand for specific emergency supplies by a coal mining enterprise according to an embodiment of the present application;

[0015] Figure 6 is a schematic diagram of the demand and supply relationship of specific emergency supplies under the shared reserve mode according to an embodiment of the present application;

[0016] Figure 7 is a flow chart of a PSO algorithm according to an embodiment of the present application;

[0017] Figure 8 is a schematic diagram of a shared reserve model for emergency supplies under a shared mode according to an embodiment of the present application;

[0018] Fig. 9 is a flow chart of a method for constructing a shared reserve model according to an embodiment of the present application;

[0019] Fig.10 It is a structural block diagram of a decision-making device for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0021] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] In the operation process of coal mining enterprises, accidents are often unpredictable and uncertain, and these low-probability events may have a significant impact on the normal operation of the enterprise. Due to this uncertainty, enterprises often face decision-making difficulties when stockpiling rescue materials, resulting in a large degree of randomness in the stockpiling of materials. First, in order to cope with possible accidents, enterprises may reserve a large amount of rescue materials, such as medical supplies, life-saving equipment, food and water. However, due to the low probability of accidents, these materials may be idle for a long time, which not only occupies the storage space of the enterprise, but may also gradually expire or be damaged due to long-term non-use, resulting in a waste of resources. Secondly, when an accident occurs, because the types or quantities of the reserved materials do not match the actual needs, or because some key materials have expired or been damaged, the enterprise may find that the existing reserve materials are insufficient to cope with the emergency. In this case, the enterprise needs to quickly mobilize rescue materials from the outside, but due to time constraints and resource constraints, this process may be very difficult, thus affecting the timeliness and effectiveness of the rescue. In addition, due to the lack of an effective material management strategy, the randomness of the enterprise in material reserves may also lead to other problems. For example, companies may overinvest in certain supplies and underinvest in other key supplies, which will further exacerbate the shortage of supplies during rescue operations. At the same time, due to the lack of regular inspection and updating of material reserves, companies may not be able to understand the actual status of materials in a timely manner, making it difficult to make correct judgments and decisions in emergency situations.

[0024] In order to improve the efficiency and effectiveness of accident rescue and reduce the impact of accidents on enterprises and society, it is particularly important for enterprises to establish a scientific material reserve and management system. Based on the statistical analysis of historical accidents and the assessment of accident risks of specific coal mining enterprises, the present invention predicts the demand and investment time of specific materials for enterprises to deal with possible accidents, establishes a joint reserve model between enterprises and shared reserve centers, and provides decision-making solutions. It can be used for rescue at the fastest speed and meet the needs for the quantity of rescue materials in a timely manner. At the same time, on the whole, it can greatly reduce the storage cost of rescue materials and the waste loss caused by long-term idleness.

[0025] At present, when my country's coal mining enterprises are storing emergency rescue materials, they mainly rely on the company's own assessment of potential accident risks, and establish their own reserve warehouses to store materials accordingly.

[0026] As introduced in the background technology, the existing technology has high storage costs and low efficiency. To solve the above problems, the embodiments of the present application provide a decision-making method for a coal mine enterprise and a sharing center to jointly reserve emergency rescue materials, a decision-making device for a coal mine enterprise and a sharing center to jointly reserve emergency rescue materials, a computer program product and a material storage system.

[0027] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0028] In this embodiment, a decision-making method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials is provided, which runs on a mobile terminal, a computer terminal or a similar computing device. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0029] Figure 1 1 is a flowchart of a decision-making method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials according to an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0030] Step S201, for a specific accident type, based on historical accident information, using a random forest algorithm, constructing a material demand model for a specific type of emergency rescue material during the accident emergency rescue period as time passes and the accident evolves, and determining the demand for emergency rescue materials through the material demand model;

[0031] Specifically, the use of random forest algorithm to determine the demand for emergency rescue materials can more accurately predict and plan the reserve of emergency rescue materials. By analyzing historical data and various influencing factors, the random forest algorithm can effectively predict the amount of rescue materials required when a coal mine accident occurs, thereby ensuring that the coal mine can provide the required materials in a timely and effective manner in an emergency, ensuring the smooth handling of the accident, and minimizing losses and casualties. Through scientific algorithm prediction, the decision-making plan for the joint reserve of emergency rescue materials by coal mine enterprises and sharing centers can be better optimized, thereby improving the efficiency and flexibility of emergency rescue work.

[0032] Specific accident types refer to clear and specific accident types, such as coal mine flooding accidents, gas explosion accidents, etc. Each accident type may require different emergency rescue supplies, so the prediction model needs to be customized for different accident types.

[0033] Historical accident information refers to data related to past accidents, including the time, location, scale, and material usage records during the accident handling process. This information is the basis for building a prediction model and is used to train the model so that it can learn the laws and patterns of material demand under specific accident types.

[0034] Step S202, obtaining basic information of the coal mine, wherein the basic information at least includes one or more of the geological conditions of the coal mine, the structure of the mine shaft, the mining method, the accident type and characteristic parameters, the scope of accident impact, the use of equipment and the working conditions of personnel, and according to the basic information, applying the demand model to determine the change curve of the demand for emergency rescue materials during the accident period as time passes and the accident evolves;

[0035] Specifically, basic information about coal mines can help ensure smooth production and operation of coal mines. Understanding the geological characteristics of the area where the coal mines are located can help determine information such as mineral reserves and coal quality, thereby providing basic data for material reserves. Understanding the structure of the mine can help plan the location and method of material reserves to ensure the safe and efficient storage of materials. Different mining methods will have an impact on the storage and use of materials, so it is necessary to adjust the material reserve plan according to the mining method. Understanding the equipment used in coal mines can help determine the types and quantities of material reserves to ensure smooth equipment operation. Understanding the work situation of coal mine personnel can help determine the timing and quantity of material reserves to ensure the normal work of personnel.

[0036] Specifically, basic information refers to all basic information related to a specific type of accident, including the time and location of the accident, the nature of the accident (such as water seepage in coal mines, gas explosions, etc.), the scale and severity of the accident, the basic conditions of the mine (such as geological conditions, mining depth, layout of underground facilities, etc.), and the emergency response situation in the early stage of the accident.

[0037] Specifically, the demand model will calculate and predict how much specific types of emergency rescue supplies will be needed at each stage of the accident based on the input basic accident information. These predicted demand quantities will change over time and as the accident status changes. For example, a large amount of rescue equipment and personnel may be needed in the early stages of the accident, while as the rescue progresses, more medical supplies and daily necessities may be needed.

[0038] Specifically, after obtaining the predicted demand at each time point, the data of demand changes over time can be plotted into a curve, namely, the emergency rescue material demand change curve. This curve can intuitively show the dynamic changes in the demand for emergency rescue materials during the accident, including the peak time of demand, the stage of demand reduction, and the time point when demand tends to stabilize.

[0039] Step S203, according to the requirements of the emergency rescue material demand change curve, when the time and material quantity meet the requirements, determine the storage information of the emergency rescue materials, wherein the above storage information includes one or more of the first reserve quantity of emergency rescue materials in the coal mine, the site selection of the shared base and the second reserve quantity of emergency rescue materials in the shared base, wherein the emergency rescue materials in the shared base can be used by multiple coal mines.

[0040] Specifically, it ensures that in the event of an emergency, coal mines can quickly obtain sufficient emergency rescue materials to meet the needs of accident emergency rescue. By determining the first reserve quantity and the second reserve quantity, it can ensure that coal mines have sufficient material reserves to deal with emergencies. At the same time, selecting a shared base and determining its site selection can improve the efficiency of material reserves. Multiple coal mines can share the same reserve base, reduce the cost of duplicate reserve materials, and improve the overall emergency rescue response capability.

[0041] Through this embodiment, based on the random forest algorithm, the demand for specific emergency rescue materials is accurately predicted, and then a joint storage model between coal mining enterprises and shared reserve centers (i.e., shared bases) is established according to the predicted demand. In this way, emergency rescue materials can be stored in both coal mines and shared bases, making the decision to store emergency rescue materials more in line with the actual situation of accident rescue, thereby improving the speed and effectiveness of emergency response to accident rescue.

[0042] During the specific implementation process, for specific accident types, based on historical accident information, a random forest algorithm is used to construct a material demand model for specific types of emergency rescue materials during the accident emergency rescue period as time passes and the accident evolves. The demand for emergency rescue materials is determined through the above material demand model. The following steps can be taken: construct the above demand model, wherein the above demand model is obtained by training with a random forest algorithm using multiple sets of training data, and each set of the above multiple sets of training data includes the above historical accident information obtained within a historical time period and the historical demand corresponding to the above historical accident information; input the current accident information into the above demand model to obtain the above demand corresponding to the current accident information.

[0043] In this solution, by using the model trained by the random forest algorithm, coal mines can more accurately determine future demand, thereby more effectively planning and managing material reserves. Coal mines can purchase and reserve the required materials in a timely manner based on the determined demand to ensure that the production process will not be affected by insufficient materials, thereby improving the production efficiency of coal mines.

[0044] Specifically, the prediction of demand for specific emergency rescue materials based on random forest aims to determine the demand for specific rescue materials in the coal mine accident rescue process based on the accident rescue process. Random forest and other methods are used to determine the accident characteristic parameters that determine the demand for specific materials for specific accident types. Through historical accident data analysis, a model for the demand for rescue materials in accident rescue is established (the relationship between the demand for materials and the required input and the main characteristic parameters of the accident is established).

[0045] The prospect of establishing a prediction model for the demand for specific emergency rescue materials is described as follows: In the emergency rescue process of coal mining enterprises facing emergencies such as mine accidents, ensuring sufficient emergency materials and dispatching the required materials in a timely and accurate manner can effectively reduce casualties and ensure the normal operation of the enterprise, which is of great significance for preventing the expansion of emergencies. Therefore, the prediction of the demand for specific emergency rescue materials in the accident rescue process of coal mining enterprises is one of the important links in the allocation of material demand, and its difficulty lies mainly in the complexity and uncertainty of the coal mine production environment. The prediction of emergency material demand of coal mining enterprises needs to take into account a variety of complex factors, and changes in these factors will affect the demand for specific emergency rescue materials.

[0046] In the early stages of mine operations, due to the complex mine operating environment, the risks generated are difficult to fully predict, so coal mining companies often find it difficult to accurately predict the demand for specific emergency rescue materials during the accident rescue process. In order to effectively respond to emergencies, it is necessary to make dynamic predictions of emergency materials in advance. Through the analysis of historical data on coal mine accidents, it can be seen that the main cause of coal mine accidents is caused by a variety of complex factors, such as mine mining methods, system structure, ventilation type, accident type and characteristics, etc. However, for different coal mine accidents, the severity of the consequences will be affected by a variety of factors. At the same time, these factors will also affect the demand for specific emergency materials during the accident rescue process to a certain extent. Therefore, it is necessary to collect and analyze the relevant data of these factors related to the demand for emergency materials during the accident rescue process, grasp the occurrence of coal mine accidents and the trend of emergency material demand, and meet the needs of timely emergency rescue.

[0047] The frequency and scale of coal mine accidents show a certain regularity. Therefore, it is possible to predict possible coal mine accidents in the future based on historical data, and then predict the demand for specific emergency rescue materials after the accident. Based on factors such as the type, scale and impact range of coal mine accidents, and in accordance with the occurrence law of coal mine accidents, historical data of coal mine accidents are used for analysis to establish a scientific and reasonable prediction model for the demand for specific emergency rescue materials, revealing the main factors affecting the demand for materials and their quantitative relationships. Then, combined with the specific coal mine situation, the relevant parameters are imported into the model to accurately predict the demand and types of emergency materials, providing a basis for the targeted preparation and deployment of emergency materials.

[0048] When predicting the demand for specific emergency rescue materials during accident rescue, the demand for specific emergency rescue materials for accident rescue is described and combined with the supply of accident rescue materials under the shared reserve model. Under the shared model, the main decision-making object is the storage quantity of specific emergency rescue materials by coal mining enterprises and shared reserve bases. To ensure the effective implementation of accident rescue, that is, the total storage volume of specific emergency rescue materials by coal mining enterprises and shared reserve bases under the shared model should meet the rescue needs. According to the demand characteristics for specific emergency rescue materials during accident rescue, for the sake of simplicity of analysis, it is only necessary to consider that the total storage volume of specific emergency rescue materials by coal mining enterprises and shared reserve bases is greater than the maximum demand for a certain specific emergency rescue material during accident rescue.

[0049] In summary, it can be concluded that when constructing a prediction model for the demand for specific emergency rescue materials, the content that needs to be predicted mainly includes the maximum demand for specific emergency rescue materials by coal mining enterprises under the most unfavorable conditions during the accident rescue process.

[0050] Based on the above situation, considering that the uncertainty and stage-by-stage demand for emergency materials in coal mining enterprises when emergencies occur is difficult to characterize, the random forest algorithm is used to realize prediction and perform data fitting, so as to accurately predict the specific emergency rescue material demand of each coal mining enterprise in different rescue stages of accident rescue, and provide a basis for the joint reserve of specific emergency rescue materials.

[0051] Specifically, the prediction model for the demand for specific emergency rescue materials assumes the following: 1) The evolution of the demand for emergency materials has certain regularities and trends, which can be described by certain mathematical models; 2) The prediction model based on the random forest algorithm is a method suitable for time series data analysis and prediction. It is assumed that the model is suitable for the dynamic prediction of the demand for emergency materials in coal mining enterprises; 3) When the specific emergency rescue material demand model is used for prediction, the selected features (such as historical demand, mine production conditions, etc.) are highly correlated with the demand for the specific emergency rescue materials, and there is no high multicollinearity (multicollinearity refers to the strong correlation between two or more features in the model. In this case, the contribution of the features to the model is difficult to distinguish, which may lead to model instability and inaccurate prediction results. When selecting features, ensure that there is no high multicollinearity between them, that is, any feature cannot be almost The prediction model is based on historical accident data of the coal industry, assuming that historical data can reflect future development trends, there is no abnormal demand caused by major events, the data set is continuous in time, and historical data can well represent future situations; 7) It is assumed that the parameters and initial values ​​in the model are known and can be estimated through historical data or empirical analysis.

[0052] The data collection in the construction of the prediction model for the demand for specific emergency rescue materials is as follows: First, all coal enterprises that have had coal mine accidents in recent years are counted, and detailed accident data of each coal mine accident is collected, including basic accident information of each accident coal mine, basic information of the coal mine, accident cause analysis, personnel involved in the accident, description of the accident process, emergency rescue situation, accident loss assessment, accident investigation and handling results; and the collected data are sorted and classified. The data sources of the collected data include internal and external information of coal mine enterprises. Internal data covers accident reports, material management records, emergency response records, etc., and external data includes meteorological information, geological exploration data, and industry reports. The data types are mainly divided into time series data, accident characteristic data, personnel and rescue data, and material storage and deployment data. These data together constitute the training data set of the model.

[0053] Afterwards, the collected accident data will be sorted and summarized, and different types of accident data will be classified to provide a reference for establishing a prediction model for the demand for specific emergency materials under different types of accidents. The data set of each accident type obtained will be used as the original data for training the prediction model, and will be used as the training set and verification set of prediction data for subsequent prediction of the demand for specific emergency materials for specific accidents in specific coal mines.

[0054] Specifically, the above basic information of the accident may include: date, location, type, severity, etc.; the basic information of the above coal mine may include: geological conditions, mine structure, mining methods, etc.; the above cause analysis of the accident may include: direct causes, indirect causes and potential root causes; the above personnel situation may include: injuries, deaths, disappearances, etc.; the above accident process description may include: specific situations, timelines and key nodes; the above emergency rescue situation may include: dispatch of rescue teams, use of emergency materials, etc.; the above accident loss assessment may include: personnel losses, economic losses, equipment and facility losses, environmental impacts, etc.; the above accident investigation and handling results may include: investigation reports, handling results and rectification measures.

[0055] The characteristic parameters of the prediction model for the demand for specific emergency rescue materials are determined as follows: Through detailed analysis of the collected historical data, through cooperation with domain experts, based on actual business experience and knowledge, preliminarily determine which factors may affect the demand for emergency materials. For example, under certain specific types of accidents, some special data indicators may directly affect the demand for specific emergency rescue materials. After the preliminary screening of features, a simple tree model or random forest model can be used to evaluate the importance of each feature, calculate the importance score of the feature, and further determine which feature indicators contribute the most to the prediction effect of the model. At the same time, features that have less impact on model performance can be identified, which may be discarded in subsequent steps.

[0056] This embodiment is explained by taking a water seepage accident in a coal mine enterprise as an example. The demand for specific emergency materials, mining drainage pumps, depends on the amount of water seepage. At the same time, the main factors affecting the water seepage amount of coal mine water seepage accidents include the basic conditions of the mine (water pressure, thickness of the impermeable layer, depth of the mining fracture zone, fault drop, aquifer, well depth, section, etc.), underground water conditions, possible water seepage, possible personnel at risk, etc., as well as the influence of other factors, and the factors affecting demand need to be further refined.

[0057] The principle of random forest algorithm is as follows: random forest uses bootstrap repeated sampling technology to randomly extract k samples with replacement from the original training sample set N to generate a new training set sample set, and then generates a random forest composed of k decision trees based on the bootstrap samples. In essence, it is an improvement on the decision tree algorithm, merging multiple decision trees together. The establishment of each tree depends on an independently extracted sample. Each tree in the forest has the same distribution, and the classification error depends on the classification ability of each tree and the correlation between them.

[0058] Figure 2 is a schematic diagram of a random forest algorithm according to an embodiment of the present application, such as Figure 2 As shown in the figure, by randomly extracting samples and features, multiple unrelated decision trees are established, and the prediction results are obtained in parallel. Each decision tree can get a prediction result based on the extracted samples and features. By averaging the results of all the trees, the regression prediction result of the entire forest is obtained.

[0059] It should be noted that the above-mentioned bootstrap method (Bootstrapping or self-service sampling method) is a uniform sampling with replacement from a given training set in statistics. That is to say, every time a sample is selected, it is equally likely to be selected again and added to the training set again. When the sample comes from the population and can be described by a normal distribution, its sampling distribution (Sampling Distribution) is a normal distribution (The Normal Distribution); but when the population from which the sample comes cannot be described by a normal distribution, it is analyzed by asymptotic analysis, bootstrap method, etc.

[0060] The two algorithms of the random forest model are as follows:

[0061] Figure 3 is a schematic diagram of a Bagging algorithm according to an embodiment of the present application, such as Figure 3As shown in the figure, the idea of ​​Bagging algorithm is to use a method similar to "democratic voting", that is, each basic model has one vote, and the final result is voted by all basic models, and the minority obeys the majority principle to produce the prediction result. From the original training data (assuming there are 10,000 data in total), randomly extract 10,000 data with replacement to form a new data set (because it is a random sampling with replacement, it is possible that a certain data is selected multiple times, or it is possible that a certain data is not selected once), and use one training sample to train a basic model each time. After performing random sampling with replacement n times, at the end of training, we can obtain n basic models trained by different data sets, also called n weak learners. Based on the results of these n weak learners, we can obtain a more accurate and reasonable result.

[0062] Figure 4 is a schematic diagram of the Boosting algorithm principle according to an embodiment of the present application, such as Figure 4 As shown in the figure, the essence of the Boosting algorithm is to promote weak learners to strong learners. The difference between it and Bagging is that Bagging treats all basic models equally. Boosting, on the other hand, achieves "differential treatment" for basic models. In layman's terms, the Boosting algorithm focuses on "cultivating elites" and "paying attention to errors". "Cultivating elites" means that in each round, a larger weight will be given to the basic model with more accurate prediction results, and the weight of the basic model with poor performance will be reduced. In this way, in the final prediction, the weight of the "excellent model" is large, which is equivalent to it can cast multiple votes, while the "general model" can only cast one vote or no vote at the time of voting. "Paying attention to errors" means changing the weight or probability distribution of the training data after each round of training, by increasing the weight of those samples that were predicted incorrectly by the basic model in the previous round and reducing the weight of the samples predicted correctly in the previous round, so that the classifier pays more attention to the misclassified data, thereby improving the overall effect of the model.

[0063] According to the principle and basic idea of ​​random forest, the generation process of random forest model mainly includes the following three steps: First, k training sample sets are extracted from the original sample set S through Bootstrap method. Generally, the sample size of each training set is consistent with S; second, the k training sets are learned to generate k decision tree models. In the process of decision tree generation, it is assumed that there are M input variables in total, and F variables are randomly extracted from M variables. Each internal node is split using the optimal splitting method on these F feature variables, and the F value is a constant in the formation process of random forest model; finally, the results of k decision trees are combined to form the final result. For classification problems, the combination method is simple majority voting; for regression problems, the combination method is simple average method.

[0064] The important parameters of the random forest model are as follows: 1) the classification strength of a single tree in the random forest and the correlation between any two trees. In a random forest, the greater the classification strength of each decision tree, that is, the more luxuriant the branches and leaves of each tree, the better the classification performance of the overall random forest; the greater the correlation between trees, that is, the more intertwined the branches and leaves between trees, the worse the classification performance of the random forest. Reducing the correlation between trees can effectively reduce the overall error rate of random forests and increase the strength of each decision tree, because it uses the Bootstrap method to form a training set, and randomly grabs features to split, and does not prune a single tree, so that the random forest model can have a higher noise tolerance and a larger classification strength, and also reduces the correlation between any two trees; 2) OOB estimation, when the Bootstrap method is applied, k simple random samplings with replacement are performed in the original sample set S to form a training sample set. When using Bootstrap to sample S, the probability p of each sample not being selected is, when n is large enough, p=0.368, indicating that nearly 37% of the samples in the original sample set S will not appear in the training sample set. These samples are called OOB (Out of Bag). The use of these samples to estimate the performance of the model is called OOB estimation. This estimation method is similar to the cross-validation method. In a random classification model, it is the error rate of the classification model; in a random regression model, it is the residual of the regression model:

[0065] ;

[0066] There are two methods for random forest to calculate the importance of variables in the model: one is based on OOB error, called MDA (Mean Decrease Accuracy); the other is based on Gini impurity, called MDG (Mean Decrease Gini). In both methods, the greater the decrease, the more important the variable.

[0067] The specific description of MDA is as follows: train the random forest model, use the out-of-bag sample data to test the OOB error of each tree in the model; randomly shuffle the value of the variable v in the out-of-bag sample data, and retest the OOB error of each tree; the average of the difference in the OOB errors of the two tests is the measure of the importance of a single tree to the variable v, and the calculation formula is:

[0068] ;

[0069] MDG is described as follows: Gini-based variable importance is measured by the reduction of Gini impurity caused by variable v. At classification node t, the calculation formula of Gini coefficient impurity is:

[0070] ,

[0071] Among them, Q represents the total number of categories of the target variable, and k / t represents the conditional probability that the target variable is of the kth category in node t. The Gini impurity reduction value of each tree is calculated according to the formula, and then the results of all trees are averaged.

[0072] Use Python to implement the random forest prediction model. First, divide the data set, create a random forest model, train the model, predict the test set, and finally evaluate the model.

[0073] Specifically, the random forest prediction implementation process is as follows: basic problem description, random forest prediction model assumptions, and random forest algorithm prediction model construction.

[0074] 1) Basic problem description:

[0075] In the process of emergency rescue when coal mining enterprises are faced with emergencies such as mine accidents, ensuring sufficient emergency supplies and dispatching the required materials in a timely and accurate manner can effectively reduce casualties and ensure the normal operation of the enterprise, which is of great significance for preventing the impact of emergencies from expanding. Therefore, the forecast of emergency material demand for coal mining enterprises is one of the important links in the allocation of material demand, and its difficulty lies in the complexity and uncertainty of the coal mine production environment. First of all, the forecast of emergency material demand for coal mining enterprises needs to take into account a variety of complex factors, and changes in these factors will affect the demand for emergency materials.

[0076] In the early stage of mine operation, due to the complex mine operation environment, the risks generated are difficult to fully predict, so it is difficult for coal mining enterprises to accurately predict the demand for emergency materials. In order to effectively respond to emergencies, it is necessary to make dynamic predictions of emergency materials in advance. Through the analysis of historical data of coal mine accidents, it can be seen that the main cause of coal mine accidents is caused by a variety of complex factors, such as mine structure, ventilation system, miner skills, etc. However, for different coal mine accidents, the severity of the consequences will be affected by a variety of factors, and to a certain extent, these factors will also affect the demand for specific emergency materials during the accident rescue process. However, in order to achieve effective accident rescue, it is necessary to master the influence weights of multiple factors, so as to achieve targeted rescue and accident prevention, and master the demand for specific rescue materials for accident rescue. Therefore, it is necessary to collect and analyze the relevant data of these factors related to the demand for emergency materials in the process of accident rescue, so as to better understand the occurrence law of coal mine accidents and their corresponding emergency material needs. In the first time after the accident, coal mining enterprises need to quickly predict and allocate the required emergency materials to achieve timely rescue and post-disaster recovery. By establishing a prediction model based on actual data and combining factors such as the type, scale and impact range of coal mine accidents, the demand and types of emergency supplies can be predicted more accurately.

[0077] Therefore, in order to solve the basic problem of dynamic prediction of emergency materials in coal mining enterprises, we can learn from the research results of predecessors in practice, combine the occurrence law and actual situation of coal mine accidents, and establish a scientific and reasonable prediction model to ensure that emergency materials can be prepared and deployed in a targeted manner when emergencies occur, so as to minimize losses and ensure production safety.

[0078] In summary, the problem of this study can be described as: through the analysis of historical data of coal mine accidents, it is found that the frequency and scale of coal mine accidents show a certain regularity. Therefore, it is possible to predict possible coal mine accidents in the future based on historical data, and then predict the demand for emergency materials after the accident. In the early stage of the accident, due to the unclear specific circumstances of the accident, there is a certain uncertainty in the prediction model. Therefore, based on the research object of the uncertain system, it is possible to extract known accurate information, simulate data behavior and evolution laws, and analyze the predicted data of coal mine accidents. It is possible to effectively evaluate the material needs of coal mine enterprises within a certain period of time and maximize the allocation of materials, thereby helping coal mine enterprises to dispatch materials in a timely and accurate manner when an accident occurs, and minimize the losses caused by the accident.

[0079] 2) Random forest prediction model assumptions:

[0080] The evolution of demand for emergency materials has certain regularity and trend, which can be described by certain mathematical models; the random forest model is a method suitable for time series data analysis and prediction, and it is assumed that the model is suitable for the dynamic prediction of emergency material demand of coal mining enterprises; when the model predicts, the selected features (such as historical demand, mine production conditions, etc.) are highly correlated with the demand for emergency rescue materials, and there is no highly multicollinear feature, that is, there is no high correlation between the features; the external environment (such as policies and regulations, economic conditions, technical level, etc.) is relatively stable during the forecast period and will not change drastically; the demand for emergency rescue materials is relatively stable within a certain time window, without extreme fluctuations; for multiple coal mining enterprises participating in the shared base, the demand for emergency rescue materials in each mine has a certain homogeneity, and the demand patterns of different mines are similar; the prediction model is based on historical data, assuming that historical data can reflect future development trends and there is no abnormal demand caused by major events; the data set is time continuous, and the historical data can well represent the future situation; it is assumed that the parameters and initial values ​​in the model are known, which can be estimated through historical data or obtained through empirical analysis.

[0081] 3) Random forest algorithm prediction model construction:

[0082] The dynamic demand forecasting model for emergency materials in coal mine enterprises based on random forest algorithm is a forecasting model based on the actual situation and data of coal mine enterprises. It is used to predict the demand for specific emergency materials at different time points when an accident occurs in coal mine enterprises. The model learns the rules in historical data through training and predicts future demand. This paper analyzes the flooding accident as a case study and analyzes other types of accidents by analogy.

[0083] Data collection: Collect historical data of coal mining enterprises that have experienced flooding accidents in recent years. The data sources include internal and external information of coal mining enterprises. Internal data covers accident reports, material management records, emergency response records, etc., and external data includes meteorological information, geological exploration data, and industry reports. The data types are mainly divided into time series data, accident characteristic data, personnel and rescue data, and material storage and deployment data. These data together constitute the training data set of the model. Each group of samples in the collected data samples contains several feature values ​​and 1 output value (the demand for this type of emergency materials at each time point after the accident or the factors that can directly reflect the demand for emergency materials).

[0084] Method for determining characteristic values: Determine characteristic values ​​based on key factors that affect material demand. For example, when a water seepage accident occurs in a coal mine enterprise, the demand for specific emergency materials, such as mine drainage pumps, depends on the amount of water seepage. At the same time, the main factors affecting the water seepage of coal mine water seepage accidents include water pressure, thickness of the impermeable layer, depth of the mining fracture zone, fault drop, aquifer, etc. There are also other factors that influence the demand, and it is necessary to further refine the factors affecting demand.

[0085] Table 1 shows the dynamic demand forecasting results of emergency materials by random forest. According to the dynamic demand forecasting model of emergency materials by random forest algorithm, the following data are obtained:

[0086] Table 1

[0087]

[0088] Based on the predicted maximum demand for specific emergency materials in each accident rescue stage, further analysis was conducted to obtain the changing trend of coal mine enterprises' demand for specific emergency materials such as mine drainage pumps during the rescue of water seepage accidents.

[0089] In the specific implementation process, according to the above basic information, the above demand model is applied to determine the change curve of the demand for emergency rescue materials as time goes by and the accident evolves during the accident period, which can be achieved through the following steps: obtaining multiple time points; obtaining multiple of the above demands, wherein each of the above demands is obtained at each of the above time points according to the random forest algorithm, and the above time points and the above demands correspond one to one; according to the multiple demands and the multiple time points, constructing a change curve of the demand for emergency rescue materials as time goes by and the accident evolves, wherein the horizontal coordinate of the change curve of the demand for emergency rescue materials is the above time point, and the vertical coordinate of the change curve of the demand for emergency rescue materials is the above demand.

[0090] In this solution, by obtaining multiple time points and corresponding demand, the random forest algorithm can be used to determine future changes in demand. Based on these results, a demand change curve can be constructed to better plan material reserve strategies. By analyzing the demand change curve, coal mines can more accurately predict future demand trends, rationally allocate material reserves, and avoid material surpluses or shortages, which helps to improve the production efficiency and operational benefits of coal mines, while reducing material reserve costs, thereby enhancing the competitiveness and sustainable development capabilities of enterprises.

[0091] Specifically, the trend analysis of changes in the demand for specific emergency rescue materials, such as when an accident occurs in a coal mining enterprise, the demand for a certain type of emergency rescue material often changes over time. The main analysis is on the changes in the demand for specific emergency rescue materials during the accident emergency rescue process. At the same time, in order to facilitate the analysis of relevant decision-making issues on the total storage volume of specific emergency rescue materials, the point where the maximum demand for specific emergency rescue materials is generated up to the time of the accident emergency rescue process is considered when analyzing its changing trend. In addition, for different types of emergency rescue materials, including consumable emergency materials and wear-and-tear emergency materials, there are also obvious differences in the changes in their demand during the accident rescue process, and their changing trends are usually manifested as: consumable emergency rescue materials and wear-and-tear emergency rescue materials.

[0092] (1) Consumable emergency rescue supplies:

[0093] Consumable emergency supplies refer to materials that are gradually consumed or disappear due to use during the emergency rescue process and cannot be recycled or reused. Such materials often need to be stored in large quantities during rescue missions and are completely consumed after use and cannot be used again.

[0094] T1 can determine the specific time range based on the actual situation of each coal mining enterprise and the nature of the accident through on-site investigation, expert review and other methods. For example, after an accident occurs in a coal mining enterprise, the specific time when the actual demand is generated can be determined by experts in related fields. The time point of the accident. For specific relief supplies demand time point, It is the time point of maximum demand for specific rescue materials. For consumable emergency rescue materials, the changing trend of their demand during the accident rescue process is mainly manifested as follows: in the early stage of accident emergency rescue, demand for this type of materials begins to arise, and the level of material demand is low and is constantly increasing; as the accident continues to develop, the impact of the accident continues to expand. At this time, the demand for consumable specific emergency rescue materials in the accident emergency rescue increases sharply, and this type of emergency materials is continuously supplied until the level of material demand reaches the maximum. At this time, the supply of emergency rescue materials per unit time reaches the maximum; as the accident rescue process continues to develop, the accident is effectively controlled, and the subsequent demand for consumable specific emergency rescue materials still exists, but compared with the early stage of accident rescue, its demand level for specific emergency rescue materials begins to decrease continuously until it reaches a relatively stable state, which lasts until the end of the accident rescue.

[0095] In addition, in order to intuitively show the changes in demand for consumable emergency rescue materials during the accident rescue process, the time node T0 is introduced, which indicates the time point when the accident occurred; the time point T1 indicates the time point when the accident rescue requires specific emergency rescue materials, and the specific value of T1 can be determined based on the coal mining enterprise's own situation through on-site investigation and combined with expert opinions; the time point T3 indicates the time point when the demand for specific emergency rescue materials for accident rescue reaches the maximum.

[0096] (2) Consumable emergency rescue supplies:

[0097] Consumable emergency supplies refer to supplies that may be damaged or worn to a certain extent during emergency rescue, but can still be used after proper repair or maintenance. Such supplies have a certain degree of durability, but may gradually be damaged under frequent use or in harsh environments, and eventually need to be replaced or scrapped.

[0098] T1 can determine the specific time range based on the actual situation of each coal mining enterprise through on-site investigation, expert review and other methods. For example, after an accident occurs in a coal mining enterprise, the specific time when the actual demand is generated can be determined by experts in related fields. The time point of the accident. For specific relief supplies demand time point, The time point of maximum demand for specific rescue materials is as follows: when the accident just happens, the impact of the accident is small, and there may be no demand for this type of emergency rescue materials temporarily; as the accident continues to develop, in the early stage of accident rescue, there is a demand for this type of emergency rescue materials, and the demand continues to increase. As the demand level continues to increase, the amount of emergency rescue materials invested is also increasing, until the amount of emergency rescue materials invested can meet the demand for emergency rescue materials in the most serious accidents; after the maximum demand for emergency rescue materials is reached, the accident is effectively controlled. At this time, the supply of this type of materials will no longer increase. It is only necessary to withdraw the invested emergency rescue materials from the accident rescue site as the impact of the accident weakens until the accident rescue is over.

[0099] Specifically, the above-mentioned consumable emergency rescue materials may include but are not limited to: rescue food, drinking water, medical supplies and fuel, etc.; the above-mentioned wear-out emergency rescue materials may include but are not limited to: protective equipment, rescue equipment and communication equipment, etc.

[0100] The demand trend of emergency materials is analyzed as follows: When an accident occurs in a coal mining enterprise, the demand curve for a specific material usually changes over time as follows: 1) Increased demand: The occurrence of an accident may lead to production interruption, equipment damage or casualties, and more materials are needed to support rescue and resume production. Therefore, in the early stages of the accident, the demand for specific materials may increase sharply; 2) Demand peaks: As the accident develops further, the demand for rescue and resumption of production will gradually decrease, but in the process, the demand for specific materials may reach a peak; 3) Demand gradually decreases: After the accident is controlled and production is resumed, the demand for specific materials will gradually decrease.

[0101] Figure 5 is a schematic diagram of changes in the demand for specific emergency materials by a coal mining enterprise according to an embodiment of the present application, such as Figure 5 As shown, The time when the accident occurred; The time point at which a specific relief supply is required (determined by the demand curve); The time when specific shared relief supplies arrive and are put into use, including the time when the shared center prepares, transports, installs and puts them into use (determined by the supplies and storage). is the time point of maximum demand for specific relief supplies (determined by the demand curve).

[0102] According to the changing trend of coal mine enterprises' demand for specific emergency materials, accident rescue can be divided into three important stages, namely, at the beginning of the accident, coal mine enterprises have not yet generated demand for specific emergency materials; at the beginning of accident rescue, the demand is within the coal mine enterprise's own reserve capacity; the accident further deteriorates, the demand exceeds the enterprise's own reserves, and the emergency materials supplied by the sharing center need to be used. Important stage: the principle of accident rescue stage division. Detailed discussion of the accident rescue stage and.

[0103] (1) Principles for the division of accident rescue stages:

[0104] The principle of dividing the stages of coal mine accident rescue is mainly based on the use of emergency materials when an accident occurs, which is divided into three stages: in the first stage, there is no temporary demand for materials in accident rescue, and it mainly relies on the emergency response of personnel and existing facilities for preliminary control and evaluation; in the second stage, accident rescue requires the use of emergency materials stored by the coal mine enterprise itself for emergency rescue and stabilization of the accident site. During this stage, the shared base responds and transports materials according to the demand for emergency materials under the most unfavorable circumstances of the coal mine; in the third stage, accident rescue requires the mobilization of emergency materials stored by the enterprise itself and emergency materials dispatched by the shared base, and cooperation with external professional rescue forces to carry out comprehensive and integrated rescue to ensure that the accident is completely resolved and normal production is restored.

[0105] (2) Detailed discussion of the accident rescue stage:

[0106] Phase 1: Initial emergency response [T0-T1];

[0107] In the initial stage of an accident, the main goal is to quickly assess the accident situation, ensure the safety of personnel, and conduct preliminary control. At this time, the first responders at the accident site should take immediate action to organize all personnel in the affected area to evacuate to a safe area. The person in charge of the site must quickly assess the type, scope and potential hazards of the accident, determine whether an upgraded response is required, and report the accident to the mine dispatch center and the emergency command center. The actions at this stage mainly rely on the emergency response of personnel and existing facilities, and there is no need for materials. Initial control measures may include shutting off gas and power sources to prevent the accident from expanding and create favorable conditions for subsequent rescue operations. The key to this stage is rapid response and initial control to ensure the safety of personnel and prepare for subsequent rescue.

[0108] The second stage: enterprise self-rescue stage [T1-T2];

[0109] After the initial emergency response, the company will enter the self-rescue stage, with the main goal of using its own emergency supplies for emergency rescue and stabilizing the accident site. At this stage, the company needs to mobilize internal resources, including rescue equipment, medical supplies and personal protective equipment, and organize internal rescue teams to carry out on-site rescue operations. These actions include searching and rescuing trapped people, controlling fire sources, and removing toxic gases. A temporary command center should be established to coordinate the rescue work of various departments to ensure smooth information flow and orderly actions. The medical rescue team must provide emergency medical care and provide on-site first aid and transfer for the injured. The successful implementation of this stage depends on the effectiveness of the company's emergency plan and the adequate preparation of emergency supplies to ensure that rescue operations can be carried out quickly and the accident can be effectively controlled.

[0110] Phase 3: Coordinated rescue phase [T2-b];

[0111] When an accident requires larger-scale rescue resources, the company will enter the coordinated rescue stage, with the goal of mobilizing shared base resources for comprehensive rescue and cooperating with external professional rescue forces to improve rescue efficiency. At this stage, the company should request rescue supplies and equipment support from the shared base, and cooperate with local, professional rescue teams, medical institutions and other external forces to develop a detailed comprehensive rescue plan. Comprehensive rescue operations include search and rescue of personnel in high-risk areas, investigation of accident causes, equipment repair and reconstruction, etc. The rescue operations at this stage are larger in scale, more extensive in scope, and more complex in coordination. The key to success lies in close cooperation among all parties and efficient integration of resources.

[0112] Figure 6is a schematic diagram of the demand and supply relationship of specific emergency materials under the shared reserve mode according to an embodiment of the present application, such as Figure 6 As shown in the figure, when an accident occurs in a coal mining enterprise, the supply curve for specific demand materials based on the shared reserve model shows that the demand for specific emergency materials is the largest in the collaborative rescue stage and then decreases.

[0113] Based on the above situation, this study takes into account the uncertainty and stages of the demand for emergency materials in coal mining enterprises during emergencies, which are difficult to characterize. Therefore, the random forest algorithm is used to realize prediction and perform data fitting, so as to accurately predict the emergency material needs of various coal mining enterprises.

[0114] Based on the summary data of historical accidents in the coal industry, a random forest algorithm is used to establish a specific emergency rescue material demand prediction model for specific accident types, so as to predict the demand for specific emergency materials in accident rescue of specific accident types of specific coal mining enterprises: 1) Collect historical data related to accidents in all coal mining enterprises in recent years, organize, summarize and classify the data, and determine the main characteristic indicators that affect the demand for specific emergency rescue materials under each accident type based on all the collected historical data. The data is further cleaned and processed, and used as the training set of the random forest prediction model. The training set composed of historical data is used to train the random forest prediction model to obtain the best prediction model; 2) For specific coal mining enterprises and specific accident types, the specific data of coal mining enterprises are collected according to characteristic parameters, and the best specific emergency rescue material demand prediction model is used to predict the demand for specific emergency materials under the specific accident type of the enterprise, and the maximum demand for specific emergency rescue materials in the accident rescue stage is determined; the prediction results provide a reference for the material reserve level decision-making of coal mining enterprises and shared bases.

[0115] Based on the historical accident data of the coal mining industry, a random forest algorithm was used to develop a specific material demand prediction model construction technology for a specific type of accident. The random forest algorithm was used to predict the demand for emergency materials in different rescue stages, and the demand prediction characteristic curve was analyzed and established: 1) First, according to the characteristics of the use of materials in accident rescue, emergency rescue was divided into three different stages, including: the initial stage of the accident, when the coal mining enterprise has not yet generated demand for specific emergency materials; the beginning of the accident rescue, when the demand is within the coal mining enterprise's own reserve capacity; and the further deterioration of the accident, when the demand exceeds the enterprise's own reserves and the emergency materials supplied by the sharing center are needed. Three important stages; 2) The divided rescue stages are further refined according to the actual situation of different coal mining enterprises. The situation is determined by the random forest algorithm, the duration of each stage and the key time nodes for the supply of emergency materials during the rescue process, including the time point T0 when the accident occurred, the time point T1 when the demand for emergency materials was generated, the time point T2 when the shared base dispatched emergency materials to the demand enterprise, the time point T3 when the demand for emergency materials for accident rescue reached the maximum, and the time point b when the accident rescue ended; 3) According to the random forest algorithm, the specific demand for emergency materials in each time interval corresponding to the comprehensive historical accident data is predicted, and the demand interval for specific materials in each accident rescue stage is determined; 4) According to the prediction results, the emergency material demand prediction curve of all coal mining enterprises in the emergency rescue process is analyzed to provide a reference for the decision-making of material reserve levels of coal mining enterprises and shared bases.

[0116] In the specific implementation process, before constructing the demand determination model, the above method also includes the following steps: constructing an initial demand determination model, wherein the initial demand determination model is obtained by training with a random forest algorithm using multiple sets of training data, and each set of training data in the multiple sets of training data includes historical basic information obtained in a historical time period and historical initial demand corresponding to the historical basic information; inputting the basic information into the initial demand determination model to obtain the initial demand corresponding to the basic information; evaluating the initial demand determination model to obtain an evaluation index, wherein the evaluation index at least includes one or more of mean square error, mean absolute error, and variation; when the evaluation index is less than a preset evaluation threshold, optimizing the initial demand determination model until the evaluation index is greater than or equal to the preset evaluation threshold, wherein the optimization method includes one or more of adjusting the number of decision trees, adjusting the maximum depth, adjusting the minimum sample split number, and adjusting the feature selection number. By establishing the initial demand determination model, the initial demand can be determined by basic information, and by optimizing the model, the accuracy of the prediction can be further improved, helping coal mines to better plan material reserves, avoiding material surplus or shortage caused by inaccurate demand estimation, and improving material utilization and production efficiency.

[0117] The construction and tuning of the prediction model for the demand for specific emergency rescue materials are as follows: After the data set is divided, the random forest algorithm is selected to construct a prediction model for the demand for specific emergency rescue materials. The random forest algorithm generates multiple decision trees and improves the accuracy and stability of the model through ensemble learning methods. In the model training stage, each decision tree is trained on the training set, and a tree with strong generalization ability is constructed by randomly extracting samples and features for splitting. Multiple such decision trees are integrated to form a random forest. Finally, the prediction results of multiple trees are averaged or majority voted to obtain the final prediction result, which can effectively reduce the bias and variance of a single decision tree and improve the prediction performance of the overall model.

[0118] After the training of the prediction model for the demand for specific emergency rescue materials is completed, the prediction model is verified using the test set to evaluate the prediction performance of the model. The common evaluation indicators are used to evaluate the model, which helps to measure the error and fit of the prediction model. If the performance of the prediction model is not ideal, it can be optimized by adjusting the prediction model parameters. Through grid search or random search and other methods, these parameters are gradually adjusted to find the optimal parameter combination to further improve the prediction accuracy and stability of the model. After completing the model training and tuning, the best prediction model is obtained to prepare for the prediction of the demand for emergency materials for specific accidents in specific coal mines.

[0119] Based on the above steps, the classified data sets are sorted according to different coal mine accident types, and a specific emergency rescue material demand prediction model under different specific accident types is established. The optimal specific emergency rescue material demand prediction model is obtained by training the prediction model and tuning the parameters.

[0120] Specifically, the commonly used evaluation indicators mentioned above may include but are not limited to: mean square error (MSE), mean absolute error (MAE) and R², etc.; the key parameters of the above model include: the number of decision trees (n-estimators), maximum depth (max-depth), minimum number of sample splits (min-samples-split) and number of feature selections (max-features), etc.

[0121] The application of the specific emergency rescue material demand prediction model is as follows: collect the characteristic parameter values ​​of the target coal mine enterprises corresponding to the characteristic accident types, input their characteristic parameter values ​​into the best specific emergency rescue material demand prediction model obtained by training, realize the prediction of the specific emergency material demand in the specific accident rescue process of a specific mine, and obtain the final prediction results. The prediction results mainly include the maximum demand for specific emergency rescue materials in the accident rescue process under the most unfavorable circumstances.

[0122] After that, according to the prediction results obtained by the prediction model of specific emergency relief material demand, the prediction results are further analyzed to evaluate the rationality and accuracy of the prediction results. By comparing the predicted values ​​with the possible specific emergency relief material demand data based on experience, the errors in the prediction are identified, and the prediction results are corrected and optimized in combination with the opinions of experts in related fields.

[0123] In some embodiments, the storage information of emergency rescue materials is determined, which can be specifically achieved through the following steps: obtaining constraint information, wherein the constraint information includes cost constraint information and time constraint information, wherein the cost constraint information is that the cost of using the emergency rescue materials is less than a preset cost threshold, the cost of using the emergency rescue materials at least includes one or more of the inventory cost of the shared base, the inventory cost of the coal mine and the transportation cost of the emergency rescue materials, and the time constraint information is that the time for the emergency rescue materials from the shared base to arrive at the coal mine is less than a preset time threshold; according to the demand and the constraint information, a particle swarm optimization algorithm is used to determine the storage information of the emergency rescue materials.

[0124] In this scheme, by obtaining cost constraint information and time constraint information and using the particle swarm optimization algorithm to determine the storage information of emergency rescue materials, the coal mine material reserve strategy can be further optimized, which in turn ensures that emergency rescue materials can be quickly and effectively deployed and used in emergency situations, and complete emergency rescue tasks under the premise of controlling costs and meeting time requirements. Coal mines can respond to emergencies more effectively and ensure production and personnel safety.

[0125] The accident rescue process under the joint reserve model of coal mining enterprises and shared reserve bases is analyzed as follows: Under the rescue form of the joint reserve model of coal mining enterprises and shared reserve bases, by analyzing the rescue characteristics under the sharing model and the core idea of ​​"enterprise emergency, social rescue", under the shared reserve model of specific emergency rescue materials, after an accident occurs, the coal mining enterprise will make an emergency response as soon as possible, and according to the accident situation, the emergency rescue materials reserved by the coal mining enterprise itself will be put into the accident rescue; at the same time, when an accident occurs, the shared reserve base will also respond as soon as possible and start to transport emergency rescue materials to the affected coal mining enterprises. During the transportation of emergency rescue materials, the accident rescue mainly depends on the specific emergency rescue materials reserved by the coal mining enterprises themselves. In order to ensure effective rescue, the coal mining enterprises' own material reserves at this stage should be able to meet the emergency rescue material needs within the time interval; after the emergency rescue materials of the shared reserve base are transported, the total amount of specific emergency rescue materials dispatched by the coal mining enterprises and the shared reserve base should meet the needs of the entire accident rescue process for specific emergency rescue materials.

[0126] In some embodiments, the inventory cost of the shared base is the sum of the first cost, the second cost and the third cost, the first cost is the construction cost of the shared base, the second cost is the product of the storage cost of the emergency rescue materials stored in the shared base and the storage quantity of the emergency rescue materials, the third cost is the product of the storage quantity of the emergency rescue materials stored in the shared base and the price of the emergency rescue materials, the inventory cost of the coal mine is the sum of the fourth cost, the fifth cost and the sixth cost, the fourth cost is the construction cost of the coal mine, the fifth cost is the product of the storage cost of the emergency rescue materials stored in the coal mine and the storage quantity of the emergency rescue materials, the sixth cost is the product of the storage quantity of the emergency rescue materials stored in the coal mine and the price of the emergency rescue materials, the transportation cost of the emergency rescue materials is the sum of the seventh cost and the eighth cost, the seventh cost is the product of the probability of an accident occurring in the coal mine and the unit price of transportation, and the eighth cost is the product of the quantity of the emergency rescue materials transported to the coal mine and the distance between the shared base and the coal mine. Of course, this is an optional method of the present application, and those skilled in the art can also make adjustments according to actual conditions.

[0127] The model is constructed as follows: clarify the cost distribution of emergency rescue materials jointly stored by coal mining enterprises and sharing centers, take the lowest cost as the overall goal, and meet the requirements of accident rescue as the core constraint, establish a decision-making model for the optimization of shared base site selection and material storage level, use a suitable algorithm to solve the model, and obtain the construction location of the sharing center and the storage level of coal mining enterprises and sharing bases; the cost distribution of emergency rescue materials jointly stored by coal mining enterprises and sharing centers mainly includes: the inventory cost and transportation cost of the shared base, and the storage cost of each enterprise: 1) Shared base inventory cost: ,in, is the total inventory cost, is the construction cost of the shared base, is the storage cost of specific emergency supplies, It is the storage quantity of specific emergency supplies in the shared base. is the price of a specific emergency supply per unit; 2) Transportation cost: ,in, is the transportation cost, is the probability of an accident occurring in coal mining enterprise j, which can be predicted through statistical analysis. In this case, the probability of an accident occurring in all enterprises is set to 0.5. is the unit price of transportation, is the number of coal mining enterprises, is the quantity of emergency supplies transported to coal mining enterprise j, is the distance between the shared base and coal mining enterprise j; 3) the storage cost of each enterprise: ,in, is the total storage cost of coal mine j, is the construction cost of the material storage depot of the coal mining enterprise, is the material storage cost of coal mining enterprise j, is the specific emergency material reserve quantity of coal mining enterprise j; 4) Total cost:

[0128]

[0129] ;

[0130] 5) Objective function:

[0131] ;

[0132] 6) Constraints: To ensure the effectiveness and feasibility of the shared reserve model of emergency materials in coal mines under the sharing mode, the model needs to ensure that emergency materials can be delivered to the disaster area within the specified time when a disaster occurs, and through cost-benefit analysis, ensure that the model can reduce the overall emergency rescue cost. The specific constraints are as follows: Supply at key time points must meet demand: ; After an accident occurs, the specific emergency supplies given to the shared base should be able to be transported to the coal mining enterprises that generate demand within the specified time:

[0133] ,

[0134] in, is the emergency dispatch time, in hours, Emergency material loading and unloading time, in hours, is the average driving speed in km / h; the quantity of specific emergency supplies stored by the coal mining enterprise itself should be able to meet the material needs of the second stage; The quantity of emergency supplies stored in the shared base should be greater than or equal to the maximum value of the difference between the maximum demand of each enterprise and the quantity of its own stored supplies, and should be able to meet the rescue needs of all coal mining enterprises; ; The location of the shared base must be within the set range:

[0135] , ,in, is the longitude of the shared base, is the latitude of the shared base, is the minimum value of longitude, is the maximum value of longitude, is the minimum value of latitude, is the maximum value of latitude.

[0136] Ideally, when an accident occurs in a coal mine enterprise, the supply curve for specific demand materials is based on the shared reserve model, where G (T1) represents the material supply at time T1, G (T2) represents the material supply at time T2, and G (T3) represents the material supply at time T3. A time node T2 is introduced, indicating that the accident rescue begins to generate demand for emergency rescue materials dispatched by the shared reserve base. Before this time node, the emergency rescue materials for accident rescue at the coal mine accident site mainly rely on the coal mine’s own reserves. The specific value of T2 is preliminarily determined based on the location distribution of the shared reserve base and the coal mine enterprise and the distance from the transportation route, and then the exact result is obtained after review by experts in related fields.

[0137] It should be noted that demand decisions mainly include two aspects: time invested and quantity of materials. Since the rescue cycle of production accidents is relatively short and is generally completed within a few days, in order to simplify decision-making, based on the actual rescue work, the law of changes in material demand during the period is not considered, only the overall demand and arrival time.

[0138] In the event of a specific type of accident, each coal mining enterprise shall make it its primary goal to put its own stored specific emergency rescue materials into use as quickly as possible, and establish a decision-making model for the specific emergency rescue material reserves of the coal mining enterprise. In the management decision-making of the coal mining enterprise for specific emergency rescue materials, the goal is to achieve the goal of putting the specific emergency rescue materials into accident rescue as quickly as possible after the accident occurs to deal with the emergency.

[0139] When a coal mining enterprise encounters a specific type of coal mining accident, it must ensure that the required specific emergency rescue materials can be immediately put into use after the accident occurs. It is also required that the coal mining enterprise's own specific emergency material storage volume can meet the material needs of the self-rescue stage. To this end, the enterprise needs to establish a decision-making model in accordance with this requirement to ensure that the specific emergency materials can arrive at the scene before the analyzed and predicted demand time point T1 to meet the needs of accident emergency rescue.

[0140] The decision-making model for the reserve of specific emergency rescue materials in coal mining enterprises includes the following two key aspects: 1) After the accident occurs, the specific emergency rescue materials can be put into accident rescue at the first time, that is, the materials in reserve can be put into use before the time point when the specific emergency rescue materials are required according to the analysis:

[0141] ;

[0142] Among them, t1 is the time when specific rescue materials are put into use; 2) In the time interval from T1 to T2, the materials stored by the coal mining enterprise itself play a role, requiring the number of specific emergency materials stored by the coal mining enterprise itself to be no less than the maximum material demand predicted in the accident rescue stage:

[0143] ;

[0144] The shared reserve base aims to meet the emergency rescue needs of the coal mines it undertakes. The establishment of its decision-making model includes the following conditions, namely: 1) After the accident occurs at time T0, the shared base immediately responds and begins to dispatch specific emergency rescue materials that may be in demand to the coal mine enterprise where the accident occurs according to the type of accident. The quantity of dispatched emergency rescue materials should meet the maximum demand of the target coal mine for specific emergency rescue materials under the most unfavorable conditions. The sum of the quantity of dispatched specific emergency rescue materials and the quantity of specific emergency rescue materials reserved by the coal mine enterprise itself should meet the accident rescue needs, namely:

[0145] ;

[0146] in, The quantity of materials to be reserved for the shared reserve base; 2) The material dispatch time should arrive before the demand for the shared base materials occurs, that is, it should arrive at the demand enterprise before T2 time:

[0147] ;

[0148] in, is the emergency dispatch time, in hours, Including the time for loading and unloading emergency materials, etc., the unit is h, is the average driving speed in km / h; 3) When providing emergency rescue to the target enterprise, the overall specific emergency rescue material reserves shall be comprehensively considered, and the reserve quantity of the shared base shall meet the accident rescue needs of all coal mining enterprises in the most unfavorable accident situation:

[0149] ;

[0150] in, The quantity of materials to be reserved for the shared reserve base.

[0151] The cost optimization decision model under the shared reserve model is as follows: clarify the cost distribution of emergency rescue materials jointly reserved by coal mining enterprises and sharing centers, take the lowest total cost as the overall goal, and meet the accident rescue as the premise, establish a decision model for the optimal location of shared bases and the optimal storage level of materials, use a suitable algorithm to solve the model, and obtain the construction location of the sharing center and the optimal storage level of specific emergency rescue materials for coal mining enterprises and sharing bases; the cost distribution of specific emergency rescue materials jointly reserved by coal mining enterprises and sharing centers mainly includes the inventory cost of shared reserve bases, the transportation cost of specific emergency rescue materials, and the storage cost of each coal mining enterprise:

[0152] 1) Shared reserve base inventory costs:

[0153] ;

[0154] in, is the storage duration of specific emergency relief supplies;

[0155] 2) Transportation costs of specific emergency relief supplies:

[0156] ;

[0157] in, is the weight of the specific emergency supplies in kg;

[0158] 3) Storage costs of various coal mining enterprises:

[0159] ;

[0160] 4) Total construction cost C under the shared reserve model:

[0161] ;

[0162] ;

[0163] 5) Construction of joint reserve decision model under shared reserve mode: Taking the lowest comprehensive rescue cost as the goal, establish the objective function:

[0164] ;

[0165] On the premise of achieving the fastest accident rescue, the time constraints of accident rescue decision-making are established: 1) The emergency material demand change curve obtained by the above prediction is used as an important reference for the material storage level decision-making planning, so that specific emergency materials can be put into rescue as soon as possible after the accident occurs, that is, during the period from T1 to T2, the materials stored by the coal mine enterprise itself play a role, and the number of materials stored by the coal mine enterprise itself should not be less than the maximum material demand predicted in this stage; 2) After the accident occurs at T0, the shared base immediately responds and begins to dispatch specific emergency materials that may be in demand, and the dispatched quantity should be dispatched in accordance with the maximum demand for the entire rescue process under the most unfavorable conditions, and the material dispatch time should arrive before the demand for the shared base materials occurs, that is, arrive at the demand enterprise before T2 time; 3) In the stage from T2 to T3, the accident rescue begins to generate demand for the emergency materials mobilized by the shared base. At this time, the number of specific emergency materials at the rescue site (including the coal mine enterprise's own reserve materials and the shared base dispatch materials) should not be less than the maximum material demand for accident rescue under the most unfavorable conditions, so that the enterprise can be put into accident emergency rescue as soon as possible after the accident occurs and the emergency materials should be able to meet the rescue needs.

[0166] On the premise of meeting the emergency rescue of multiple target enterprises, the constraint conditions for the storage level decision of accident rescue decision-making materials are established, and a decision-making method for the specific rescue material demand of the rescue sharing base is established to meet the emergency rescue needs of multiple target enterprises. At this time, the material reserve level decision should not only meet the overall material demand of all coal mining enterprises, but also meet the emergency rescue needs of individual coal mining enterprises. During the analysis, at most only one coal mining enterprise has material demand, that is, there are no two or more enterprises with the same type of accidents at the same time: 1) When considering the whole, that is, when performing emergency rescue for the target enterprise, the overall material reserves must be considered comprehensively, and the reserve quantity of the shared base should meet the accident rescue needs of all coal mining enterprises in the most unfavorable situation of the accident; 2) Considering a single enterprise, that is, the sum of its own reserve quantity and the materials dispatched by the shared base should be able to fully meet the accident rescue, which is used as one of the reference bases for the decision on the level of emergency material reserves.

[0167] Consider a coal mine emergency material single-cycle inventory management system. The research object is a specific emergency material, and the decision-making subject is the coal mine enterprise ( ) Emergency material storage and sharing bases, which require decision-making coal mining enterprises ( ) Storage quantity of specific relief supplies and the number of specific relief supplies stored at the shared base , and shared base locations The environmental characteristics of the decision-making are that the demand for specific emergency materials of each coal mining enterprise in different rescue stages in a certain type of accident has been determined through the random forest algorithm prediction model, and the storage allocation problem is solved with the maximum demand as the total inventory (that is, the total demand is known, and the storage capacity of each coal mining enterprise and the shared base storage capacity are determined).

[0168] In some embodiments, based on the above-mentioned demand and the above-mentioned constraint information, a particle swarm optimization algorithm is used to determine the above-mentioned storage information of the above-mentioned emergency rescue materials, which can be specifically achieved through the following steps: using a particle swarm optimization algorithm to solve according to the above-mentioned demand, adjusting the positions of particles that do not meet the above-mentioned constraint information until the termination condition is met, outputting the above-mentioned first reserve quantity of the above-mentioned emergency rescue materials in the above-mentioned coal mine, the site selection situation of the above-mentioned shared base and the above-mentioned second reserve quantity of the above-mentioned emergency rescue materials in the above-mentioned shared base, wherein the above-mentioned termination condition includes that the number of iterations reaches the maximum number of iterations and / or the quality of the solution reaches the optimal quality.

[0169] In this scheme, in order to ensure the feasibility of the solution, it is necessary to adjust the position of particles that violate the constraints or apply a penalty function. The algorithm continues to iterate until the termination condition is met, such as reaching the maximum number of iterations or the quality of the solution reaches a predetermined standard. Finally, the algorithm outputs the optimal or suboptimal material reserve plan, which includes the material storage volume of the sharing center, the location of the sharing center, and the reserve volume of each coal mining enterprise. Throughout the process, the parameters of the PSO algorithm are carefully adjusted, and the appropriate inertia weight, learning factor, and number of particles are selected to ensure the convergence and search ability of the algorithm.

[0170] Joint reserve model between coal mining enterprises and shared reserve bases: Based on the predicted demand for specific emergency rescue materials of each coal mine, it is assumed that the shared material base (center) is responsible for the rescue material needs of n surrounding coal mines, and a shared decision-making model is established for specific rescue materials.

[0171] The solution of the joint reserve model of coal mining enterprises and shared reserve bases is as follows: Under the premise of effectively ensuring accident rescue, the objective function is established with the lowest total construction cost of coal mining enterprises and shared bases as the goal, and corresponding constraints are established. Select a suitable optimization algorithm model to solve the joint reserve model of coal mining enterprises and shared reserve bases, and obtain the optimal storage level of specific emergency rescue materials and the construction location of the shared base for coal mining enterprises and shared reserve bases. When solving this decision model, the algorithm models that can be used include simulated annealing algorithm, genetic algorithm, and particle swarm optimization algorithm. The main algorithm model used in this scheme is the particle swarm optimization algorithm.

[0172] Particle Swarm Optimization (PSO), also known as Particle Swarm Optimization or Bird Swarm Foraging Algorithm, is a group intelligence optimization technology that simulates the foraging behavior of bird flocks. It is a new evolutionary algorithm (EA) developed by J. Kennedy and RC Eberhart in recent years. It is used to solve complex optimization problems. PSO algorithm is a kind of evolutionary algorithm. Similar to simulated annealing algorithm, it also starts from random solution and finds the optimal solution through iteration. It also evaluates the quality of solution by fitness, but its rules are simpler than genetic algorithm. It does not have the "crossover" and "mutation" operations of genetic algorithm. It finds the global optimum by following the currently searched optimal value. This algorithm has attracted the attention of the academic community for its advantages of easy implementation, high precision and fast convergence, and has demonstrated its superiority in solving practical problems. PSO is a parallel algorithm.

[0173] The particle swarm algorithm simulates birds in a flock by designing a massless particle. The particle has only two properties: speed and position. The speed represents the speed of movement, and the position represents the direction of movement.

[0174] The bird is abstracted as a particle (point) without mass and volume and extended to N-dimensional space. The position of particle i in N-dimensional space is represented by vector X i =(x 1 , x 2 , …, x N ), the flight speed is represented by vector V i =(v 1 , v 2 , …, v N ). Each particle has a fitness value determined by the objective function and knows its best position (pbest) found so far and its current position X i This can be regarded as the particle's own flight experience. In addition, each particle also knows the best position (gbest) found by all particles in the entire group so far (gbest is the best value in pbest), which can be regarded as the experience of the particle's companions. The particle decides the next movement based on its own experience and the best experience of its companions.

[0175] When applying PSO to solve the coal mine emergency rescue material reserve model, it is necessary to first define the objective function that needs to be optimized, that is, the objective function established with the goal of minimizing the comprehensive rescue cost, and then establish constraints with the goal of storing specific emergency rescue materials in each mine as quickly as possible and sharing the reserve base to meet the emergency rescue needs of the coal mines undertaken, initialize a group of randomly distributed particles, each particle represents a potential solution, and assign an initial velocity to each particle.

[0176] Particle swarm algorithm speed and position update: PSO is initialized as a group of random particles (random solution), and then finds the optimal solution through iteration. In each iteration, the particles update themselves by tracking two "extreme values" (pbest, gbest). After finding these two optimal values, the particles update their speed and position through the following formula. During the iteration process, each particle adjusts its flight direction and speed according to its own historical optimal position (pBest) and the group's historical optimal position (gBest). The particle position update follows the following formula:

[0177] ;

[0178] ;

[0179] Where i=1, 2, 3…, N, N is the total number of particles in this group, is the velocity of the particle, is a random number between (0, 1), is the current position of the particle, and is the learning factor, usually , The maximum value of (greater than 0), if Greater than ,but .

[0180] Based on the above formula, further in-depth analysis can be obtained to obtain the iterative optimization formula of the particle swarm optimization algorithm:

[0181] ,in, is the inertia factor, its value is non-negative, its value is large, the global optimization ability is strong, the local optimization ability is weak, its value is small, the global optimization ability is weak, the local optimization ability is strong, and the dynamic Can obtain better optimization results than fixed values, dynamic It can change linearly during the PS0 search process, or it can change dynamically according to a certain measurement function of PS0 performance.

[0182] The most commonly used strategy is the linear decreasing weight strategy, namely:

[0183] ,

[0184] in, is the maximum number of iterations, is the initial inertia weight, is the inertia weight when iterating to the maximum evolutionary generation, With the introduction of PSO, the performance of the PSO algorithm has been greatly improved. The global and local search capabilities can be adjusted for different search problems, which also makes the PSO algorithm successfully applied to many practical problems.

[0185] Figure 7 is a flow chart of the PSO algorithm according to an embodiment of the present application, such as Figure 7 As shown in the figure, the standard PSO algorithm process is as follows: 1) Initialize a group of particles (the group size is N), including random positions and speeds; 2) Evaluate the fitness of each particle; 3) For each particle, compare its fitness value with the best position pbest it has passed. If it is better, use it as the current best position pbest; 4) For each particle, compare its fitness value with the best position gbest it has passed. If it is better, use it as the current best position gbest; 5) Adjust the particle speed and position according to the above formula; 6) If the end condition is not met, go to step 2).

[0186] The iteration termination condition is generally selected as the maximum number of iterations Gk or (and) the optimal position searched by the particle swarm so far meets the predetermined minimum adaptation threshold according to the specific problem.

[0187] Local and global PSO algorithms: When C1=0, the particles lose their cognitive ability and become a social-only model: , which is called the global PSO algorithm. Particles have the ability to expand the search space and have a faster convergence speed, but due to the lack of local search for complex problems; it is more likely to fall into local optimality than the standard PSO. When C2=0, there is no social information between particles, and the model becomes a cognition-only model: It is called the local PSO algorithm. Since there is no information exchange between individuals, the entire group is equivalent to multiple particles performing blind random searches, with a slow convergence speed, and thus the possibility of obtaining the optimal solution is small.

[0188] To ensure the feasibility of the solution, it is necessary to adjust the position of particles that violate the constraints or apply penalty functions. The algorithm continues to iterate until the termination condition is met, such as reaching the maximum number of iterations or the quality of the solution reaches a predetermined standard. Finally, the algorithm outputs the optimal or suboptimal material reserve plan, which includes the material storage volume of the sharing center, the location of the sharing center, and the reserve volume of each coal mining enterprise. Throughout the process, the parameters of the PSO algorithm are carefully adjusted, and the appropriate inertia weight, learning factor, and number of particles are selected to ensure the convergence and search ability of the algorithm.

[0189] Based on the optimal reserve quantity of coal mining enterprises and shared reserve bases and the construction location of the sharing center obtained by the model, the optimal reserve plan for the joint reserve of specific emergency rescue materials by coal mining enterprises and shared reserve bases is determined.

[0190] The demand curve for specific emergency rescue materials is determined as follows: Based on the results obtained from the specific emergency rescue material demand prediction model and the joint reserve model of coal mining enterprises and shared reserve bases, the results are analyzed in depth to determine the changing trend of demand for specific emergency rescue materials after an accident in a coal mining enterprise. Among them, the specific emergency rescue material demand prediction model based on random forest can be used to obtain the maximum demand of coal mining enterprises for specific emergency rescue materials, and the joint reserve model of coal mining enterprises and shared reserve bases can be used to determine the specific location of the coal mining enterprise and the distance between the shared reserve base and each coal mining enterprise.

[0191] An optimization model for the joint reserve of specific emergency rescue materials is constructed with the lowest total cost as the goal. The model comprehensively considers the inventory construction cost, transportation cost and storage cost of shared bases and coal mining enterprises to establish constraints to ensure effective accident rescue, and establishes a joint reserve optimization model for specific emergency rescue materials with the lowest total input cost as the objective function.

[0192] The particle swarm optimization algorithm is used, combined with the established objective function and constraints, to solve the objective function, determine the location of specific emergency rescue material reserves and shared bases, and formulate an optimization plan for the joint reserve of specific emergency rescue materials.

[0193] According to the prediction results of accident demand changes and the constraints of accident rescue, a decision-making model is established with the guarantee of accident rescue as the core and the minimum emergency input cost as the goal. On the basis of the constraints established above, the inventory cost, transportation cost and storage cost are comprehensively considered, and the objective function is established with the minimum total input cost. In this way, a decision-making model for the joint reserve of emergency materials is established. This model aims to realize the decision-making of the emergency material storage level of coal mining enterprises and shared bases and the site selection of shared storage bases.

[0194] By selecting a suitable optimization algorithm to solve the model, the location of the shared base and the storage level of the enterprise and the shared base are determined, and the optimal joint reserve decision plan is obtained. The particle swarm optimization algorithm is used to solve the established emergency material joint reserve model, and the material reserve and shared base location are determined within the set constraints, thereby formulating a technical plan for the joint reserve of emergency materials.

[0195] According to the prediction results of random forest on demand, a shared reserve model for emergency materials under the sharing mode is constructed to determine the reserve quantity of specific emergency materials for coal mining enterprises and shared bases and the optimal construction location of shared bases, so as to achieve reasonable distribution of materials, improve the supply efficiency of emergency materials, and reduce the risk of material shortages. In an emergency, reserve systems at all levels can quickly mobilize materials to ensure that demand is met in the shortest time. At the same time, the model also takes into account factors such as the inventory cost of specific emergency materials for coal mining enterprises, the inventory cost of specific emergency materials for shared bases, and transportation costs. By optimizing the reserve quantity and layout, the rationalization of emergency material allocation is achieved, which can not only reduce the inventory pressure of enterprises, but also improve the allocation efficiency of emergency materials and reduce resource waste.

[0196] In the process of constructing the accident rescue decision-making model, we must first ensure that the rescue action responds quickly to meet the time urgency; secondly, ensure that the material reserves are sufficient and updated in time to cope with diverse rescue needs. Then, combined with the accident demand prediction and constraints, a multi-objective optimization model with the lowest cost as the goal is established. Finally, by selecting an appropriate optimization algorithm, the model is solved to determine the optimal rescue base location and material reserve level, so as to achieve efficient joint reserve decision-making.

[0197] The model assumes the following: is a random variable of the time when the emergency demand occurs The value of The time when the accident occurred; The time point at which a specific relief supply is required (determined by the demand curve); The time when specific shared relief supplies arrive and are put into use, including the time when the shared center prepares, transports, installs and puts them into use (determined by the supplies and storage). The time point of maximum demand for specific relief supplies (determined by the demand curve); The second phase of the accident rescue has a greater demand for materials, namely, specific coal mining enterprises ( ) At the time of arrival of shared materials The maximum demand for the material before (determined by the demand curve), and Coal mining enterprises ( ) Storage quantity of specific emergency supplies The maximum demand before the shared supplies arrive needs to be met; The peak point of the third stage of accident rescue demand, that is, the peak point of the specific coal mining enterprise ( ) the maximum demand for a particular emergency supply (determined by the demand curve), To meet the maximum demand for specific rescue materials in each mine to respond to accidents, and , that is, the shared base can meet the maximum demand of a coal mining enterprise for a specific emergency material; , that is, the number of specific emergency supplies stored in the shared base is sufficient to meet the radiation The maximum value of the difference between the maximum demand of a coal mining enterprise for specific emergency materials and the coal mining enterprise’s own reserves; if , that is, there is no need to establish a shared base for the storage of emergency supplies; if , that is, specific coal mining enterprises ( ) The reserve volume of specific emergency supplies can meet their maximum demand and does not need to be shared from a sharing center.

[0198] Figure 8is a schematic diagram of a shared reserve model for emergency supplies in a shared mode according to an embodiment of the present application, such as Figure 8 As shown in the figure, the shared reserve model of emergency materials under the sharing mode composed of shared bases and coal mining enterprises is generally distributed in a star shape, with the shared base in the middle and five coal mining enterprises located around it. The shared base transports specific emergency materials to each coal mining enterprise.

[0199] In some embodiments, obtaining basic information of a coal mine can be specifically achieved through the following steps: obtaining initial basic information of the coal mine; preprocessing the initial basic information to obtain basic information, wherein the preprocessing method includes at least one or more of data cleaning, standardization processing, integration processing and normalization processing.

[0200] In this solution, by preprocessing the data, the validity and accuracy of the data can be guaranteed, so that more accurate demand can be obtained later.

[0201] Specifically, by analyzing the demand for the use of specific emergency materials during the coal mine accident rescue process, the important time nodes at different stages of the accident rescue are determined, and the random forest algorithm is used to predict the material demand at different rescue stages; a specific material reserve decision method is established with the goal of the enterprise to invest in emergency rescue as quickly as possible after the accident, ensuring that the enterprise's self-stored materials can meet the demand in the early stage of rescue; a shared base material reserve decision method for multi-target enterprises is proposed to ensure that the shared base can effectively support the enterprise to achieve accident rescue under the most unfavorable circumstances. Based on this, an emergency material joint reserve optimization model with the lowest total cost as the goal is constructed, comprehensively considering the inventory cost, transportation cost and storage cost, and finally determining the shared base site selection and material reserve level through the particle swarm optimization algorithm. The present invention provides an efficient and economical solution for coal mine enterprises and shared bases to jointly reserve emergency materials, which can effectively improve the response speed and resource utilization efficiency of accident rescue.

[0202] Specifically, the historical accident data preprocessing and data division are as follows: For the collected historical data sets, the data needs to be cleaned, standardized and integrated to ensure the integrity and consistency of the data, retain the characteristic parameter-related data that has been determined to have a greater impact on the demand for specific emergency materials, and remove other indicator data with less relevance to the predicted result variables. Then clean, standardize and integrate the data to ensure the integrity and consistency of the data. When performing standardization, the selected data set input can be normalized to between [0,1], the training set output can be normalized to the default interval [-1,1], and the test set input uses the same normalization method as the training set input to reduce the impact of the order of magnitude difference between different indicators. Data normalization processing:

[0203] ,

[0204] Among them, x is the dimensionless value of the indicator, x' is the measured value of the indicator, is the maximum value of the indicator, is the minimum value of the indicator.

[0205] Model training: After completing data processing, the data set is divided into a training set and a validation set during the model training process. The training set is used to train the model, and the validation set is used to evaluate the performance of the model. The hyperparameters of the random forest are adjusted by cross-validation, such as the number of trees (n-estimators), maximum depth (max-depth), and minimum number of samples of nodes (min-samples split). The model parameters are adjusted to optimize the performance until the model performance reaches the best. That is, the sorted data set is divided into a training set and a test set according to a certain ratio, generally divided in a ratio of 8:2 or 7:3. The training set is used to train a specific emergency relief material demand prediction model, while the test set is used to verify the prediction model. In order to ensure the generalization ability of the model, the representativeness of the data should be paid attention to when dividing. For example, for time series data, it can be divided in chronological order, using earlier data as the training set and newer data as the test set to simulate the actual prediction scenario. In addition, if the feature categories in the data set are unevenly distributed, stratified sampling can be used to ensure that the distribution ratio of each category in the training set and the test set is consistent.

[0206] The implementation ideas of the coal mine emergency material sharing reserve model are as follows: 1) First, the emergency rescue is divided into different stages according to the characteristics of the use of materials in accident rescue, including the initial stage of the accident, when the coal mine enterprise has not yet generated demand for specific emergency materials; the beginning of the accident rescue, when the demand is within the coal mine enterprise's own reserve capacity; and the accident further deteriorates, when the demand exceeds the enterprise's own reserves and the emergency materials supplied by the sharing center are needed. The divided rescue stages are further refined, and according to the actual situation of different coal mine enterprises, the duration of each stage and the key time nodes for the supply of emergency materials during the rescue process are determined, including: the time point T of the accident 0 , the time point when the demand for emergency supplies arises T 1 , the time point T when the shared base dispatches emergency supplies to the enterprise in need 2 , the time point T when the demand for emergency supplies for accident rescue reaches the maximum 3, the end time of the accident rescue b. According to the random forest algorithm, the historical accident data is combined to predict the demand for specific emergency materials in each time interval, and the demand interval for specific materials in each accident rescue stage is determined; according to the prediction results, the forecast curve of emergency material demand of all coal mining enterprises in the emergency rescue process is analyzed to provide a reference for the decision-making of material reserve levels of coal mining enterprises and shared bases.

[0207] 2) The material storage level is planned based on the predicted curve of emergency material demand, so that specific emergency materials can be put into rescue as soon as possible after the accident occurs. That is, during the period from T1 to T2, the materials stored by the coal mine enterprise itself play a role, and the quantity of its reserve materials should not be less than the maximum demand for this stage obtained by prediction; during the period from T2 to T3, there is a demand for the mobilization of materials from the shared base, and at this time, the number of materials and the total number of materials stored by the coal mine should not be less than the maximum demand for rescue materials. In order to achieve the goal of the enterprise in the accident to invest in emergency rescue as soon as possible.

[0208] 3) In order to meet the emergency rescue needs of multiple target enterprises, the rescue sharing base needs to establish relevant constraints for the decision-making method of specific rescue material needs. At this time, the decision on the level of material reserves must meet both the overall material needs and the emergency rescue of individual coal mining enterprises; when analyzing, at most only one coal mining enterprise has material needs, that is, there are no two or more enterprises with the same type of accidents at the same time. When considering the whole, that is, when performing emergency rescue for the target enterprise, the overall material reserves need to be considered comprehensively, and the reserve quantity of the shared base should meet the accident rescue needs of all coal mining enterprises in the most unfavorable situation of the accident; considering a single enterprise, that is, the sum of its own reserve quantity and the materials dispatched by the shared base should be able to fully meet the accident rescue, which is used as one of the reference bases for the decision-making of the emergency material reserve level.

[0209] 4) On the basis of satisfying the constraints of multi-objective rescue and the fastest accident rescue, the inventory cost, transportation cost and storage cost are comprehensively considered, and the objective function is established with the lowest total cost. In this way, a decision model for the joint reserve of emergency materials is established. This model aims to realize the decision-making of the storage level of emergency materials for coal mining enterprises and shared bases and the location of shared storage bases. The particle swarm optimization algorithm is then used to solve the joint reserve model of emergency materials, determine the material reserves and the location of shared bases within the set constraints, and further formulate a technical plan for the joint reserve of emergency materials.

[0210] Fig. 9 is a flow chart of a method for constructing a shared reserve model according to an embodiment of the present application, such as Fig. 9As shown in the figure, the demand for emergency materials in coal mine enterprises mainly refers to the types and quantities of materials required to effectively respond to emergencies during the emergency rescue period after the occurrence of an emergency. The effectiveness here mainly refers to the types, quantities and time points of the materials provided being adapted to the needs of the emergency rescue process, that is, providing the right types and quantities of materials at the right time, which is manifested in high emergency rescue efficiency and the efficiency of emergency material use in the process of responding to emergencies. The demand for the quantity of emergency materials mainly refers to the quantity of materials required to respond to emergencies. In the process of responding to emergencies, according to the principle of obtaining a higher efficiency of emergency material use, the quantity of emergency materials is preferably the minimum quantity that meets the emergency rescue needs. In the process of emergency rescue, some emergency materials are instantaneous needs, while some emergency materials are continuous needs. For emergency materials with instantaneous needs, it is necessary to clarify the time point of demand; for emergency materials with continuous needs, in order to rationally use emergency logistics resources and improve the efficiency of emergency rescue under limited resources, the demand for emergency materials at each stage or time point of emergency rescue should also be reasonably arranged according to the rescue process of the emergency.

[0211] From the above, it can be seen that in an embodiment of the present invention, a dynamic demand prediction model for specific emergency rescue materials is established based on the random forest algorithm, a database is constructed by collecting detailed data of all coal mine accidents in recent years and basic parameters of corresponding enterprises, the prediction model is trained using the database, and the best model obtained by training is used to quickly and accurately predict the demand for specific emergency rescue materials of a specific coal mine enterprise at various stages of emergency rescue under a specific accident type; on the premise that coal mine enterprises can invest in accident emergency as quickly as possible after an accident and shared bases can effectively meet the emergency rescue needs of multiple target enterprises, an optimization model with the lowest total cost as the goal is constructed, and the objective function is solved by the particle swarm optimization algorithm to achieve the optimal reserve level determination and shared base site selection, thereby ensuring efficient use of resources and reducing rescue costs.

[0212] That is, the technical solution provided by the embodiment of the present invention comprehensively considers multiple factors, based on the statistical analysis of historical accidents and the assessment of accident risks of specific coal mining enterprises, predicts the demand and investment time of specific materials for enterprises to deal with possible accidents, establishes a joint reserve model between enterprises and shared reserve centers, and provides decision-making solutions. It is possible to invest in rescue at the fastest speed and meet the demand for the amount of rescue materials in a timely manner. At the same time, overall, it greatly reduces the storage cost of rescue materials and the waste loss caused by long-term idleness.

[0213] The embodiment of the present application also provides a decision-making device for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials. It should be noted that the decision-making device for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials in the embodiment of the present application can be used to execute the decision-making method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and those that have been explained will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0214] The following introduces the decision-making device for the coal mining enterprise and the sharing center to jointly reserve emergency rescue materials provided in the embodiment of the present application.

[0215] Fig.10 This is a structural block diagram of a decision-making device for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials according to an embodiment of the present application. Fig.10 As shown, the device comprises:

[0216] The processing unit 10 is used to construct a material demand model of a specific type of emergency rescue materials over time and the evolution of the accident during the emergency rescue period of the accident based on historical accident information and using a random forest algorithm for a specific accident type, and determine the demand for emergency rescue materials through the material demand model;

[0217] The acquisition unit 20 is used to acquire basic information of the coal mine, wherein the basic information at least includes one or more of the geological conditions of the coal mine, the mine structure, the mining method, the accident type and characteristic parameters, the accident impact range, the equipment usage and the personnel work conditions, and according to the basic information, the demand model is applied to determine the change curve of the demand for emergency rescue materials during the accident period as time passes and the accident evolves;

[0218] The determination unit 30 is used to determine the storage information of the emergency rescue materials according to the requirements of the emergency rescue material demand change curve, when the time and the material quantity meet the requirements, wherein the above storage information includes one or more of the first reserve quantity of the emergency rescue materials of the coal mine, the site selection of the shared base and the second reserve quantity of the emergency rescue materials of the shared base, wherein the emergency rescue materials of the shared base can be used by multiple coal mines.

[0219] Through this embodiment, based on the random forest algorithm, the demand for specific emergency rescue materials is accurately predicted, and then a joint storage model between coal mining enterprises and shared reserve centers (i.e., shared bases) is established according to the predicted demand. In this way, emergency rescue materials can be stored in both coal mines and shared bases, making the decision to store emergency rescue materials more in line with the actual situation of accident rescue, thereby improving the speed and effectiveness of emergency response to accident rescue.

[0220] During the specific implementation process, the processing unit includes a first construction module and a first processing module. The first construction module is used to construct the above-mentioned demand model, wherein the above-mentioned demand model is obtained by training with a random forest algorithm using multiple sets of training data, and each set of training data in the above-mentioned multiple sets of training data includes the above-mentioned historical accident information obtained within a historical time period and the historical demand corresponding to the above-mentioned historical accident information; the first processing module is used to input the current accident information into the above-mentioned demand model to obtain the above-mentioned demand corresponding to the current accident information.

[0221] In this solution, by using the model trained by the random forest algorithm, coal mines can more accurately determine future demand, thereby more effectively planning and managing material reserves. Coal mines can purchase and reserve the required materials in a timely manner based on the determined demand to ensure that the production process will not be affected by insufficient materials, thereby improving the production efficiency of coal mines.

[0222] During the specific implementation process, the acquisition unit includes a first acquisition module, a second acquisition module and a second construction module. The first acquisition module is used to acquire multiple time points; the second acquisition module is used to acquire multiple of the above-mentioned demands, wherein each of the above-mentioned demands is obtained according to the random forest algorithm at each of the above-mentioned time points, and the above-mentioned time points and the above-mentioned demands correspond one to one; the second construction module is used to construct a curve of changes in the demand for emergency rescue materials over time and the evolution of accidents based on the multiple above-mentioned demands and the multiple above-mentioned time points, wherein the horizontal coordinate of the curve of changes in the demand for emergency rescue materials is the above-mentioned time point, and the vertical coordinate of the curve of changes in the demand for emergency rescue materials is the above-mentioned demand.

[0223] In this solution, by obtaining multiple time points and corresponding demand, the random forest algorithm can be used to determine future changes in demand. Based on these results, a demand change curve can be constructed to better plan material reserve strategies. By analyzing the demand change curve, coal mines can more accurately predict future demand trends, rationally allocate material reserves, and avoid material surpluses or shortages, which helps to improve the production efficiency and operational benefits of coal mines, while reducing material reserve costs, thereby enhancing the competitiveness and sustainable development capabilities of enterprises.

[0224] In some embodiments, the determination unit includes a third acquisition module and a determination module, the third acquisition module is used to acquire constraint information, wherein the constraint information includes cost constraint information and time constraint information, wherein the cost constraint information is that the use cost of the emergency rescue materials is less than a preset cost threshold, the use cost of the emergency rescue materials at least includes one or more of the inventory cost of the shared base, the inventory cost of the coal mine and the transportation cost of the emergency rescue materials, and the time constraint information is that the time for the emergency rescue materials from the shared base to arrive at the coal mine is less than a preset time threshold; the determination module is used to determine the storage information of the emergency rescue materials according to the demand and the constraint information by using a particle swarm optimization algorithm.

[0225] In this scheme, by obtaining cost constraint information and time constraint information and using the particle swarm optimization algorithm to determine the storage information of emergency rescue materials, the coal mine material reserve strategy can be further optimized, which in turn ensures that emergency rescue materials can be quickly and effectively deployed and used in emergency situations, and complete emergency rescue tasks under the premise of controlling costs and meeting time requirements. Coal mines can respond to emergencies more effectively and ensure production and personnel safety.

[0226] In some embodiments, the inventory cost of the shared base is the sum of the first cost, the second cost and the third cost, the first cost is the construction cost of the shared base, the second cost is the product of the storage cost of the emergency rescue materials stored in the shared base and the storage quantity of the emergency rescue materials, the third cost is the product of the storage quantity of the emergency rescue materials stored in the shared base and the price of the emergency rescue materials, the inventory cost of the coal mine is the sum of the fourth cost, the fifth cost and the sixth cost, the fourth cost is the construction cost of the coal mine, the fifth cost is the product of the storage cost of the emergency rescue materials stored in the coal mine and the storage quantity of the emergency rescue materials, the sixth cost is the product of the storage quantity of the emergency rescue materials stored in the coal mine and the price of the emergency rescue materials, the transportation cost of the emergency rescue materials is the sum of the seventh cost and the eighth cost, the seventh cost is the product of the probability of an accident occurring in the coal mine and the unit price of transportation, and the eighth cost is the product of the quantity of the emergency rescue materials transported to the coal mine and the distance between the shared base and the coal mine. Of course, this is an optional method of the present application, and those skilled in the art can also make adjustments according to actual conditions.

[0227] In some embodiments, the determination module includes a processing submodule, which is used to use a particle swarm optimization algorithm to solve according to the above-mentioned demand, adjust the positions of particles that do not meet the above-mentioned constraint information until the termination condition is met, and output the above-mentioned first reserve quantity of the above-mentioned emergency rescue materials in the above-mentioned coal mine, the site selection of the above-mentioned shared base and the above-mentioned second reserve quantity of the above-mentioned emergency rescue materials in the above-mentioned shared base, wherein the above-mentioned termination condition includes that the number of iterations reaches the maximum number of iterations and / or the quality of the solution reaches the optimal quality.

[0228] In this scheme, in order to ensure the feasibility of the solution, it is necessary to adjust the position of particles that violate the constraints or apply a penalty function. The algorithm continues to iterate until the termination condition is met, such as reaching the maximum number of iterations or the quality of the solution reaches a predetermined standard. Finally, the algorithm outputs the optimal or suboptimal material reserve plan, which includes the material storage volume of the sharing center, the location of the sharing center, and the reserve volume of each coal mining enterprise. Throughout the process, the parameters of the PSO algorithm are carefully adjusted, and the appropriate inertia weight, learning factor, and number of particles are selected to ensure the convergence and search ability of the algorithm.

[0229] In some embodiments, the acquisition unit includes a fourth acquisition module and a second processing module, the fourth acquisition module is used to obtain the initial basic information of the above-mentioned coal mine; the second processing module is used to preprocess the above-mentioned initial basic information to obtain the above-mentioned basic information, wherein the preprocessing method includes at least one or more of data cleaning, standardization processing, integration processing and normalization processing.

[0230] In this solution, by preprocessing the data, the validity and accuracy of the data can be guaranteed, so that more accurate demand can be obtained later.

[0231] The decision-making device for the joint reserve of emergency rescue materials by coal mining enterprises and sharing centers includes a processor and a memory. The first acquisition unit, the first determination unit, and the second determination unit are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions. The above modules are all located in the same processor; or, the above modules are located in different processors in any combination.

[0232] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problems of high storage cost and low efficiency in the prior art can be solved by adjusting kernel parameters.

[0233] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0234] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute a decision-making method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials.

[0235] An embodiment of the present invention provides a processor, which is used to run a program, wherein when the program is run, a decision method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials is executed.

[0236] A computer program product includes a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the decision-making method for the joint reserve of emergency rescue materials by a coal mining enterprise and a sharing center in each embodiment of the present application are implemented.

[0237] The present application also provides a material storage system, including one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include a decision-making method for executing any one of the above-mentioned coal mining enterprises and sharing centers to jointly reserve emergency rescue materials.

[0238] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0239] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0240] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0241] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0242] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0243] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0244] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0245] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0246] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0247] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A decision-making method for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials, characterized in that: include: For specific accident types, based on historical accident information, a random forest algorithm is used to construct a material demand model for specific types of emergency rescue materials during the accident emergency rescue period over time and as the accident evolves, and the demand for emergency rescue materials is determined through the material demand model; Obtaining basic information of the coal mine, wherein the basic information includes at least one or more of the geological conditions of the coal mine, the structure of the mine shaft, the mining method, the accident type and characteristic parameters, the scope of accident impact, the use of equipment and the working conditions of personnel, and applying the demand model to determine the change curve of the demand for emergency rescue materials during the accident period as time passes and the accident evolves according to the basic information; According to the requirements of the emergency rescue material demand change curve, when the time and material quantity meet the requirements, determine the storage information of the emergency rescue materials, wherein the storage information includes one or more of the first reserve quantity of the emergency rescue materials of the coal mine, the site selection of the shared base and the second reserve quantity of the emergency rescue materials of the shared base, wherein the emergency rescue materials of the shared base can be used by multiple coal mines; Constructing an initial demand determination model, wherein the initial demand determination model is obtained by training with a random forest algorithm using multiple sets of training data, and each set of training data in the multiple sets of training data includes historical basic information obtained in a historical time period and historical initial demand corresponding to the historical basic information; inputting the basic information into the initial demand determination model to obtain the initial demand corresponding to the basic information; evaluating the initial demand determination model to obtain an evaluation index, wherein the evaluation index includes at least one or more of mean square error, mean absolute error, and variance; when the evaluation index is less than a preset evaluation threshold, optimizing the initial demand determination model until the evaluation index is greater than or equal to the preset evaluation threshold, wherein the optimization method includes one or more of adjusting the number of decision trees, adjusting the maximum depth, adjusting the minimum number of sample splits, and adjusting the number of feature selections; Determining the storage information of emergency rescue materials, including: obtaining constraint information, wherein the constraint information includes cost constraint information and time constraint information, wherein the cost constraint information is that the use cost of the emergency rescue materials is less than a preset cost threshold, the use cost of the emergency rescue materials at least includes one or more of the inventory cost of the shared base, the inventory cost of the coal mine and the transportation cost of the emergency rescue materials, and the time constraint information is that the time for the emergency rescue materials of the shared base to arrive at the coal mine is less than a preset time threshold; according to the demand and the constraint information, using a particle swarm optimization algorithm to determine the storage information of the emergency rescue materials; The inventory cost of the shared base is the sum of the first cost, the second cost and the third cost, the first cost is the construction cost of the shared base, the second cost is the product of the storage cost of the emergency rescue materials stored in the shared base and the storage quantity of the emergency rescue materials, the third cost is the product of the storage quantity of the emergency rescue materials stored in the shared base and the price of the emergency rescue materials, the inventory cost of the coal mine is the sum of the fourth cost, the fifth cost and the sixth cost, the fourth cost is the construction cost of the coal mine, the fifth cost is the product of the storage cost of the emergency rescue materials stored in the coal mine and the storage quantity of the emergency rescue materials, the sixth cost is the product of the storage quantity of the emergency rescue materials stored in the coal mine and the price of the emergency rescue materials, the transportation cost of the emergency rescue materials is the sum of the seventh cost and the eighth cost, the seventh cost is the product of the probability of an accident occurring in the coal mine and the unit price of transportation, and the eighth cost is the product of the quantity of the emergency rescue materials transported to the coal mine and the distance between the shared base and the coal mine; According to the demand and the constraint information, a particle swarm optimization algorithm is used to determine the storage information of the emergency rescue materials, including: solving the problem according to the demand using a particle swarm optimization algorithm, adjusting the positions of particles that do not meet the constraint information until a termination condition is met, and outputting the first reserve quantity of the emergency rescue materials in the coal mine, the site selection of the shared base, and the second reserve quantity of the emergency rescue materials in the shared base, wherein the termination condition includes that the number of iterations reaches a maximum number of iterations and / or the quality of the solution reaches an optimal quality.

2. The method according to claim 1, characterized in that For specific accident types, based on historical accident information, a random forest algorithm is used to construct a material demand model for specific types of emergency rescue materials during the accident emergency rescue period over time and the evolution of the accident. The demand for emergency rescue materials is determined through the material demand model, including: Constructing the demand model, wherein the demand model is obtained by training with a random forest algorithm using multiple sets of training data, each set of training data in the multiple sets of training data includes the historical accident information obtained in a historical time period and the historical demand corresponding to the historical accident information; The current accident information is input into the demand model to obtain the demand corresponding to the current accident information.

3. The method according to claim 2, characterized in that Based on the basic information, the demand model is applied to determine the change curve of the demand for emergency rescue materials during the accident period as time passes and the accident evolves, including: Get multiple time points; Acquire a plurality of the demands, wherein each of the demands is obtained at each time point according to a random forest algorithm, and the time points correspond to the demands one by one; According to the multiple demands and the multiple time points, a curve of changes in demand for emergency rescue materials as time passes and accidents evolve is constructed, wherein the horizontal coordinate of the curve of changes in demand for emergency rescue materials is the time point, and the vertical coordinate of the curve of changes in demand for emergency rescue materials is the demand.

4. The method according to any one of claims 1 to 3, characterized in that Get basic information about coal mines, including: Obtaining initial basic information of the coal mine; The initial basic information is preprocessed to obtain the basic information, wherein the preprocessing method includes at least one or more of data cleaning, standardization processing, integration processing and normalization processing.

5. A decision-making device for a coal mining enterprise and a sharing center to jointly reserve emergency rescue materials, characterized in that: include: A processing unit is used to construct a material demand model for a specific type of emergency rescue materials over time and during the emergency rescue period of an accident based on historical accident information and using a random forest algorithm, and determine the demand for emergency rescue materials through the material demand model; An acquisition unit is used to acquire basic information of the coal mine, wherein the basic information includes at least one or more of the geological conditions of the coal mine, the mine structure, the mining method, the accident type and characteristic parameters, the scope of accident impact, the equipment usage and the personnel work conditions, and according to the basic information, the demand model is applied to determine the change curve of the demand for emergency rescue materials during the accident period as time passes and the accident evolves; A determination unit, configured to determine storage information of emergency rescue materials according to the requirements of the emergency rescue material demand change curve, when the time and material quantity meet the requirements, wherein the storage information includes one or more of a first reserve quantity of emergency rescue materials of the coal mine, a site selection condition of the shared base, and a second reserve quantity of emergency rescue materials of the shared base, wherein the emergency rescue materials of the shared base can be used by multiple coal mines; The device is also used to construct an initial demand determination model, wherein the initial demand determination model is obtained by training with a random forest algorithm using multiple sets of training data, and each set of training data in the multiple sets of training data includes historical basic information obtained in a historical time period and historical initial demand corresponding to the historical basic information; the basic information is input into the initial demand determination model to obtain the initial demand corresponding to the basic information; the initial demand determination model is evaluated to obtain an evaluation index, wherein the evaluation index includes at least one or more of mean square error, mean absolute error, and variance; when the evaluation index is less than a preset evaluation threshold, the initial demand determination model is optimized until the evaluation index is greater than or equal to the preset evaluation threshold, wherein the optimization method includes one or more of adjusting the number of decision trees, adjusting the maximum depth, adjusting the minimum sample splitting number, and adjusting the feature selection number; The determination unit includes a third acquisition module and a determination module, the third acquisition module is used to acquire constraint information, wherein the constraint information includes cost constraint information and time constraint information, wherein the cost constraint information is that the use cost of the emergency rescue materials is less than a preset cost threshold, the use cost of the emergency rescue materials at least includes one or more of the inventory cost of the shared base, the inventory cost of the coal mine and the transportation cost of the emergency rescue materials, and the time constraint information is that the time for the emergency rescue materials of the shared base to arrive at the coal mine is less than a preset time threshold; the determination module is used to determine the storage information of the emergency rescue materials according to the demand and the constraint information by using a particle swarm optimization algorithm; The inventory cost of the shared base is the sum of the first cost, the second cost and the third cost, the first cost is the construction cost of the shared base, the second cost is the product of the storage cost of the emergency rescue materials stored in the shared base and the storage quantity of the emergency rescue materials, the third cost is the product of the storage quantity of the emergency rescue materials stored in the shared base and the price of the emergency rescue materials, the inventory cost of the coal mine is the sum of the fourth cost, the fifth cost and the sixth cost, the fourth cost is the construction cost of the coal mine, the fifth cost is the product of the storage cost of the emergency rescue materials stored in the coal mine and the storage quantity of the emergency rescue materials, the sixth cost is the product of the storage quantity of the emergency rescue materials stored in the coal mine and the price of the emergency rescue materials, the transportation cost of the emergency rescue materials is the sum of the seventh cost and the eighth cost, the seventh cost is the product of the probability of an accident occurring in the coal mine and the unit price of transportation, and the eighth cost is the product of the quantity of the emergency rescue materials transported to the coal mine and the distance between the shared base and the coal mine; The determination module includes a processing submodule, which is used to use a particle swarm optimization algorithm to solve according to the demand, adjust the positions of particles that do not meet the constraint information until the termination condition is met, and output the first reserve quantity of the emergency rescue materials of the coal mine, the site selection of the shared base and the second reserve quantity of the emergency rescue materials of the shared base, wherein the termination condition includes that the number of iterations reaches the maximum number of iterations and / or the quality of the solution reaches the optimal quality.

6. A computer program product comprising a computer program, characterized in that When the computer program is executed by the processor, the steps of the decision-making method for the coal mining enterprise and the sharing center to jointly reserve emergency rescue materials as described in any one of claims 1 to 4 are implemented.

7. A material storage system, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a decision-making method for executing the joint reserve of emergency rescue materials by a coal mining enterprise and a sharing center as described in any one of claims 1 to 4.

Images