Coal mine disaster fusion monitoring and intelligent decision-making system and method

Abstract

The invention provides a coal mine disaster fusion monitoring and intelligent decision-making system and method. The system comprises an index obtaining assembly, a reasoning, studying and judging assembly and a man-machine interaction assembly. The method comprises the following steps: carrying out real-time data acquisition on disaster indexes of an internal environment of a coal mine through monitoring equipment, and carrying out data integration on the acquired disaster index data to obtain integrated disaster index data; on the basis of the integrated disaster index data, information reasoning and intelligent study and judgment are applied to speculate the possibility and severity of coal mine disasters, and possible disaster risks are predicted; according to the prediction result, emergency decision making is carried out, and corresponding measures are taken for disaster prevention and control; through a man-machine interaction interface and a visualization technology, monitoring data, a prediction result and a decision scheme are displayed in a graphical mode, so that a decision maker knows a disaster state, the monitoring data and the prediction result and makes an interactive decision. According to the invention, the efficiency and accuracy of coal mine disaster prevention and control are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of information interaction processing, and in particular to a coal mine disaster fusion monitoring and intelligent decision-making system and method. Background Art

[0002] Coal mine disasters refer to accidents or disasters that occur during the production process of coal mines, including coal mine explosions, mine collapses, and coal mine fires. These disasters may cause serious consequences such as casualties, property losses, and environmental pollution. Through monitoring, coal mines can be monitored and managed in real time and comprehensively, and timely warnings and responses to coal mine disasters can be given; the occurrence of coal mine disasters can be minimized, and the lives of miners and the sustainable development of coal mines can be protected. However, the existing technology lacks a unified support system, resulting in limited means of disaster monitoring and inability to effectively deal with disasters.

[0003] Prior art 1, application number: CN 202310077185.5 discloses a risk monitoring method, device, electronic device and storage medium for coal mine disasters, including: obtaining initial production data and initial static data of the area to be detected, processing the initial production data and initial static data, obtaining disaster precursor information corresponding to multiple types of disasters in the area to be detected, determining a coal mine disaster early warning processing model according to the mine condition information of the area to be detected, and performing joint analysis and processing on multiple disaster precursor information according to the coal mine disaster early warning processing model to obtain comprehensive disaster risk information of the area to be detected. Although it is possible to jointly analyze the predicted data of multiple disasters in the area to be detected, before the disaster occurs, the possibility and harm of the disaster are evaluated based on the obtained disaster hidden dangers and precursor information, and the real-time and accurate early warning of coal mine disasters is realized, effectively improving the safety of coal mine production. However, the information processing speed is inefficient and effective monitoring of disasters cannot be formed.

[0004] Prior art 2, application number: CN 202310511207.4 discloses a method for producing a thematic network map for coal mine disaster monitoring and early warning analysis, including the following steps: obtaining and cleaning irrelevant data in a coal mine two-dimensional CAD vector image under the 2000 national geodetic coordinate system; transferring the cleaned vector image to a raster image according to the outermost boundary of the frame to obtain a raster image; obtaining the attribute parameters of the raster image; performing tile cutting to obtain multi-resolution layered tile data; publishing the multi-resolution layered tile data and its corresponding attribute parameters on the network and using them as a thematic base map for coal mine disaster monitoring and early warning analysis, and completing the production of the thematic network map for coal mine disaster monitoring and early warning analysis. Although the thematic maps required for coal mine disaster monitoring and early warning can be quickly loaded and displayed, and can be accurately spatially matched with the monitoring data, the problem of cumbersome processing and inconvenient sharing when coal mine production technology mapping is used as the production of thematic maps for coal mine disaster monitoring and early warning analysis is solved. However, there is a lack of collection and processing of disaster occurrence information, and only location recognition can be achieved, and disaster occurrence cannot be predicted.

[0005] Prior art three, application number: CN 202410002795.3 discloses a coal mine disaster data quality evaluation model construction and evaluation method and device, including: obtaining the evaluation target, evaluation index and evaluation sub-indicator of the target coal mine disaster data quality; obtaining the first importance between the evaluation target and the evaluation index, and obtaining the second importance between the evaluation index and the evaluation sub-indicator; establishing a coal mine disaster data quality evaluation model based on the evaluation target, the evaluation index, the evaluation sub-indicator, the first importance and the second importance. Although a more complete coal mine disaster data quality evaluation model can be obtained by obtaining the evaluation target, the evaluation index and the evaluation sub-indicator, and considering multiple evaluation indicators and weights for model construction, compared with the evaluation method of a single indicator or angle, it can better reflect the true situation of data quality, and at the same time can improve the efficiency and accuracy of quality evaluation, can reflect the true situation of data quality, and provide a data basis for subsequent decision makers to make decisions and improvements based on the results. However, the data is cumbersome, and the model has low efficiency in processing data, which is not conducive to obtaining monitoring results as soon as possible.

[0006] At present, the existing technologies 1, 2 and 3 lack a unified support system, which leads to limited means of disaster monitoring and the inability to effectively deal with disasters. Therefore, the present invention provides a coal mine disaster fusion monitoring and intelligent decision-making system and method, which manages coal mine disaster fusion monitoring and intelligent decision-making, establishes a unified construction and distributed application business support system, develops information reasoning and intelligent analysis technology, and performs logical induction and deductive reasoning on the prevention and control process of five types of disasters in coal mines, namely water, fire, gas, roof and dust, to achieve human-computer interaction and visual communication in the emergency decision-making process, and form a technical solution for coal mine disaster fusion monitoring and intelligent decision-making. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a coal mine disaster fusion monitoring and intelligent decision-making system, comprising:

[0008] The indicator acquisition component is responsible for collecting real-time data on the disaster indicators of the internal environment of the coal mine through monitoring equipment, integrating the collected disaster indicator data, and obtaining the integrated disaster indicator data;

[0009] The reasoning and judgment component is responsible for inferring the possibility and severity of coal mine disasters and predicting possible disaster risks based on the integrated disaster indicator data using information reasoning and intelligent judgment;

[0010] The human-computer interaction component is responsible for displaying monitoring data, prediction results and decision-making plans in a graphical way through the human-computer interaction interface and visualization technology, so as to make interactive decisions.

[0011] Optional reasoning and analysis components, including:

[0012] The model building module is responsible for building a random forest model based on the integrated disaster indicator data using a machine learning algorithm. The random forest model learns the laws and characteristics of coal mine disasters, as well as the factors related to and influencing factors of disasters based on historical data and real-time data.

[0013] The inference and prediction module is responsible for using the established random forest model and probability statistics to infer and predict the possibility and severity of coal mine disasters, and evaluate the current disaster risks; according to the prediction results, make corresponding decisions and control measures, carry out disaster prevention and control, and formulate corresponding emergency decision-making plans;

[0014] The feedback optimization module is responsible for providing feedback and optimizing the random forest model based on the decision-making effect, and evaluating and improving the random forest model based on the actual disaster situation and decision-making results.

[0015] Optional, model building module, including:

[0016] The feature extraction submodule is responsible for receiving the integrated disaster indicator data, collecting data related to coal mine disasters, including historical data and real-time data, cleaning the collected data, and extracting target features;

[0017] The model training submodule is responsible for selecting the random forest machine learning algorithm and training the random forest model using the disaster indicator data set; the trained random forest model is evaluated using the accuracy and recall evaluation indicators;

[0018] The retraining submodule is responsible for learning the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters, updating the data set, and retraining the random forest model based on the target characteristics.

[0019] Optional, model training submodule, including:

[0020] The quantity confirmation unit is responsible for selecting random forest as the machine learning algorithm and determining the number of decision trees; random forest makes predictions by building multiple decision trees;

[0021] The autonomous sampling unit is responsible for random autonomous sampling from the training set when building each decision tree, that is, sampling the training set with replacement;

[0022] The decision tree construction unit is responsible for randomly and autonomously sampling samples from the data set when building each decision tree. According to the selected target features, the data set is divided into different nodes using the Gini index partitioning criterion, and decision trees are constructed and combined into a random forest. The random forest model is trained using the disaster indicator data set.

[0023] Optional, inference prediction module, including:

[0024] The result output submodule is responsible for using the established random forest model to predict real-time data and obtain the prediction results of the possibility and severity of disasters;

[0025] The analysis and statistics submodule is responsible for comparing the output of the random forest model with the actual data, calculating the probability of coal mine disasters, and counting the frequency of disasters of different severity;

[0026] The risk assessment submodule is responsible for evaluating the current disaster risk and determining the probability and possible severity of the disaster based on the results of probability and statistical analysis.

[0027] Optional, feedback optimization module, including:

[0028] The problem analysis submodule is responsible for analyzing the problems existing in the random forest model based on the evaluation results. The prediction results of the random forest model are compared with the actual data to confirm whether they exceed the error threshold.

[0029] The model update submodule is responsible for retraining based on the improved random forest model; retraining the improved random forest model using the updated data set; and validating the improved random forest model using the validation set;

[0030] The loop iteration submodule is responsible for repeatedly performing the process of evaluating, improving, training, and validating the improved random forest model until the improved random forest model reaches the expected performance indicators.

[0031] Optionally, in the problem analysis submodule, if the prediction result of the random forest model exceeds the error threshold, it means that there is a problem with the random forest model, and increasing the amount of sample data can improve the performance of the random forest model.

[0032] Optional, human-computer interaction module, including:

[0033] The visualization submodule is responsible for displaying the prediction results of the probability and severity of disasters obtained by the reasoning and judgment component;

[0034] The plan display submodule is responsible for sending the emergency decision-making plan based on the prediction results to the decision maker by displaying the risk area through a map and the emergency process through a flowchart;

[0035] The decision adjustment submodule is responsible for enabling decision makers to interact with the visualization interface through the human-computer interaction interface, understand the detailed information of monitoring data and prediction results, and select and adjust decision plans.

[0036] Optionally, the visualization submodule uses color coding, labeling and heat mapping to represent different levels of disaster risks and presents them in a visual form.

[0037] The present invention provides a coal mine disaster fusion monitoring and intelligent decision-making method comprising the following steps:

[0038] Through monitoring equipment, real-time data collection is performed on the five disaster indicators of water, fire, gas, roof and dust in the internal environment of the coal mine, and the collected disaster indicator data is integrated to obtain the integrated disaster indicator data;

[0039] Based on the integrated disaster indicator data, information reasoning and intelligent judgment are used to speculate the possibility and severity of coal mine disasters and predict possible disaster risks; based on the prediction results, emergency decisions are made, including taking corresponding measures for disaster prevention and control;

[0040] Through the human-computer interaction interface and visualization technology, monitoring data, prediction results and decision-making plans are displayed in a graphical way, allowing decision makers to understand the disaster status, monitoring data and prediction results and make interactive decisions.

[0041] The index acquisition component of the present invention collects disaster index data of the internal environment of the coal mine in real time through monitoring equipment, and integrates it to obtain accurate disaster index data; it provides a real-time and accurate data basis, and provides reliable information support for subsequent reasoning, judgment and decision-making. Based on the integrated disaster index data, the reasoning and judgment component uses information reasoning and intelligent judgment to speculate and predict the possibility and severity of coal mine disasters; it provides accurate predictions of potential disaster risks, helps decision makers better understand the state of disasters, and take corresponding measures to prevent and control disasters in a timely manner, thereby avoiding or reducing the occurrence and impact of disasters. The human-computer interaction component displays the monitoring data and prediction results in a graphical manner through visualization technology, and allows decision makers to intuitively understand the disaster state, monitoring data and prediction results through the human-computer interaction interface, and make interactive decisions; it provides an intuitive and easy-to-understand information display method, so that decision makers can better understand and analyze data and make scientific and effective decisions.

[0042] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0043] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0045] Figure 1 This is a block diagram of a coal mine disaster fusion monitoring and intelligent decision-making system in Example 1 of the present invention;

[0046] Figure 2 This is a block diagram of an indicator acquisition component in Embodiment 2 of the present invention;

[0047] Figure 3 This is a block diagram of a data integration module in Embodiment 3 of the present invention;

[0048] Figure 4 This is a block diagram of the reasoning and analysis component in Example 4 of the present invention;

[0049] Figure 5 A block diagram of a model building module in Embodiment 5 of the present invention;

[0050] Figure 6 This is a block diagram of the model training submodule in Example 6 of the present invention;

[0051] Figure 7 This is a block diagram of the inference prediction module in Example 7 of the present invention;

[0052] Figure 8 This is a block diagram of a feedback optimization module in Embodiment 8 of the present invention;

[0053] Fig. 9 This is a block diagram of a human-computer interaction component in Embodiment 9 of the present invention;

[0054] Fig.10 This is a flow chart of the coal mine disaster integrated monitoring and intelligent decision-making method in Example 10 of the present invention. DETAILED DESCRIPTION

[0055] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0056] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0057] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0058] Example 1: Figure 1 As shown, an embodiment of the present invention provides a coal mine disaster fusion monitoring and intelligent decision-making system, comprising:

[0059] The indicator acquisition component is responsible for real-time data collection of five types of disaster indicators in the internal environment of the coal mine, including water, fire, gas, roof and dust, through monitoring equipment, and integrating the collected disaster indicator data to obtain the integrated disaster indicator data;

[0060] The reasoning and judgment component is responsible for inferring the possibility and severity of coal mine disasters and predicting possible disaster risks based on the integrated disaster indicator data, using information reasoning and intelligent judgment; making emergency decisions based on the prediction results, including taking corresponding measures for disaster prevention and control;

[0061] The human-computer interaction component is responsible for displaying the monitoring data, prediction results and decision-making plans in a graphical way through the human-computer interaction interface and visualization technology, so that decision makers can understand the disaster status, monitoring data and prediction results and make interactive decisions.

[0062] The working principle and beneficial effects of the above technical solution are as follows: the indicator acquisition component of this embodiment collects real-time data of five types of disaster indicators such as water, fire, gas, roof and dust in the internal environment of the coal mine through monitoring equipment, integrates the collected disaster indicator data, and obtains the integrated disaster indicator data; the reasoning and judgment component uses information reasoning and intelligent judgment based on the integrated disaster indicator data to speculate the possibility and severity of coal mine disasters and predict possible disaster risks; according to the predicted results, emergency decisions are made, including taking corresponding measures for disaster prevention and control; the human-computer interaction component uses the human-computer interaction interface and visualization technology to display the monitoring data, prediction results and decision-making plans in a graphical way, so that decision makers can understand the disaster status, monitoring data and prediction results and make interactive decisions. The indicator acquisition component of the above scheme collects the disaster indicator data of the internal environment of the coal mine in real time through monitoring equipment, and integrates it to obtain accurate disaster indicator data; it provides a real-time and accurate data basis, and provides reliable information support for subsequent reasoning and judgment and decision-making. The reasoning and judgment component uses information reasoning and intelligent judgment based on the integrated disaster indicator data to speculate and predict the possibility and severity of coal mine disasters; it provides accurate predictions of potential disaster risks, helping decision makers to better understand the state of disasters and take appropriate measures to prevent and control disasters in a timely manner, thereby avoiding or reducing the occurrence and impact of disasters. The human-computer interaction component uses visualization technology to display monitoring data and prediction results in a graphical manner, and through the human-computer interaction interface, decision makers can intuitively understand the state of disasters, monitoring data and prediction results, and make interactive decisions; it provides an intuitive and easy-to-understand information display method, allowing decision makers to better understand and analyze data and make scientific and effective decisions.

[0063] To sum up, the various modules of this embodiment provide comprehensive and accurate disaster monitoring and decision-making support by integrating data, speculating predictions and visual displays; improve the efficiency and accuracy of coal mine disaster prevention and control, reduce the occurrence of human errors and disasters, ensure the safe production of coal mines, and protect the lives and property of employees; can realize real-time monitoring, early warning and decision-making support for coal mine disasters, and improve the efficiency and accuracy of coal mine safety management; perform logical induction and deductive reasoning on the five types of disaster prevention and control processes of coal mines, namely water, fire, gas, roof and dust, realize human-computer interaction and visual communication in the emergency decision-making process, and form a technical solution for integrated monitoring and intelligent decision-making of coal mine disasters.

[0064] In this embodiment, disaster refers to adverse effects such as severe damage, casualties, and property losses caused by emergencies or natural disasters; in the coal mining industry, disasters usually refer to coal mine accidents, including gas explosions, coal dust explosions, mine collapses, etc.

[0065] Measures refer to emergency prevention and control measures taken to reduce or prevent the occurrence of disasters and the expansion of losses. Specific measures include but are not limited to: Safety production management measures: strengthen the coal mine safety management system to ensure that operating specifications and safety production procedures are strictly implemented; construction of protective facilities: install gas extraction devices, ventilation equipment, dust prevention equipment, etc. to improve the ventilation conditions and gas extraction effects of mines; emergency rescue measures: establish a complete coal mine accident emergency rescue system, train professional rescue personnel, equip emergency rescue equipment, and improve accident response capabilities; by taking these measures, the probability of coal mine disasters can be reduced, and emergency decisions and control measures can be made in a timely manner when disasters occur, minimizing casualties and property losses caused by disasters.

[0066] Example 2: Figure 2 As shown, based on Example 1, the indicator acquisition component provided by the embodiment of the present invention includes:

[0067] The sensor monitoring module is responsible for deploying various sensors inside the coal mine to monitor various disaster indicators of the internal environment of the coal mine in real time;

[0068] The wireless communication module is responsible for transmitting various disaster indicators collected by various sensors to wireless communication nodes through wireless communication technology, realizing remote acquisition and real-time monitoring of data;

[0069] The data integration module is responsible for obtaining various disaster indicators through wireless communication nodes, processing missing values ​​of the collected data of various disaster indicators, aligning the data from different data sources in time with consistent timestamps, and using the decision-level fusion method to merge the preprocessed and aligned data to obtain the integrated disaster indicator data.

[0070] The working principle and beneficial effects of the above technical scheme are as follows: the sensor monitoring module of this embodiment deploys various sensors inside the coal mine to monitor various disaster indicators of the internal environment of the coal mine in real time; the wireless communication module transmits various disaster indicators collected by various sensors to the wireless communication node through wireless communication technology to realize remote acquisition and real-time monitoring of data; the wireless communication node of the data integration module obtains various disaster indicators, processes missing values ​​of the data collected for various disaster indicators, aligns the data from different data sources in time, has consistent timestamps, and uses the decision-level fusion method to merge the pre-processed and aligned data to obtain the integrated disaster indicator data. The sensor monitoring module of the above scheme monitors various disaster indicators of the internal environment of the coal mine in real time by deploying various sensors, such as temperature, humidity, gas concentration, etc.; obtains accurate and real-time disaster indicator data, and provides a basis for subsequent data processing and decision-making. The wireless communication module transmits the disaster indicator data collected by the sensor to the wireless communication node through wireless communication technology, which can realize remote acquisition and real-time monitoring of data, avoid the limitations of traditional wired transmission methods, and improve the efficiency and convenience of data collection. The data integration module processes missing values ​​for the collected disaster indicator data, aligns the data from different data sources in time to ensure consistent timestamps, and then uses the decision-level fusion method to merge the preprocessed and aligned data to obtain integrated disaster indicator data; it unifies data from different sensors and data sources to improve data accuracy and availability, and provide reliable data support for subsequent disaster prediction, warning, and decision-making.

[0071] To sum up, the sub-modules of this embodiment cooperate with each other to realize the whole process from sensor monitoring to data integration, provide accurate, real-time and reliable disaster indicator data, and provide a scientific basis for disaster management departments and relevant personnel, helping them to better understand the disaster situation, predict disaster trends, and take corresponding measures, thereby reducing the occurrence and losses of disasters.

[0072] Example 3: Figure 3 As shown, based on Example 2, the data integration module provided in this embodiment of the present invention includes:

[0073] The relationship determination submodule is responsible for determining the logical relationship for merging data, and uses the logical AND or method to merge. The logical AND method means that the data from all sensors will be merged into a group of data only when they meet a certain condition; the logical OR method means that as long as the data from one sensor meets a certain condition, they will be merged into a group of data;

[0074] Among them, a certain condition refers to the judgment condition set when merging data. The judgment condition is a threshold. If there are two sensors measuring the same physical quantity, only the data whose difference between the two sensor measurements is less than a certain threshold will be merged. Then this condition is that the absolute value obtained by subtracting the measurement values ​​of the two sensors is less than the threshold. The data is merged using the logical AND method or the logical OR method.

[0075] The logic judgment submodule is responsible for comparing the data of each sensor one by one for the logic and method, and judging whether the merging conditions are met. Only when the data of all sensors meet the conditions, they are merged into a group of data; for the logic or method, it only needs to find the sensor data that meets the conditions;

[0076] The data merging submodule is responsible for merging the sensor data that meets the conditions into a set of data based on the results of logical judgment, and saving it as a new data set.

[0077] The working principle and beneficial effects of the above technical solution are as follows: the relationship determination submodule of this embodiment determines the logical relationship for merging data, and merges them by using the logical AND or method. The logical AND method indicates that the data of all sensors are merged into a group of data only when they meet a certain condition; the logical OR method indicates that as long as the data of one sensor meets a certain condition, they are merged into a group of data; wherein a certain condition refers to the judgment condition set when merging data, and the judgment condition is a threshold value. There are two sensors measuring the same physical quantity, and only those data whose difference between the measured values ​​of the two sensors is less than a certain threshold value are merged. Then the condition is that the absolute value obtained by subtracting the measured values ​​of the two sensors is less than the threshold value, and the data are merged by using the logical AND method or the logical OR method; for the logical judgment submodule, for the logical AND method, compares the data of each sensor one by one, and determines whether the merging condition is met. Only when the data of all sensors meet the condition, they are merged into a group of data; for the logical OR method, it is only necessary to find the sensor data that meets the condition; the data merging submodule merges the sensor data that meets the condition into a group of data according to the result of the logical judgment; and saves it as a new data set. The relationship determination submodule of the above scheme determines the logical relationship of data merging, and selects the logical AND or OR method for merging according to the needs of specific applications; by setting the logical relationship, the merging method can be reasonably selected according to the characteristics and requirements of the sensor data. The logical judgment submodule compares and judges the sensor data one by one according to the set logical relationship; for the logical AND method, it judges whether the data of all sensors meet the conditions; for the logical OR method, it finds the sensor data that meets the conditions; through logical judgment, the data that meets the requirements can be screened out for merging. The data merging submodule merges the sensor data that meets the conditions into a group of data and saves it as a new data set; through data merging, the data of multiple sensors can be integrated to provide more comprehensive and accurate information.

[0078] In summary, in the process of data integration, this embodiment selects a suitable merging method according to specific application requirements, and screens and integrates data that meets the requirements through logical judgment and data merging, which can improve the accuracy, availability and practicality of the data and provide a reliable foundation for subsequent data analysis and application.

[0079] Example 4: Figure 4 As shown, based on Example 1, the reasoning and judgment component provided by the embodiment of the present invention includes:

[0080] The model building module is responsible for building a random forest model based on the integrated disaster indicator data using a machine learning algorithm. The random forest model learns the laws and characteristics of coal mine disasters, as well as the factors related to and influencing factors of disasters based on historical data and real-time data.

[0081] The inference and prediction module is responsible for using the established random forest model and probability statistics to infer and predict the possibility and severity of coal mine disasters, and evaluate the current disaster risks; according to the prediction results, make corresponding decisions and control measures, carry out disaster prevention and control, and formulate corresponding emergency decision-making plans;

[0082] The feedback optimization module is responsible for providing feedback and optimizing the random forest model based on the decision-making effect, and evaluating and improving the random forest model based on the actual disaster situation and decision-making results.

[0083] The working principle and beneficial effects of the above technical solution are as follows: the model building module of this embodiment uses a machine learning algorithm to build a random forest model based on the integrated disaster indicator data. The random forest model learns the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters based on historical data and real-time data; the inference and prediction module uses the established random forest model to use probability statistics to infer and predict the possibility and severity of coal mine disasters, and evaluate the current disaster risk; according to the predicted results, corresponding decisions and control measures are made to carry out disaster prevention and control, and corresponding emergency decision-making plans are formulated; the feedback optimization module feeds back and optimizes the random forest model based on the decision effect, and evaluates and improves the random forest model based on the actual disaster situation and decision results. The model building module of the above scheme can reveal the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters, and establish an accurate and reliable random forest model by learning historical data and real-time data. The inference prediction module uses probability statistics to infer and predict the possibility and severity of coal mine disasters, assess the current disaster risk, and make corresponding decisions and control measures based on the predicted results, and formulate emergency decision-making plans; accurately predict the probability and severity of coal mine disasters, and provide a scientific basis for decision-making. The feedback optimization module can improve and optimize the random forest model by evaluating the actual disaster situation and decision-making results, improve the accuracy and reliability of the prediction, and continuously optimize the random forest model to make it more adaptable to the actual situation.

[0084] In summary, this embodiment provides a complete disaster prediction and decision support system. By establishing a random forest model and performing inference prediction, the possibility and severity of disasters can be discovered in advance, and corresponding control measures can be taken in time to reduce disaster risks. At the same time, by improving the prediction model through the feedback optimization module, the accuracy and practicality of the prediction can be continuously improved, and the efficiency of the entire system can be further improved. Therefore, the ability of disaster prediction and prevention and control can be improved, and the safety of people's lives and property can be protected. The possibility and severity of coal mine disasters can be speculated, and corresponding decision support can be provided to reduce disaster risks and ensure coal mine safety.

[0085] Example 5: Figure 5 As shown, based on Example 4, the model building module provided in this embodiment of the present invention includes:

[0086] The feature extraction submodule is responsible for receiving the integrated disaster indicator data, collecting data related to coal mine disasters, including historical data and real-time data, cleaning the collected data, and extracting target features;

[0087] The model training submodule is responsible for selecting the random forest machine learning algorithm and training the random forest model using the disaster indicator data set; the trained random forest model is evaluated using the accuracy and recall evaluation indicators;

[0088] The retraining submodule is responsible for learning the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters, updating the data set, and retraining the random forest model based on the target characteristics.

[0089] The working principle and beneficial effects of the above technical solution are as follows: the feature extraction submodule of this embodiment receives the integrated disaster index data, collects data related to coal mine disasters, including historical data and real-time data, cleans the collected data, and extracts target features; the model training submodule is responsible for selecting the random forest machine learning algorithm, and uses the data set of disaster index data to train the random forest model; the trained random forest model is evaluated using the accuracy and recall evaluation indicators; the retraining submodule random forest model learns the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters according to the target features, updates the data set, and realizes the retraining of the random forest model. The feature extraction submodule of the above scheme extracts the features related to coal mine disasters from the integrated disaster index data. Through data cleaning and feature extraction, the factors that have a greater impact on the occurrence of disasters can be screened out, helping the model to learn and predict the laws and characteristics of disasters more accurately. The model training submodule uses the random forest machine learning algorithm to train the disaster index data set and establish a random forest model; through training, the model can learn the laws and characteristics of disasters in historical data and real-time data, so as to predict the occurrence of future disasters. The retraining submodule updates the data set and retrains the random forest model based on the target characteristics and actual disaster situations. Through retraining, the model can be continuously optimized and improved to improve the accuracy and reliability of the prediction.

[0090] In summary, this embodiment improves the accuracy and timeliness of coal mine disaster prediction. Through feature extraction, model training and retraining, a model can be established that can predict the occurrence of coal mine disasters based on historical data and real-time data, and be optimized and improved according to the actual disaster situation; it can help relevant departments and personnel to timely assess the current disaster risks and formulate corresponding emergency decision-making plans, thereby improving the prevention and control capabilities of coal mine disasters and reducing the occurrence and impact of disasters.

[0091] Example 6: Figure 6 As shown, based on Example 5, the model training submodule provided in this embodiment of the present invention includes:

[0092] The quantity confirmation unit is responsible for selecting random forest as the machine learning algorithm and determining the number of decision trees; random forest makes predictions by building multiple decision trees;

[0093] The autonomous sampling unit is responsible for random autonomous sampling from the training set when building each decision tree, that is, sampling the training set with replacement;

[0094] The decision tree construction unit is responsible for randomly sampling samples from the data set when constructing each decision tree. According to the selected target features, the data set is divided into different nodes using the Gini index partitioning criterion, and decision trees are constructed and combined into a random forest. The random forest model is trained using the disaster indicator data set.

[0095] Among them, the impurity of the target feature in the initial data set is calculated, and the Gini index is used to measure the impurity. The calculation formula of the Gini index is: Gini§=1-∑(p i 2 ), where p i The target feature i For each target feature and its corresponding eigenvalue, calculate the weighted Gini index after partitioning using the feature and eigenvalue; based on the calculated weighted Gini index, select the partition feature and eigenvalue that minimizes the Gini index as the basis for partitioning the node.

[0096] The working principle and beneficial effects of the above technical solution are as follows: the quantity confirmation unit of this embodiment selects random forest as the machine learning algorithm and determines the number of decision trees; random forest makes predictions by constructing multiple decision trees; the autonomous sampling unit performs random autonomous sampling from the training set when constructing each decision tree, that is, sampling the training set with replacement; the decision tree construction unit randomly and autonomously samples samples from the data set when constructing each decision tree, divides the data set into different nodes according to the selected target features, constructs decision trees, and combines them into a random forest, and uses the data set of disaster indicator data to train the random forest model. The quantity confirmation unit of the above scheme determines the basic parameters of the random forest model, that is, the number of decision trees used, and the autonomous sampling unit increases the diversity of the model through autonomous sampling to avoid overfitting. Through sampling with replacement, the samples used by each decision tree will be different, thereby increasing the generalization ability of the model. The decision tree construction unit constructs decision trees and combines multiple decision trees into a random forest model. The decision tree uses the Gini index partitioning criterion to divide the data set into different nodes according to the selected target features; by constructing multiple decision trees and training them with disaster indicator data, a powerful random forest model can be obtained for disaster prediction.

[0097] In summary, this embodiment constructs a powerful and generalizable random forest model. By determining the number of decision trees, performing autonomous sampling, and using the Gini index partitioning criterion, the accuracy, generalization ability, and stability of the model can be improved; it can be used for disaster prediction to help people take timely measures to deal with potential disaster risks.

[0098] Example 7: Figure 7 As shown, based on Example 4, the inference prediction module provided by the embodiment of the present invention includes:

[0099] The result output submodule is responsible for using the established random forest model to predict real-time data and obtain the prediction results of the possibility and severity of disasters;

[0100] The analysis and statistics submodule is responsible for comparing the output of the random forest model with the actual data, calculating the probability of coal mine disasters, and counting the frequency of disasters of different severity;

[0101] The risk assessment submodule is responsible for evaluating the current disaster risk and determining the probability and possible severity of the disaster based on the results of probability and statistical analysis.

[0102] The working principle and beneficial effects of the above technical solution are as follows: the result output submodule of this embodiment uses the established random forest model to predict the real-time data and obtain the prediction results of the possibility and severity of the disaster; the analysis and statistics submodule compares the output results of the random forest model with the actual data, calculates the probability of coal mine disasters, and counts the frequency of disasters of different severity; the risk assessment submodule evaluates the current disaster risk based on the results of probability statistical analysis, and determines the probability and possible severity of the disaster. The result output submodule of the above solution can quickly and accurately predict the possible occurrence of disasters based on the input data, providing a basis for subsequent analysis and decision-making. The analysis and statistics submodule can verify and correct the prediction results of the model based on actual data to improve the accuracy of the prediction. The risk assessment submodule can quantitatively evaluate the risk level of current coal mine disasters and provide a scientific basis for the formulation of decision-making and control measures.

[0103] In summary, this embodiment provides a comprehensive disaster prediction and assessment system, which can infer and predict the possibility and severity of coal mine disasters from multiple perspectives; through statistical analysis and risk assessment, it can timely assess the current disaster risk and take corresponding decision-making and control measures to reduce the occurrence of coal mine disasters and ensure the safety of miners. At the same time, it can also provide scientific basis for coal mine managers to help them formulate disaster prevention and emergency measures to improve the safety and production efficiency of coal mines.

[0104] Example 8: Figure 8As shown, based on Example 4, the feedback optimization module provided by the embodiment of the present invention includes:

[0105] The problem analysis submodule is responsible for analyzing the problems of the random forest model based on the evaluation results. The prediction results of the random forest model are compared with the actual data to confirm whether they exceed the error threshold. If the prediction results of the random forest model exceed the error threshold, it means that there is a problem with the random forest model. Increasing the amount of sample data can improve the performance of the random forest model.

[0106] The model update submodule is responsible for retraining based on the improved random forest model; retraining the improved random forest model using the updated data set; and validating the improved random forest model using the validation set;

[0107] The loop iteration submodule is responsible for repeatedly performing the process of evaluating, improving, training, and validating the improved random forest model until the improved random forest model reaches the expected performance indicators.

[0108] The working principle and beneficial effects of the above technical solution are as follows: the problem analysis submodule of this embodiment analyzes the problems existing in the random forest model according to the evaluation results, and compares the prediction results of the random forest model with the actual data to confirm whether it exceeds the error threshold. If the prediction results of the random forest model exceed the error threshold, it indicates that there is a problem with the random forest model, and the performance of the random forest model is improved by increasing the amount of sample data; the model update submodule retrains according to the improved random forest model; retrains the improved random forest model with an updated data set; verifies the improved random forest model with a validation set; and the loop iteration submodule repeatedly performs the process of improving the random forest model evaluation, improvement, training and verification until the improved random forest model reaches the expected performance index. The problem analysis submodule of the above scheme mainly compares with the actual data to confirm whether the prediction results of the model exceed the error threshold. If it exceeds, it indicates that there is a problem with the model; it can accurately identify the shortcomings of the model and provide guidance for the improvement of the model. The model update submodule improves the performance of the random forest model by increasing the amount of sample data or other improvement measures; retrains the model with an updated data set, and verifies the improved model with a validation set; it can update the model to improve the accuracy and prediction ability of the model. The loop iteration submodule gradually optimizes the model through continuous loop iteration to make it more accurate and reliable; it can achieve continuous optimization and improvement of the model and improve the stability and reliability of the model.

[0109] In summary, this embodiment can better assess and predict the risk of coal mine disasters and provide reliable support for decision-making and emergency measures by analyzing the problems of the model, continuously improving the model, and improving the accuracy and prediction ability of the model. At the same time, through a cyclic iteration method, the model is continuously optimized to adapt to the ever-changing actual situation and improve the effect and efficiency of disaster prevention and control.

[0110] Example 9: Fig. 9 As shown, based on Example 1, the human-computer interaction module provided by the embodiment of the present invention includes:

[0111] The visualization submodule is responsible for visualizing the prediction results of the probability and severity of disasters obtained by the reasoning and judgment component, using color coding, marking and heat maps to represent different levels of disaster risks;

[0112] Color coding uses different colors to represent different levels of disaster risk. For example, low-risk areas can be marked as green, medium-risk areas can be marked as yellow, and high-risk areas can be marked as red. Decision makers can intuitively see different levels of risk areas;

[0113] Markers Use markers or icons on the map to indicate different levels of disaster risk. For example, you can use small icons or symbols to indicate the risk level of each area so that decision makers can quickly scan the map to understand the risk level of different areas.

[0114] Heatmaps use heatmaps to show the density and distribution of disaster risks. Heatmaps can present different shades of colors on the map according to the risk levels of different areas, so that decision makers can intuitively understand the distribution of disaster risks;

[0115] The plan display submodule is responsible for sending the emergency decision-making plan based on the prediction results to the decision maker by displaying the risk area through a map and the emergency process through a flowchart;

[0116] Among them, on the map, risk areas can be marked by boundaries or overlays, so that decision makers can clearly understand which areas have higher disaster risks; the emergency process is displayed through a flowchart, which can show the execution order and steps of various emergency measures, as well as the resources and personnel dispatch that may be involved. Through the display of the flowchart, decision makers can better understand the execution path and related elements of the emergency decision-making plan;

[0117] The decision adjustment submodule is responsible for enabling decision makers to interact with the visualization interface through the human-computer interaction interface, understand the detailed information of monitoring data and prediction results, and select and adjust decision plans.

[0118] The working principle and beneficial effects of the above technical solution are as follows: the visualization display submodule of this embodiment uses color coding, marking and heat map to represent different levels of disaster risks in a visual form, using the predicted results of the possibility and severity of disasters obtained by the reasoning and judgment component; wherein, color coding uses different colors to represent different levels of disaster risks. For example, low-risk areas can be marked as green, medium-risk areas can be marked as yellow, and high-risk areas can be marked as red, so that decision makers can intuitively see risk areas of different levels; marking on the map uses marks or icons to represent different levels of disaster risks. For example, small icons or symbols can be used to represent the risk level of each area, so that decision makers can understand the risk situation of different areas by quickly scanning the map. Heat maps use heat maps to display the density and distribution of disaster risks. Heat maps can present different shades of colors on the map according to the risk levels of different areas, so that decision makers can intuitively understand the distribution of disaster risks; the plan display submodule sends the emergency decision plan based on the prediction results to the decision maker by displaying the risk area on the map and the emergency process through the flowchart; Among them, on the map, the risk area can be marked with boundaries or overlays so that decision makers can clearly understand which areas have a higher disaster risk; the emergency process is displayed through a flowchart, which can show the execution order and steps of various emergency measures, as well as the resources and personnel scheduling that may be involved. Through the display of the flowchart, decision makers can better understand the execution path and related elements of the emergency decision plan; the decision adjustment submodule uses the human-computer interaction interface to allow decision makers to interact with the visualization interface, understand the detailed information of the monitoring data and prediction results, and select and adjust the decision plan. The visualization submodule of the above scheme displays the disaster risk prediction results in a visual form, using color coding, marking and heat maps to represent different levels of disaster risks; through this visualization, decision makers can intuitively understand the risk situation in different areas, quickly identify high-risk areas, and make corresponding decisions. The scheme display submodule displays the emergency decision-making scheme based on the prediction results through maps and flowcharts, so that decision makers can clearly understand the risk areas and emergency processes. By displaying the risk areas on the map, decision makers can directly see which areas have a higher disaster risk, so as to determine the scope and focus of emergency measures; by displaying the emergency process through flowcharts, decision makers can understand the execution order and steps of emergency measures, as well as the resources and personnel scheduling involved, so as to better understand and evaluate the feasibility and effectiveness of emergency decision-making schemes.The decision adjustment submodule provides detailed monitoring data and prediction results through the human-computer interaction interface, so that decision makers can have an in-depth understanding of them and select and adjust decision plans. Decision makers can view detailed information on monitoring data and prediction results, such as specific values ​​and trend changes of disaster risks, through interaction with the visualization interface, so as to more accurately assess the risk situation and adjust the decision plan as needed.

[0119] In summary, this embodiment helps decision makers better understand and assess disaster risk situations, quickly make emergency decisions, and make adjustments when necessary; through visual display and interactive interface, it provides intuitive, accurate and convenient information display and operation methods, improves decision makers' decision efficiency and quality, thereby effectively responding to disaster risks and reducing losses and risks.

[0120] Example 10: Fig.10 As shown, based on Example 1 to Example 9, the coal mine disaster fusion monitoring and intelligent decision-making method provided by the embodiment of the present invention includes the following steps:

[0121] S100: Through monitoring equipment, real-time data collection is performed on five types of disaster indicators in the internal environment of the coal mine, including water, fire, gas, roof and dust, and the collected disaster indicator data is integrated to obtain the integrated disaster indicator data;

[0122] S200: Based on the integrated disaster indicator data, use information reasoning and intelligent analysis to speculate the possibility and severity of coal mine disasters and predict possible disaster risks; make emergency decisions based on the prediction results, including taking corresponding measures for disaster prevention and control;

[0123] S300: Through the human-computer interaction interface and visualization technology, monitoring data, prediction results and decision-making plans are displayed in a graphical way, allowing decision makers to understand the disaster status, monitoring data and prediction results and make interactive decisions.

[0124] The working principle and beneficial effects of the above technical solution are as follows: first, the present embodiment collects real-time data of five disaster indicators, such as water, fire, gas, roof and dust, in the internal environment of the coal mine through monitoring equipment, integrates the collected disaster indicator data, and obtains the integrated disaster indicator data; secondly, based on the integrated disaster indicator data, information reasoning and intelligent judgment are used to speculate the possibility and severity of coal mine disasters and predict possible disaster risks; according to the predicted results, emergency decisions are made, including taking corresponding measures to prevent and control disasters; finally, through the human-computer interaction interface, through visualization technology, the monitoring data, prediction results and decision-making schemes are displayed in a graphical manner, so that decision makers can understand the disaster status, monitoring data and prediction results, and make interactive decisions. Step S100 of the above scheme collects and integrates real-time data, collects data of five disaster indicators, such as water, fire, gas, roof and dust, in the internal environment of the coal mine through monitoring equipment, and integrates the collected data into a comprehensive disaster indicator data, which can provide real-time and comprehensive disaster indicator data, and provide a basis for subsequent analysis and decision-making. Step S200: Information reasoning and intelligent judgment. By using reasoning algorithms and intelligent technologies to analyze and infer the integrated disaster indicator data, possible disaster risks can be predicted, which can provide prediction results of the possibility and severity of disasters, and provide reference for decision makers to make emergency decisions. Step S300: Visual display and human-computer interaction. Through visualization technology, the monitoring data and prediction results are displayed graphically on the human-computer interaction interface, so that decision makers can intuitively understand the disaster status, monitoring data and prediction results, which can help decision makers better understand and analyze the situation, make interactive decisions, and improve decision efficiency and accuracy.

[0125] To sum up, the significance of the entire coal mine disaster integrated monitoring and intelligent decision-making method in this embodiment lies in providing comprehensive and accurate coal mine disaster monitoring and prediction information through technical means such as real-time data collection, information reasoning and visualization display, helping decision makers to make emergency decisions in a timely manner and take corresponding measures to prevent and control disasters; it can improve the safety and production efficiency of coal mines, reduce the occurrence of coal mine disasters and accidents, and protect the lives and property safety of miners.

[0126] Embodiment 11: Based on Embodiment 10, the disaster indicator data integration process provided in the embodiment of the present invention comprises the following steps:

[0127] S101: Various sensors are deployed inside the coal mine to monitor various disaster indicators of the internal environment of the coal mine in real time;

[0128] S102: various disaster indicators collected by various sensors are transmitted to wireless communication nodes through wireless communication technology to achieve remote acquisition and real-time monitoring of data;

[0129] S103: The wireless communication node obtains various disaster indicators, processes missing values ​​for the collected data of various disaster indicators, aligns the data from different data sources in time to have consistent timestamps, and uses a decision-level fusion method to merge the preprocessed and aligned data to obtain integrated disaster indicator data.

[0130] The working principle and beneficial effects of the above technical solution are as follows: In this embodiment, various sensors are first deployed inside the coal mine to monitor various disaster indicators of the internal environment of the coal mine in real time; secondly, various disaster indicators collected by various sensors are transmitted to wireless communication nodes through wireless communication technology to realize remote acquisition and real-time monitoring of data; finally, the wireless communication node obtains various disaster indicators, performs missing value processing on the data collected for various disaster indicators, aligns the data from different data sources in time, has consistent timestamps, and uses decision-level fusion method to merge the preprocessed and aligned data to obtain integrated disaster indicator data. Step S101 of the above solution realizes real-time monitoring of the internal environment of the coal mine by deploying various sensors inside the coal mine to monitor various disaster indicators of the internal environment of the coal mine in real time. Step S102 transmits various disaster indicators collected by various sensors to wireless communication nodes through wireless communication technology, realizes remote acquisition and real-time monitoring of data, and improves the efficiency and accuracy of data acquisition. In step S103, the wireless communication node obtains various disaster indicators, processes missing values ​​for the collected data of various disaster indicators, aligns the data from different data sources in time, and merges the preprocessed and aligned data through a decision-level fusion method to obtain integrated disaster indicator data, thereby realizing the integration and merging of disaster indicator data and ensuring the consistency and integrity of the data.

[0131] To sum up, this embodiment provides real-time, accurate and complete coal mine disaster indicator data, which provides a reliable data basis for subsequent disaster prediction and decision-making; through real-time monitoring of the internal environment of coal mines and integration of disaster indicator data, disaster risks can be discovered in a timely manner, possible disasters can be predicted, and a scientific basis can be provided for taking corresponding prevention and control measures, thereby minimizing the occurrence of coal mine disasters and ensuring the life safety of coal miners.

[0132] Embodiment 12: Based on Embodiment 10, the process of estimating the possibility and severity of coal mine disasters provided in the embodiment of the present invention comprises the following steps:

[0133] S201: Based on the integrated disaster indicator data, a random forest model is established using a machine learning algorithm. The random forest model learns the laws and characteristics of coal mine disasters, as well as the factors related to and influencing factors of disasters based on historical data and real-time data;

[0134] S202: Using the established random forest model, probability statistics are used to infer and predict the possibility and severity of coal mine disasters, and to assess the current disaster risk; based on the predicted results, corresponding decisions and control measures are taken to carry out disaster prevention and control, and corresponding emergency decision-making plans are formulated;

[0135] S203: Provide feedback and optimize the random forest model based on the decision-making effect, and evaluate and improve the random forest model based on the actual disaster situation and decision-making results.

[0136] The working principle and beneficial effects of the above technical solution are as follows: First, based on the integrated disaster indicator data, this embodiment uses a machine learning algorithm to establish a random forest model. The random forest model learns the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters based on historical data and real-time data; secondly, using the established random forest model, probability statistics are used to infer and predict the possibility and severity of coal mine disasters, and evaluate the current disaster risk; according to the predicted results, corresponding decisions and control measures are made, disaster prevention and control are carried out, and corresponding emergency decision-making plans are formulated; finally, according to the decision effect, the random forest model is fed back and optimized, and the random forest model is evaluated and improved according to the actual disaster situation and decision results. Step S201 of the above scheme establishes a random forest model, which can learn the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters, and improve the accuracy and prediction ability of the possibility and severity of disasters. Step S202 can formulate corresponding decisions and control measures according to the prediction results, carry out disaster prevention and control, and formulate emergency decision-making plans, which helps to take measures in time to minimize the occurrence and impact of disasters. Step S203 helps to improve the accuracy and predictive ability of the model, so that it can better adapt to changes in the coal mine environment and the characteristics of disasters.

[0137] In summary, this embodiment provides a method based on data and machine learning, which can speculate and predict the possibility and severity of coal mine disasters; through these technologies, the effect of disaster prevention and control can be improved, and the life safety of coal mine workers and the stability of coal mine production can be guaranteed.

[0138] Embodiment 13: Based on Embodiment 10, the interactive decision-making process provided by the embodiment of the present invention includes the following steps:

[0139] S301: The prediction results of the disaster probability and severity obtained by the reasoning and judgment component are displayed in a visual form by using color coding, marking and heat map to represent different levels of disaster risks;

[0140] S302: Sending an emergency decision plan based on the prediction results to the decision maker by displaying the risk area on a map and the emergency process through a flowchart;

[0141] S303: Through the human-computer interaction interface, the decision maker interacts with the visualization interface to understand the detailed information of the monitoring data and the prediction results, and selects and adjusts the decision plan.

[0142] The working principle and beneficial effects of the above technical solution are as follows: First, the prediction results of the possibility and severity of disasters obtained by the reasoning and judgment component are displayed in a visual form by using color coding, marking and heat map to represent different levels of disaster risks; secondly, the emergency decision-making plan based on the prediction results is sent to the decision maker by displaying the risk area on the map and the emergency process through the flowchart; finally, through the human-computer interaction interface, the decision maker interacts with the visualization interface to understand the detailed information of the monitoring data and the prediction results, and selects and adjusts the decision-making plan. Step S301 of the above solution: By color coding, marking and heat map, the distribution of disaster risks can be intuitively displayed, helping decision makers to quickly understand the severity and possibility of disaster risks. Step S302 displays the risk area on the map and displays the emergency process through the flowchart, and conveys the emergency decision-making plan based on the prediction results to the decision maker, helping the decision maker to intuitively understand the risk area and emergency measures, so as to make corresponding decisions and actions. Step S303: Through the human-computer interaction interface, decision makers can interact with the visualization interface, view detailed information on monitoring data and forecast results, select and adjust decision plans, and provide instant interaction between decision makers and the system, helping decision makers to understand disaster risks and response measures more comprehensively and accurately, and improve the accuracy and efficiency of decision-making.

[0143] In summary, this embodiment provides decision makers with an intuitive, comprehensive, and real-time view to help them understand disaster risks and emergency decision-making plans; it provides decision makers with effective tools and information to enable them to better respond to disaster risks and take timely measures to reduce the losses and impacts caused by disasters. At the same time, the interactive decision-making process can also improve the flexibility and accuracy of decision-making and better adapt to different disaster situations and needs.

[0144] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A coal mine disaster fusion monitoring and intelligent decision-making system, characterized in that: Include: The indicator acquisition component is responsible for collecting real-time data on the disaster indicators of the internal environment of the coal mine through monitoring equipment, integrating the collected disaster indicator data, and obtaining the integrated disaster indicator data; The reasoning and judgment component is responsible for inferring the possibility and severity of coal mine disasters and predicting possible disaster risks based on the integrated disaster indicator data using information reasoning and intelligent judgment; The human-computer interaction component is responsible for displaying monitoring data, prediction results and decision-making plans in a graphical way through the human-computer interaction interface and visualization technology, so as to make interactive decisions.

2. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 1, characterized in that: Reasoning and judgment components, including: The model building module is responsible for building a random forest model based on the integrated disaster indicator data using a machine learning algorithm. The random forest model learns the laws and characteristics of coal mine disasters, as well as the factors related to and influencing factors of disasters based on historical data and real-time data. The inference and prediction module is responsible for using the established random forest model and probability statistics to infer and predict the possibility and severity of coal mine disasters, and evaluate the current disaster risks; according to the prediction results, make corresponding decisions and control measures, carry out disaster prevention and control, and formulate corresponding emergency decision-making plans; The feedback optimization module is responsible for providing feedback and optimizing the random forest model based on the decision-making effect, and evaluating and improving the random forest model based on the actual disaster situation and decision-making results.

3. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 2, characterized in that: Model building module, including: The feature extraction submodule is responsible for receiving the integrated disaster indicator data, collecting data related to coal mine disasters, including historical data and real-time data, cleaning the collected data, and extracting target features; The model training submodule is responsible for selecting the random forest machine learning algorithm and training the random forest model using the disaster indicator data set; the trained random forest model is evaluated using the accuracy and recall evaluation indicators; The retraining submodule is responsible for learning the laws and characteristics of coal mine disasters, as well as the factors and influencing factors related to the occurrence of disasters, updating the data set, and retraining the random forest model based on the target characteristics.

4. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 3, characterized in that: Model training submodule, including: The quantity confirmation unit is responsible for selecting random forest as the machine learning algorithm and determining the number of decision trees; random forest makes predictions by building multiple decision trees; The autonomous sampling unit is responsible for random autonomous sampling from the training set when building each decision tree, that is, sampling the training set with replacement; The decision tree construction unit is responsible for randomly and autonomously sampling samples from the data set when building each decision tree. According to the selected target features, the data set is divided into different nodes using the Gini index partitioning criterion, and decision trees are constructed and combined into a random forest. The random forest model is trained using the disaster indicator data set.

5. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 2, characterized in that: Inference prediction module, including: The result output submodule is responsible for using the established random forest model to predict real-time data and obtain the prediction results of the possibility and severity of disasters; The analysis and statistics submodule is responsible for comparing the output of the random forest model with the actual data, calculating the probability of coal mine disasters, and counting the frequency of disasters of different severity; The risk assessment submodule is responsible for evaluating the current disaster risk and determining the probability and possible severity of the disaster based on the results of probability and statistical analysis.

6. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 2, characterized in that: Feedback optimization module, including: The problem analysis submodule is responsible for analyzing the problems existing in the random forest model based on the evaluation results. The prediction results of the random forest model are compared with the actual data to confirm whether they exceed the error threshold. The model update submodule is responsible for retraining based on the improved random forest model; Improve the random forest model by retraining it with an updated dataset; Use the validation set to validate the improved random forest model; The loop iteration submodule is responsible for repeatedly performing the process of evaluating, improving, training, and validating the improved random forest model until the improved random forest model reaches the expected performance indicators.

7. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 6, characterized in that: In the problem analysis submodule, if the prediction result of the random forest model exceeds the error threshold, it means that there is a problem with the random forest model. Increasing the amount of sample data can improve the performance of the random forest model.

8. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 1, characterized in that: Human-computer interaction module, including: The visualization submodule is responsible for displaying the prediction results of the probability and severity of disasters obtained by the reasoning and judgment component; The plan display submodule is responsible for sending the emergency decision-making plan based on the prediction results to the decision maker by displaying the risk area through a map and the emergency process through a flowchart; The decision adjustment submodule is responsible for enabling decision makers to interact with the visualization interface through the human-computer interaction interface, understand the detailed information of monitoring data and prediction results, and select and adjust decision plans.

9. The coal mine disaster fusion monitoring and intelligent decision-making system according to claim 8, characterized in that: The visualization submodule uses color coding, marking and heat map to represent different levels of disaster risks in a visual form.

10. A coal mine disaster fusion monitoring and intelligent decision-making method, characterized in that: The following steps are involved: Through monitoring equipment, real-time data collection is performed on the five disaster indicators of water, fire, gas, roof and dust in the internal environment of the coal mine, and the collected disaster indicator data is integrated to obtain the integrated disaster indicator data; Based on the integrated disaster indicator data, information reasoning and intelligent judgment are used to speculate the possibility and severity of coal mine disasters and predict possible disaster risks; based on the prediction results, emergency decisions are made, including taking corresponding measures for disaster prevention and control; Through the human-computer interaction interface and visualization technology, monitoring data, prediction results and decision-making plans are displayed in a graphical way, allowing decision makers to understand the disaster status, monitoring data and prediction results and make interactive decisions.

Images