Coal mine production process management and control system

Abstract

The invention provides a coal mine production process management and control system, and relates to the technical field of process management, and the system comprises a data collection module which is used for obtaining the data of each production process in a mine field in real time, and outputting initial data; the preprocessing module is used for preprocessing the initial data to obtain to-be-analyzed data; the model training module is used for training a to-be-trained model selected from the model database in combination with a preset data resource, outputting an alternative model, and outputting the alternative model meeting a preset performance condition as an analysis model; and the data analysis module is used for analyzing the to-be-analyzed data based on the analysis model, outputting a process analysis result, and performing parameter updating and performance optimization on the analysis model in combination with preset data resources and cloud data. According to the invention, data of each production flow in the mine field can be monitored in real time, potential risks of each production flow can be found in time, parameter updating and performance optimization are continuously carried out on the model, and the real-time performance and accuracy of a flow analysis result are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of process management, and in particular to a coal mine production process control system. Background Art

[0002] The coal mine production process is a complex and semi-continuous industrial process, covering tunneling, coal mining, belt conveyor, buffer coal bunker, hoisting (or inclined shaft transportation), washing, loading and other links. Each link is closely related, and any abnormality in any link may lead to the interruption of the entire production chain, which in turn has a direct impact on the overall production efficiency of the coal mine. Ensuring the smooth and reasonable operation of the coal mine production line is not only a need for the daily output of the mine, but also affects the safe production of the entire coal mine.

[0003] Traditional production process control often adopts manual experience, overly relying on the subjective experience of dispatchers, and has obvious lag. When a process is abnormal, it will be fed back to the process management personnel, and only after reporting the decision-making department can it be adjusted. This model is slow to respond and lacks a global optimization perspective. When faced with an increasingly complex production environment, the limitations of the traditional process control model are becoming more and more prominent, making it difficult to effectively respond to emergencies in the production process.

[0004] Therefore, the present invention provides a coal mine production process control system. Summary of the invention

[0005] The present invention provides a coal mine production process control system, which is used to monitor the data of each production process in the mine in real time, and perform real-time analysis and predictive analysis on the data through an analysis model to promptly discover the potential risks of each production process. The model can also be continuously updated with parameters and performance optimization to ensure the real-time and accuracy of the process analysis results, thereby ensuring production safety.

[0006] The present invention provides a coal mine production process control system, comprising: Data acquisition module, used to obtain data of various production processes in the mine in real time and output initial data; A preprocessing module, used to preprocess the initial data using a preset preprocessing method to obtain data to be analyzed; A model training module is used to train the model to be trained selected from the model database in combination with preset data resources, output candidate models, and output the candidate models that meet preset performance conditions as analysis models; The data analysis module is used to analyze the data to be analyzed based on the analysis model, output the process analysis results, and update the parameters and optimize the performance of the analysis model based on the process analysis results and in combination with the preset data resources and cloud data.

[0007] Preferably, the data acquisition module comprises: The process classification submodule is used to divide the production processes in the mine according to the monitoring priority of each production process and obtain a process classification table; The data acquisition submodule is used to obtain the data of each production process in the process classification table in real time by using the preset monitoring equipment set at the preset position, and output the initial data.

[0008] Preferably, the data acquisition submodule includes: A demand acquisition unit, used to acquire the monitoring requirements corresponding to each production process, and summarize and output the system monitoring requirements in combination with the monitoring priorities corresponding to each production process; An equipment selection unit is used to obtain preset monitoring equipment corresponding to each production process based on the system monitoring requirements and in combination with the preset requirements-equipment matching table, determine the preset location for installing each preset monitoring equipment, and output a process-equipment comparison table; The data acquisition unit is used to obtain the data of each production process in the mine in real time through the preset monitoring equipment at each preset position based on the process-equipment comparison table, and comprehensively output the initial data.

[0009] Preferably, the preprocessing module comprises: A method selection submodule is used to extract features from the initial data, construct a first feature set based on the extracted features, and obtain a preset preprocessing method matching the first feature set in combination with a preset feature-method mapping table; The preprocessing submodule is used to preprocess the initial data based on the preset preprocessing method to obtain first data, and output the first data that meets the first threshold condition as data to be analyzed.

[0010] Preferably, the model training module includes: A model selection submodule is used to select a model to be trained that matches the monitoring requirements of the system in the model database in combination with a preset requirement-model matching table; A model training submodule, used to select preset data resources from a preset resource pool, train the to-be-trained model based on the preset data resources, and output a candidate model; The model output submodule is used to obtain the performance requirement information of the model, output the preset performance conditions based on the performance requirement information, and output the candidate model that meets the preset performance judgment conditions as the analysis model.

[0011] Preferably, the model training submodule includes: A data selection unit, used to obtain a data selection instruction, and select a corresponding preset data resource in a target resource pool based on the data selection instruction; A data partitioning unit, used to partition the preset data resources using a preset data partitioning method, and output a training data set, a verification data set, and a test data set; The model training unit is used to train and optimize the performance of the model to be trained based on the training data set, the verification data set and the test data set, and output an alternative model.

[0012] Preferably, the data analysis module includes: A data analysis submodule, used to analyze the data to be analyzed to obtain sub-data corresponding to each production process; The level information acquisition submodule is used to obtain the monitoring level information of each sub-data and summarize and output the process monitoring level table; A comparison table generation submodule, for combining the process monitoring level table and the sub-data corresponding to each production process under each monitoring level, and outputting a monitoring level-data comparison table; The abnormality identification submodule is used to extract features from the sub-data of the production process at each monitoring level in the monitoring level-data comparison table, and construct a second feature set, and perform feature matching on the data features in the second feature set in combination with a preset abnormality feature database to obtain abnormality identification sub-results corresponding to the sub-data of the production process at each monitoring level, and summarize and output abnormality analysis results; A historical data acquisition submodule, used to select from a historical database first historical process data matching the data numerical feature in the second feature set and second historical process data matching the data temporal feature in the second feature set; a format conversion submodule, for obtaining data format information of the sub-data, the first historical process data, and the second historical process data corresponding to each monitoring level, and performing format conversion on the sub-data, the first historical process data, and the second historical process data corresponding to each monitoring level in combination with the format requirement information of the analysis model, to obtain first standard data, second standard data, and third standard data, respectively; A real-time analysis submodule, configured to perform real-time analysis on the first standard data corresponding to the production process at each monitoring level based on the analysis model, and output a real-time analysis result; A prediction analysis submodule, used to perform prediction analysis on the second standard data and the third standard data based on the analysis model, obtain prediction data of each production process at each monitoring level within a future preset time period, and output the prediction analysis results; A comprehensive analysis submodule is used to analyze the abnormal analysis results, real-time analysis results and predictive analysis results using a preset comprehensive analysis method, and output the comprehensive analysis results; A historical prediction data acquisition submodule is used to acquire historical prediction data matching the analysis model in a historical database; A result verification submodule is used to perform time calibration and data verification on the historical prediction data and the real-time analysis results under each same production process, and output the prediction data verification results; The model optimization submodule is used to optimize the comprehensive analysis results and the prediction data verification results in combination with preset optimization indicators, and adjust the parameters and optimize the performance of the analysis model based on the optimization analysis results.

[0013] Preferably, the model optimization submodule includes: An analysis unit, configured to analyze the comprehensive analysis result to obtain a first sub-result corresponding to the abnormal analysis result, a second sub-result corresponding to the real-time analysis result, and a third sub-result corresponding to the predictive analysis result; A feature extraction unit, used to extract features from the first sub-result, the second sub-result, the third sub-result and the prediction data verification result, and construct an abnormal feature set, a real-time feature set, a prediction feature set and a verification feature set based on the extracted features; An indicator selection unit, used to select preset optimization indicators in an indicator database, and construct a first indicator set matching the abnormal feature set, a second indicator set matching the real-time feature set, a third indicator set matching the prediction feature set, and a fourth indicator set matching the verification feature set; A relationship graph acquisition unit, used to acquire a mapping relationship between preset optimization indicators in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set, and output an indicator relationship graph; an optimization analysis unit, configured to perform optimization analysis on the comprehensive analysis result and the prediction data verification result based on the first indicator set, the second indicator set, the third indicator set, the fourth indicator set and the indicator relationship diagram to obtain an optimization analysis result; ; in, Indicates the optimization analysis results; represents the exponential function; Represents a comprehensive correction coefficient corresponding to the data calculation error under each preset optimization indicator in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set; represents the total number of preset optimization indicators in the i-th indicator set; i=1 corresponds to the first indicator set; i=2 corresponds to the second indicator set; i=3 corresponds to the third indicator set; i=4 corresponds to the fourth indicator set; represents the optimization standard parameter corresponding to the jth preset optimization indicator in the i-th indicator set; represents the actual parameter value corresponding to the jth preset optimization indicator in the ith indicator set in the comprehensive analysis results and the prediction data verification results; represents the optimization weight coefficient corresponding to the jth preset optimization indicator in the i-th indicator set; represents the complex correlation coefficient between the jth preset optimization indicator in the i-th indicator set and the remaining t-1 preset optimization indicators in each indicator set, and t represents the total number of preset optimization indicators in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set; The model optimization unit is used to combine the preset result-strategy-instruction comparison table to obtain the model parameter performance optimization strategy and the optimization instructions under the corresponding strategy that match the optimization analysis results, and adjust the parameters and optimize the performance of the analysis model based on the model parameter performance optimization strategy and the optimization instructions.

[0014] The present invention provides a coal mine production process control system, which realizes monitoring and optimization of the coal mine production process through the collaboration of data collection, preprocessing, model training and data analysis modules. From initial data to data to be analyzed and then to the training and optimization of the analysis model, the system continuously improves itself to better adapt to changes in the coal mine production environment. The present invention can monitor each production process in real time and provide accurate process analysis results to help managers make more accurate decisions. It can also realize real-time optimization of the model, improve the production efficiency of each production process, and ensure the efficient operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a framework schematic diagram of a coal mine production process control system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] like Figure 1As shown, a coal mine production process control system provided by an embodiment of the present invention includes: Data acquisition module, used to obtain data of various production processes in the mine in real time and output initial data; A preprocessing module is used to preprocess the initial data using a preset preprocessing method to obtain data to be analyzed; A model training module is used to train the model to be trained selected from the model database in combination with preset data resources, output candidate models, and output the candidate models that meet preset performance conditions as analysis models; The data analysis module is used to analyze the data to be analyzed based on the analysis model, output the process analysis results, and update the parameters and optimize the performance of the analysis model based on the process analysis results and in combination with preset data resources and cloud data.

[0019] In this embodiment, initial data refers to the original data of each production process in the mine obtained from the data acquisition module, including but not limited to the equipment operation parameters of each production process in the mine and the temperature, humidity, gas concentration and other parameters of the surrounding environment; In this embodiment, the preset preprocessing method: a preset data preprocessing method is used to process the initial data, including but not limited to data cleaning, denoising, normalization and other operations; In this embodiment, the data to be analyzed: data prepared for analysis obtained after being processed by the preprocessing module; In this embodiment, preset data resources: pre-set data resources used for model training and data analysis, including but not limited to real-time process data and historical process data corresponding to each production process; In this embodiment, the model database: a database storing various models, from which the model training module selects the model to be trained; In this embodiment, the model to be trained: the model to be trained selected from the model database; In this embodiment, the candidate model: a model that meets the preset performance conditions after being trained by the model training module; In this embodiment, the preset performance condition: a preset model performance index is used to screen out candidate models that meet the condition; In this embodiment, the analysis model: the candidate model after screening is used by the data analysis module to analyze the data to be analyzed; In this embodiment, process analysis result: the result obtained by the data analysis module after analyzing the data to be analyzed based on the analysis model; In this embodiment, cloud data: data resources stored in the cloud, used to update parameters and optimize performance of the analysis model in combination with preset data resources and process analysis results; In this embodiment, parameter update and performance optimization: based on the cloud data and process analysis results, the parameters of the analysis model are updated and the performance is optimized to improve the accuracy and efficiency of the model.

[0020] The implementation principle and beneficial effects of this embodiment: The present invention realizes the monitoring and optimization of the coal mine production process through the collaboration of data collection, preprocessing, model training and data analysis modules. From the initial data to the data to be analyzed and then to the training and optimization of the analysis model, the system continuously improves itself to better adapt to the changes in the coal mine production environment. The present invention can monitor each production process in real time and provide accurate process analysis results to help managers make more accurate decisions. It can also realize real-time optimization of the model, improve the production efficiency of each production process, and ensure the efficient operation of the system.

[0021] The embodiment of the present invention provides a coal mine production process control system, a data acquisition module, including: The process classification submodule is used to divide the production processes in the mine according to the monitoring priority of each production process and obtain a process classification table; The data acquisition submodule is used to obtain the data of each production process in the process classification table in real time by using the preset monitoring equipment set at the preset position, and output the initial data.

[0022] In this embodiment, monitoring priority: a priority order set according to the importance and urgency of the production process; In this embodiment, the process classification table: a table divided according to the monitoring priority and the characteristics of the production process, used to record the monitoring priority of each production process; In this embodiment, preset position: pre-set position information, used to determine the layout position of the monitoring equipment to ensure the comprehensiveness and effectiveness of data collection; In this embodiment, preset monitoring equipment: pre-set monitoring equipment is used to obtain data of each production process in real time for subsequent analysis and control. For example, in a coal mine production scenario, the monitoring priority is set to coal mining process as the highest, mine ventilation process as the second, and mine drainage process as the lowest. According to this priority order, the system establishes a process classification table and installs corresponding monitoring equipment at preset locations. These devices obtain data of each process in real time. The system collects and manages data in a timely manner according to the monitoring priority to ensure the smooth progress of the mine production process.

[0023] The implementation principle and beneficial effects of this embodiment: The present invention realizes the real-time acquisition and classification management of each production process data in the mine through the collaboration of the process classification submodule and the data acquisition submodule. The monitoring priority is used to determine the working order of the monitoring equipment, and the process classification table helps the system to effectively classify and manage the data. The present invention can give priority to monitoring and acquiring the data of key production processes to ensure the stability and safety of the production process. At the same time, it helps the system to classify and manage the data of each production process, facilitate subsequent analysis and decision-making, respond to changes in the production environment in a timely manner, and improve the efficiency and accuracy of production control.

[0024] The embodiment of the present invention provides a coal mine production process control system, a data collection submodule, including: The demand acquisition unit is used to obtain the monitoring requirements corresponding to each production process, and summarize and output the system monitoring requirements in combination with the monitoring priorities corresponding to each production process; The equipment selection unit is used to obtain the preset monitoring equipment corresponding to each production process based on the system monitoring requirements and in combination with the preset requirements-equipment matching table, determine the preset location for installing each preset monitoring equipment, and output the process-equipment comparison table; The data acquisition unit is used to obtain the data of each production process in the mine in real time through the preset monitoring equipment at each preset location based on the process-equipment comparison table, and comprehensively output the initial data.

[0025] In this embodiment, monitoring requirements: the requirements of each production process for monitoring data, including the parameters, frequency, accuracy, etc. that need to be monitored; In this embodiment, the system monitoring demand: summarizes the demand for monitoring data of each production process, which is the comprehensive demand of the entire system for monitoring data; In this embodiment, the preset requirement-equipment matching table: a table containing the mapping relationship between monitoring requirements and monitoring equipment, which is pre-set and helps the equipment selection unit determine the preset monitoring equipment corresponding to each production process and the installation location of these equipment; In this embodiment, the process-equipment comparison table records the correspondence between each production process and the preset monitoring equipment, and is used to guide the data acquisition unit to obtain the data of each production process in real time. For example, in a coal mine production scenario, the monitoring requirements include the need to monitor temperature and humidity in the mine ventilation process, and the need to monitor geological parameters in the coal mining process. Based on these monitoring requirements, the system combines the monitoring priority and the preset requirements-equipment matching table to determine the monitoring equipment and installation location corresponding to each production process, and establishes a process-equipment comparison table. The data acquisition unit obtains the data of each production process in real time through this table to ensure that the system can effectively monitor and control the mine production process.

[0026] The implementation principle and beneficial effects of this embodiment: The present invention determines the monitoring equipment and location according to the monitoring requirements and obtains the data of each production process in real time through the collaboration of the demand acquisition unit, the equipment selection unit and the data acquisition unit. The present invention can provide a customized monitoring solution according to the monitoring requirements of each production process to ensure the accuracy and comprehensiveness of the monitoring data. By presetting the demand-equipment matching table and the process-equipment comparison table, the selection and installation location of the monitoring equipment can be continuously optimized to improve the efficiency of data collection.

[0027] An embodiment of the present invention provides a coal mine production process control system, a preprocessing module, including: A method selection submodule is used to extract features from the initial data, construct a first feature set based on the extracted features, and obtain a preset preprocessing method matching the first feature set in combination with a preset feature-method mapping table; The preprocessing submodule is used to preprocess the initial data based on a preset preprocessing method to obtain first data, and output the first data that meets the first threshold condition as data to be analyzed.

[0028] In this embodiment, the first feature set: a feature set constructed by extracting features from the initial data through the method selection submodule, which is used to describe the important features of the data. For example, for vibration data in coal mine production, the system will extract features such as frequency and amplitude from the original vibration data, and then select a suitable preprocessing method for data processing according to a preset feature-method mapping table, and finally output data that meets the threshold conditions for subsequent analysis; In this embodiment, the preset feature-method mapping table is used to record the mapping relationship between the preset features and the preset preprocessing methods, so as to help the system select an appropriate preprocessing method; In this embodiment, first data: data obtained after preprocessing; In this embodiment, the first threshold condition is a condition used to filter data. Data that meets the condition will be output as data to be analyzed. It is usually used to filter out data with higher quality or that meets specific standards.

[0029] Implementation principle and beneficial effects of this embodiment: The present invention extracts features and preprocesses the initial data through the collaboration of the method selection submodule and the preprocessing submodule to obtain data that meets the analysis requirements. The preset feature-method mapping table can help the system select a suitable preprocessing method to ensure the accuracy and effectiveness of data processing. By setting the first threshold condition, the system can automatically screen the data, further analyze the data that meets the threshold condition, and improve the quality and availability of the data.

[0030] The embodiment of the present invention provides a coal mine production process control system, a model training module, including: The model selection submodule is used to select a model to be trained that matches the system monitoring requirements in the model database in combination with the preset requirements-model matching table; The model training submodule is used to select preset data resources from a preset resource pool, train the model to be trained based on the preset data resources, and output a candidate model; The model output submodule is used to obtain the performance requirement information of the model, output the preset performance conditions based on the performance requirement information, and output the alternative models that meet the preset performance judgment conditions as analysis models.

[0031] In this embodiment, the preset demand-model matching table records the matching relationship between the system monitoring demand and the model, and helps the system select a suitable model to be trained according to the monitoring demand; In this embodiment, the preset resource pool: stores preset data resources, which are used for the model training submodule to select data for model training; In this embodiment, performance requirement information: performance requirements that the model needs to meet during training and application, including but not limited to accuracy, recall rate, error range, etc. Specifically, for example, a neural network model is selected for coal mine production data prediction, and the model training submodule selects historical production data from a preset resource pool for model training, and based on the performance requirement information, requires the model accuracy to reach more than 90%. Finally, the model output submodule outputs an analysis model that meets the preset performance conditions, which can be used for predictive analysis of production data.

[0032] The implementation principle and beneficial effects of this embodiment: The present invention selects a model to be trained that meets the system monitoring requirements through the collaboration of a model selection submodule, a model training submodule, and a model output submodule, and performs model training based on data resources in a preset resource pool, and outputs an analysis model that meets the preset performance conditions according to performance requirement information. The present invention can output an alternative model that meets the performance requirements by setting performance requirement information, thereby improving the prediction accuracy and stability of the model and helping decision makers better understand the data and make accurate decisions.

[0033] A coal mine production process control system provided by an embodiment of the present invention includes a model training submodule, including: A data selection unit, used to obtain a data selection instruction, and select corresponding preset data resources in a target resource pool based on the data selection instruction; A data partitioning unit, used to partition the preset data resources using a preset data partitioning method, and output a training data set, a verification data set, and a test data set; The model training unit is used to train and optimize the performance of the model to be trained based on the training data set, the validation data set, and the test data set, and output the candidate model.

[0034] In this embodiment, the data selection instruction: instructs the data selection unit to select a specific data resource from the target resource pool, and may be a selection instruction for a specific time period, specific parameters, etc.; In this embodiment, the target resource pool: stores various data resources required by the system, such as historical production data, sensor data, etc.; In this embodiment, the preset data partitioning method specifies how to partition the entire data set into a training data set, a validation data set, and a test data set, such as random partitioning, proportional partitioning, time partitioning, etc.; In this embodiment, training data set: a data set used for model training, through which the model learns parameters; In this embodiment, the validation data set is used to evaluate the performance of the model and adjust the hyperparameters during the training process to avoid overfitting of the model; In this embodiment, the test data set is used to finally evaluate the generalization ability and performance of the model. The model is tested on this part of the data to verify its performance on unseen data. For example, in the coal mine production management and control system, the data selection unit selects the production data of the past year from the target resource pool as the preset data resource according to the data selection instruction. The data partitioning unit divides the data into a training data set (80%), a verification data set (10%) and a test data set (10%) by time partitioning. The model training unit uses these data sets to train and optimize the performance of the model to be trained, and outputs an alternative model for subsequent analysis.

[0035] The implementation principle and beneficial effects of this embodiment: The present invention can select preset data resources in the target resource pool according to instructions, and divide the data into training, verification and test data sets according to a preset data division method, and perform model training and performance optimization based on these data sets. Through the collaboration of the data selection unit and the data division unit, the system can efficiently utilize the data resources in the target resource pool to avoid data waste. At the same time, the training data set, verification data set and test data set are used for model training and performance evaluation, which helps to optimize the performance and generalization ability of the model.

[0036] The embodiment of the present invention provides a coal mine production process control system, a data analysis module, including: The data analysis submodule is used to analyze the data to be analyzed and obtain the sub-data corresponding to each production process; The level information acquisition submodule is used to obtain the monitoring level information of each sub-data and summarize and output the process monitoring level table; The comparison table generation submodule is used to combine the process monitoring level table and the sub-data corresponding to each production process under each monitoring level to output the monitoring level-data comparison table; The abnormality identification submodule is used to extract features from the sub-data of the production process at each monitoring level in the monitoring level-data comparison table, and construct a second feature set, and perform feature matching on the data features in the second feature set in combination with a preset abnormality feature database, to obtain abnormality identification sub-results corresponding to the sub-data of the production process at each monitoring level, and summarize and output abnormality analysis results; A historical data acquisition submodule, used to select and obtain, from the historical database, first historical process data that matches the data numerical feature in the second feature set and second historical process data that matches the data temporal feature in the second feature set; The format conversion submodule is used to obtain the data format information of the sub-data, the first historical process data and the second historical process data corresponding to each monitoring level, and perform format conversion on the sub-data, the first historical process data and the second historical process data corresponding to each monitoring level in combination with the format requirement information of the analysis model to obtain the first standard data, the second standard data and the third standard data respectively; A real-time analysis submodule, configured to perform real-time analysis on the first standard data corresponding to the production process at each monitoring level based on the analysis model, and output the real-time analysis result; The prediction analysis submodule is used to perform prediction analysis on the second standard data and the third standard data based on the analysis model, obtain the prediction data of each production process under each monitoring level in the future preset time period, and output the prediction analysis results; The comprehensive analysis submodule is used to analyze the abnormal analysis results, real-time analysis results and predictive analysis results using a preset comprehensive analysis method, and output the comprehensive analysis results; A historical prediction data acquisition submodule is used to acquire historical prediction data matching the analysis model in the historical database; The result verification submodule is used to perform time calibration and data verification on the historical forecast data and real-time analysis results under each same production process, and output the forecast data verification results; The model optimization submodule is used to optimize the comprehensive analysis results and the prediction data verification results in combination with preset optimization indicators, and to adjust the parameters and optimize the performance of the analysis model based on the optimization analysis results.

[0037] In this embodiment, sub-data: sub-data corresponding to each production process obtained after parsing the data to be analyzed, that is, specific data for each production process extracted from the original data. For example, assuming that a certain production process is "coal mining", the corresponding sub-data is the mining depth data in the process; In this embodiment, monitoring level information: that is, the monitoring level of different production processes, is used to determine the corresponding monitoring level and data processing method In this embodiment, the process monitoring level table: summarizes the monitoring level information corresponding to each production process, clarifies the monitoring priority of each production process, and helps the system to monitor and analyze; In this embodiment, the monitoring level-data comparison table includes the monitoring level information and sub-data comparison table corresponding to each production process, which is convenient for subsequent data analysis and abnormality identification. In this embodiment, the second feature set is a feature set constructed by extracting features from the sub-data, and is used for abnormality identification and analysis. For example, assuming that a certain production process is "coal mining", the second feature set includes features such as mining speed and mining depth change rate; In this embodiment, the preset abnormal feature database includes predefined abnormal feature information, which is used to match the data features in the second feature set to identify abnormal situations; In this embodiment, the abnormality identification sub-result is the result obtained after feature extraction and abnormality identification of the sub-data of the production process at the monitoring level, for example, abnormal fluctuation of the mining depth exceeds a threshold, etc.; In this embodiment, the abnormal analysis result: the analysis result of the abnormal identification sub-result is integrated to guide subsequent decision-making and processing; In this embodiment, the first historical process data: historical process data matching the data numerical feature in the second feature set; In this embodiment, the second historical process data: historical process data matching the temporal features in the second feature set; In this embodiment, data format information refers to format information of various data, such as data type, data structure, etc.; In this embodiment, the format requirement information: that is, the requirement information of the data format of the data to be processed by the analysis model, to ensure the consistency and compatibility of the data format; In this embodiment, the first standard data: standard data matching the format requirement information obtained after format conversion of the sub-data, used for subsequent real-time analysis; In this embodiment, the second standard data: standard data matching the format requirement information obtained after format conversion of the first historical process data, used for subsequent prediction analysis; In this embodiment, the third standard data: standard data matching the format requirement information obtained after format conversion of the second historical process data, corresponding to the second standard data, and used for subsequent prediction analysis; In this embodiment, the real-time analysis result: the result obtained after real-time analysis of the first standard data, outputting the current status evaluation result of a certain production process, such as normal, abnormal, etc.; In this embodiment, the prediction data refers to data obtained by performing a prediction analysis on the second standard data and the third standard data through an analysis model, and is used to predict the production situation or data of each production process in a certain time period in the future, for example, predicting the production change in the next week; In this embodiment, the forecast analysis results: after analyzing the forecast data, the forecast status of a future production process is output, for example, the forecast of coal mine output, equipment wear, etc.; In this embodiment, the preset comprehensive analysis method is a method that is preset and used to comprehensively analyze the abnormal analysis results, real-time analysis results, and predictive analysis results, which can be a weighted average method that comprehensively considers abnormal situations, real-time data, and predictive data, or a comprehensive judgment method based on rules and models; In this embodiment, comprehensive analysis results: the final results obtained by the comprehensive analysis method, combined with the results of abnormal analysis, real-time analysis and predictive analysis, provide a comprehensive data analysis conclusion, for example, the comprehensive analysis results show that the production process under a certain monitoring level is abnormal, the real-time analysis results show that the data of the process is abnormal, and the predictive analysis results also show that there may be problems with the process in the future; In this embodiment, historical prediction data refers to historical prediction data that matches the analysis model and is stored in the history database. In other words, the historical prediction data stored in the history database by the analysis model is the prediction data obtained by the analysis model at a previous time point through prediction analysis, and is used for comparison and verification with the real-time analysis results corresponding to the same time node, so as to continuously reduce the prediction error of the analysis model and optimize the prediction performance of the analysis model. In this embodiment, time calibration and data verification: verify the accuracy of the prediction data by comparing the timestamps and data content of the historical prediction data and the real-time analysis results to ensure the accuracy and reliability of the prediction data; In this embodiment, the prediction data verification result: the verification conclusion drawn based on the time calibration and data verification results indicates the quality and accuracy of the prediction data. For example, the verification result shows that the prediction data is highly consistent with the actual data, and the prediction accuracy rate reaches 95%; In this embodiment, the preset optimization index is an index or standard used to optimize the comprehensive analysis results and the prediction data verification results, for example, accuracy, recall rate, error range and other indicators; In this embodiment, the optimization analysis results: the optimization conclusions obtained after analyzing the comprehensive analysis results and the prediction data verification results according to the optimization indicators, including parameter adjustment or performance optimization suggestions, etc. For example, the conclusion obtained according to the optimization indicators is that certain parameters of the model need to be adjusted to improve the prediction accuracy.

[0038] Implementation principles and beneficial effects of this embodiment: The present invention implements comprehensive monitoring and analysis of the production process by parsing data, acquiring level information, identifying anomalies, acquiring historical data, converting formats, performing real-time analysis, predictive analysis, comprehensive analysis, acquiring historical predictive data, verifying results, and optimizing models through the collaboration of various submodules. The present invention can comprehensively monitor and analyze abnormal conditions, real-time data, and future predictive data of the production process, and optimize the analysis according to optimization indicators, and adjust the model parameters and optimize the performance based on the optimization analysis results, thereby continuously improving the performance of the analysis model.

[0039] The embodiment of the present invention provides a coal mine production process control system, a model optimization submodule, including: An analysis unit, used to analyze the comprehensive analysis result to obtain a first sub-result corresponding to the abnormal analysis result, a second sub-result corresponding to the real-time analysis result, and a third sub-result corresponding to the predictive analysis result; A feature extraction unit, used to extract features from the first sub-result, the second sub-result, the third sub-result and the prediction data verification result, and construct an abnormal feature set, a real-time feature set, a prediction feature set and a verification feature set based on the extracted features; An indicator selection unit, used to select preset optimization indicators in the indicator database, and construct a first indicator set matching the abnormal feature set, a second indicator set matching the real-time feature set, a third indicator set matching the prediction feature set, and a fourth indicator set matching the verification feature set; A relationship graph acquisition unit, used to acquire a mapping relationship between preset optimization indicators in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set, and output an indicator relationship graph; An optimization analysis unit, configured to optimize and analyze the comprehensive analysis result and the prediction data verification result based on the first indicator set, the second indicator set, the third indicator set, the fourth indicator set and the indicator relationship diagram, to obtain an optimization analysis result; ; in, Indicates the optimization analysis results; represents the exponential function; Represents a comprehensive correction coefficient corresponding to the data calculation error under each preset optimization indicator in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set; represents the total number of preset optimization indicators in the i-th indicator set; i=1 corresponds to the first indicator set; i=2 corresponds to the second indicator set; i=3 corresponds to the third indicator set; i=4 corresponds to the fourth indicator set; represents the optimization standard parameter corresponding to the jth preset optimization indicator in the i-th indicator set; represents the actual parameter value corresponding to the jth preset optimization indicator in the ith indicator set in the comprehensive analysis results and the prediction data verification results; represents the optimization weight coefficient corresponding to the jth preset optimization indicator in the i-th indicator set; represents the complex correlation coefficient between the jth preset optimization indicator in the i-th indicator set and the remaining t-1 preset optimization indicators in each indicator set, and t represents the total number of preset optimization indicators in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set; The model optimization unit is used to combine the preset result-strategy-instruction comparison table to obtain the model parameter performance optimization strategy and the optimization instructions under the corresponding strategy that match the optimization analysis results, and adjust the parameters and optimize the performance of the analysis model based on the model parameter performance optimization strategy and the optimization instructions.

[0040] In this embodiment, the first sub-result: a sub-result corresponding to the abnormal analysis result parsed from the comprehensive analysis result, including detailed information about the abnormal situation in the production process. For example, assuming that the comprehensive analysis result shows that an abnormality occurs in a certain production link, the first sub-result may be information such as the specific abnormality type, abnormality duration, and impact range of the link; In this embodiment, the second sub-result: a sub-result corresponding to the real-time analysis result parsed from the comprehensive analysis result, providing detailed information about the current production status. For example, if the real-time analysis result shows that the operation status of a certain equipment is unstable, the second sub-result may include information such as equipment operation parameters, sensor data, and equipment operation time; In this embodiment, the third sub-result: a sub-result corresponding to the forecast analysis result parsed from the comprehensive analysis result provides forecast information of future production conditions. For example, assuming that the forecast analysis result shows that a certain production indicator will decline within the next week, the third sub-result may include the predicted decline range, possible reasons and recommended countermeasures; In this embodiment, abnormal feature set: a feature set extracted from the first sub-result, used to describe the characteristics of the abnormal situation, including but not limited to the abnormality type, abnormality occurrence time, abnormality impact range, abnormality duration and other characteristics; In this embodiment, real-time feature set: a feature set extracted from the second sub-result, used to describe the characteristics of the real-time production status, including but not limited to real-time information such as equipment operating parameters, sensor data, and environmental conditions; In this embodiment, the prediction feature set is a feature set extracted from the third sub-result, which is used to describe the characteristics of future production conditions, including but not limited to prediction indicators, prediction time periods, prediction trends, possible influencing factors, and other characteristics; In this embodiment, verification feature set: a feature set used to verify the accuracy of predicted data, used to compare the difference between predicted data and actual data, including but not limited to features such as verification time period, verification index, actual data, and predicted data; In this embodiment, the indicator database: a database storing preset optimization indicators, used to provide selectable indicators in the subsequent indicator selection process, including but not limited to various indicators that can be used for optimization analysis, such as production equipment operating parameters, environmental monitoring data, and production indicators; In this embodiment, the first indicator set: a set of preset optimization indicators corresponding to the abnormal feature set; In this embodiment, the second indicator set: a set of preset optimization indicators corresponding to the real-time feature set; In this embodiment, the third indicator set: a set of preset optimization indicators corresponding to the prediction feature set; In this embodiment, the fourth indicator set: a set of preset optimization indicators corresponding to the verification feature set; In this embodiment, the indicator relationship diagram: that is, a mapping relationship diagram between indicators, is used to help understand the correlation between the indicators; In this embodiment, the preset result-strategy-instruction comparison table: a comparison table containing the mapping relationship between the optimization analysis results and the model parameter performance optimization strategies and optimization instructions, which is used to match the input optimization analysis results to the corresponding model parameter performance optimization strategies and optimization instructions; In this embodiment, the model parameter performance optimization strategy: a strategy for adjusting model parameters and optimizing performance determined according to the optimization analysis results; In this embodiment, the optimization instruction is to adjust and update the specific parameters of the model according to the specific operation instructions provided by the model parameter performance optimization strategy.

[0041] The implementation principle and beneficial effects of this embodiment: The present invention can achieve parameter adjustment and performance optimization of the analysis model through the collaboration of various units such as the parsing unit, the feature extraction unit, the indicator selection unit, and the relationship diagram acquisition unit. The present invention can perform real-time parameter adjustment and performance optimization on the analysis model, thereby continuously improving the data analysis performance of the analysis model, ensuring the accuracy of the model analysis results, and thereby improving the efficiency and accuracy of the production process control system.

[0042] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A coal mine production process control system, characterized in that: include: Data acquisition module, used to obtain data of various production processes in the mine in real time and output initial data; A preprocessing module, used to preprocess the initial data using a preset preprocessing method to obtain data to be analyzed; A model training module is used to train the model to be trained selected from the model database in combination with preset data resources, output candidate models, and output the candidate models that meet preset performance conditions as analysis models; The data analysis module is used to analyze the data to be analyzed based on the analysis model, output the process analysis results, and update the parameters and optimize the performance of the analysis model based on the process analysis results and in combination with the preset data resources and cloud data.

2. A coal mine production process control system according to claim 1, characterized in that: The data acquisition module comprises: The process classification submodule is used to divide the production processes in the mine according to the monitoring priority of each production process and obtain a process classification table; The data acquisition submodule is used to obtain the data of each production process in the process classification table in real time by using the preset monitoring equipment set at the preset position, and output the initial data.

3. A coal mine production process control system according to claim 2, characterized in that: The data acquisition submodule comprises: A demand acquisition unit, used to acquire the monitoring requirements corresponding to each production process, and summarize and output the system monitoring requirements in combination with the monitoring priorities corresponding to each production process; An equipment selection unit is used to obtain preset monitoring equipment corresponding to each production process based on the system monitoring requirements and in combination with the preset requirements-equipment matching table, determine the preset location for installing each preset monitoring equipment, and output a process-equipment comparison table; The data acquisition unit is used to obtain the data of each production process in the mine in real time through the preset monitoring equipment at each preset position based on the process-equipment comparison table, and comprehensively output the initial data.

4. A coal mine production process control system according to claim 1, characterized in that: The preprocessing module comprises: A method selection submodule is used to extract features from the initial data, construct a first feature set based on the extracted features, and obtain a preset preprocessing method matching the first feature set in combination with a preset feature-method mapping table; The preprocessing submodule is used to preprocess the initial data based on the preset preprocessing method to obtain first data, and output the first data that meets the first threshold condition as data to be analyzed.

5. A coal mine production process control system according to claim 3, characterized in that: The model training module includes: A model selection submodule is used to select a model to be trained that matches the monitoring requirements of the system in the model database in combination with a preset requirement-model matching table; A model training submodule, used to select preset data resources from a preset resource pool, train the to-be-trained model based on the preset data resources, and output a candidate model; The model output submodule is used to obtain the performance requirement information of the model, output the preset performance conditions based on the performance requirement information, and output the candidate model that meets the preset performance judgment conditions as the analysis model.

6. A coal mine production process control system according to claim 5, characterized in that: The model training submodule includes: A data selection unit, used to obtain a data selection instruction, and select a corresponding preset data resource in a target resource pool based on the data selection instruction; A data partitioning unit, used to partition the preset data resources using a preset data partitioning method, and output a training data set, a verification data set, and a test data set; The model training unit is used to train and optimize the performance of the model to be trained based on the training data set, the verification data set and the test data set, and output an alternative model.

7. A coal mine production process control system according to claim 1, characterized in that: The data analysis module comprises: A data analysis submodule, used to analyze the data to be analyzed to obtain sub-data corresponding to each production process; The level information acquisition submodule is used to obtain the monitoring level information of each sub-data and summarize and output the process monitoring level table; A comparison table generation submodule, for combining the process monitoring level table and the sub-data corresponding to each production process under each monitoring level, and outputting a monitoring level-data comparison table; The abnormality identification submodule is used to extract features from the sub-data of the production process at each monitoring level in the monitoring level-data comparison table, and construct a second feature set, and perform feature matching on the data features in the second feature set in combination with a preset abnormality feature database to obtain abnormality identification sub-results corresponding to the sub-data of the production process at each monitoring level, and summarize and output abnormality analysis results; A historical data acquisition submodule, used to select from a historical database first historical process data matching the data numerical feature in the second feature set and second historical process data matching the data temporal feature in the second feature set; a format conversion submodule, for obtaining data format information of the sub-data, the first historical process data, and the second historical process data corresponding to each monitoring level, and performing format conversion on the sub-data, the first historical process data, and the second historical process data corresponding to each monitoring level in combination with the format requirement information of the analysis model, to obtain first standard data, second standard data, and third standard data, respectively; A real-time analysis submodule, configured to perform real-time analysis on the first standard data corresponding to the production process at each monitoring level based on the analysis model, and output a real-time analysis result; A prediction analysis submodule, used to perform prediction analysis on the second standard data and the third standard data based on the analysis model, obtain prediction data of each production process at each monitoring level within a future preset time period, and output the prediction analysis results; A comprehensive analysis submodule is used to analyze the abnormal analysis results, real-time analysis results and predictive analysis results using a preset comprehensive analysis method, and output the comprehensive analysis results; A historical prediction data acquisition submodule is used to acquire historical prediction data matching the analysis model in a historical database; A result verification submodule is used to perform time calibration and data verification on the historical prediction data and the real-time analysis results under each same production process, and output the prediction data verification results; The model optimization submodule is used to optimize the comprehensive analysis results and the prediction data verification results in combination with preset optimization indicators, and adjust the parameters and optimize the performance of the analysis model based on the optimization analysis results.

8. A coal mine production process control system according to claim 7, characterized in that: The model optimization submodule includes: An analysis unit, configured to analyze the comprehensive analysis result to obtain a first sub-result corresponding to the abnormal analysis result, a second sub-result corresponding to the real-time analysis result, and a third sub-result corresponding to the predictive analysis result; A feature extraction unit, used to extract features from the first sub-result, the second sub-result, the third sub-result and the prediction data verification result, and construct an abnormal feature set, a real-time feature set, a prediction feature set and a verification feature set based on the extracted features; An indicator selection unit, used to select preset optimization indicators in an indicator database, and construct a first indicator set matching the abnormal feature set, a second indicator set matching the real-time feature set, a third indicator set matching the prediction feature set, and a fourth indicator set matching the verification feature set; A relationship graph acquisition unit, used to acquire a mapping relationship between preset optimization indicators in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set, and output an indicator relationship graph; An optimization analysis unit, configured to perform optimization analysis on the comprehensive analysis result and the prediction data verification result based on the first indicator set, the second indicator set, the third indicator set, the fourth indicator set and the indicator relationship diagram to obtain an optimization analysis result; ； in, Indicates the optimization analysis results; represents the exponential function; Represents a comprehensive correction coefficient corresponding to the data calculation error under each preset optimization indicator in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set; represents the total number of preset optimization indicators in the i-th indicator set; i=1 corresponds to the first indicator set; i=2 corresponds to the second indicator set; i=3 corresponds to the third indicator set; i=4 corresponds to the fourth indicator set; represents the optimization standard parameter corresponding to the jth preset optimization indicator in the i-th indicator set; represents the actual parameter value corresponding to the jth preset optimization indicator in the ith indicator set in the comprehensive analysis results and the prediction data verification results; represents the optimization weight coefficient corresponding to the jth preset optimization indicator in the i-th indicator set; represents the complex correlation coefficient between the jth preset optimization indicator in the i-th indicator set and the remaining t-1 preset optimization indicators in each indicator set, and t represents the total number of preset optimization indicators in the first indicator set, the second indicator set, the third indicator set, and the fourth indicator set; The model optimization unit is used to combine the preset result-strategy-instruction comparison table to obtain the model parameter performance optimization strategy and the optimization instructions under the corresponding strategy that match the optimization analysis results, and adjust the parameters and optimize the performance of the analysis model based on the model parameter performance optimization strategy and the optimization instructions.

Images