Coal preparation plant production data full-process management system and method

Abstract

The invention provides a coal preparation plant production data full-process management system and method, and the method comprises the steps: a data collection subsystem collects equipment data and production data of different data sources of each plant through a preset data collection interface; the data transmission subsystem encrypts and transmits equipment data and production data acquired by the data acquisition subsystem to the data cleaning subsystem for cleaning, and the data storage subsystem stores the data cleaned by the data cleaning subsystem to a plurality of databases of different types in a mixed storage mode. And the data application subsystem obtains the target data from the data storage subsystem and performs production plan optimization, equipment fault early warning and quality control and optimization based on the target data and preset data. By the adoption of the method, efficient processing and deep application of the whole process of production data of the coal preparation plant are achieved, powerful support is provided for intelligent and refined production management of the coal preparation plant, and the overall operation level and economic benefits of the coal preparation plant can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a full-process management system and method for production data of a coal preparation plant. Background Art

[0002] In the modern coal preparation industry, with the widespread application of automation equipment and information technology, the production process generates a massive amount of data. These data cover all aspects from raw coal input to clean coal output, including equipment operating status, process parameters, coal quality indicators and other information. For coal preparation plants, how to make full use of these data to optimize production processes, improve product quality, reduce operating costs and ensure stable equipment operation has become a key factor in improving the competitiveness of coal preparation plants. At present, some more advanced coal preparation plants have begun to try to apply production data to a certain extent, but the existing solutions still have many limitations.

[0003] In terms of data collection and integration, although some coal preparation plants have adopted a variety of data collection methods, these data collection methods are often isolated, and there is a lack of unified standards for data interfaces and communication protocols between different systems and different equipment, resulting in low data collection efficiency, difficulty in ensuring data accuracy and integrity, and a lack of effective security protection mechanisms during data transmission.

[0004] In terms of data storage and management, most coal preparation plants simply store the collected data in traditional databases. This storage method does not fully consider the characteristics of coal preparation data (such as time series, massive data, and complex correlations between different data). This makes data query and retrieval inefficient, making it difficult to meet the needs of rapid analysis and mining of large amounts of historical data, and it is also unable to effectively support real-time processing and analysis applications of data.

[0005] In terms of data application, the existing implementation solutions are mainly in the following forms: (1) Production decision support: Production managers usually rely on manual statistics and analysis of some key data, and then formulate production plans, adjust production process parameters, and arrange equipment maintenance and overhaul time based on these limited data and personal experience. This decision-making method is highly subjective and lacks comprehensive and in-depth data analysis of the entire production process, making it difficult to achieve optimal allocation of production resources and maximize production efficiency.

[0006] (2) Equipment maintenance management: Equipment maintenance mainly relies on a regular inspection system, that is, to conduct routine maintenance work such as appearance inspection, performance testing, and replacement of wearing parts at fixed time intervals (such as once a week or once a month). When equipment fails, maintenance personnel diagnose and repair the fault based on the fault phenomenon and personal experience. There is a lack of real-time monitoring and analysis of equipment operation data, and it is impossible to predict the occurrence of equipment failure in advance, resulting in long downtime due to sudden equipment failure, which seriously affects production progress and increases equipment maintenance costs and the workload of maintenance personnel.

[0007] (3) Quality control measures: At present, the main method of quality control is manual sampling and testing, that is, coal samples are taken at specific stages of the production process and sent to the laboratory for testing and analysis, and then the test results are used to determine whether the product quality meets the standard requirements. If quality problems are found, they are corrected by manually adjusting the production process parameters. This quality control method has obvious lags and cannot monitor changes in product quality in real time. In addition, the accuracy and timeliness of manual adjustment of process parameters are difficult to guarantee, which can easily lead to large fluctuations in product quality and affect the company's economic benefits and market reputation. Summary of the invention

[0008] In view of this, an object of the present invention is to provide a full-process management system and method for production data of a coal preparation plant to alleviate the above-mentioned problems existing in the related technology.

[0009] In the first aspect, an embodiment of the present invention provides a full-process management system for production data of a coal preparation plant, including: a data acquisition subsystem, a data transmission subsystem, a data cleaning subsystem, a data storage subsystem and a data application subsystem; the data acquisition subsystem is pre-set with a data acquisition interface for docking with the data transmission subsystem; the data acquisition subsystem is used to collect equipment data and production data from different data sources in each plant through the data acquisition interface; the data transmission subsystem is used to encrypt and transmit the equipment data and production data collected by the data acquisition subsystem to the data cleaning subsystem; the data cleaning subsystem is used to clean the received equipment data and production data; wherein the cleaning includes outlier detection, outlier processing, missing value detection and missing value processing; the data storage subsystem is used to use a hybrid storage method to store the data cleaned by the data cleaning subsystem to multiple different types of databases; the data application subsystem is used to obtain target data from the data storage subsystem, and perform production plan optimization, equipment failure warning and quality control and optimization based on the target data and preset data.

[0010] In a second aspect, an embodiment of the present invention further provides a method for managing the entire process of production data of a coal preparation plant, which is applied to the entire process management system of production data of a coal preparation plant described in the first aspect above, and includes: the data acquisition subsystem collects equipment data and production data from different data sources in each plant through the data acquisition interface; the data transmission subsystem encrypts and transmits the equipment data and production data collected by the data acquisition subsystem to the data cleaning subsystem; the data cleaning subsystem cleans the received equipment data and production data; wherein the cleaning includes outlier detection, outlier processing, missing value detection and missing value processing; the data storage subsystem uses a hybrid storage method to store the data cleaned by the data cleaning subsystem in multiple different types of databases; the data application subsystem obtains target data from the data storage subsystem, and performs production plan optimization, equipment failure warning, and quality control and optimization based on the target data and preset data.

[0011] The embodiment of the present invention provides a full-process management system and method for production data of a coal preparation plant. It uses a unified data acquisition interface to collect multi-source data, which improves the data acquisition efficiency compared with the existing technology; detects and processes abnormal values ​​and missing values ​​in the data cleaning link, which improves the accuracy and effectiveness of data cleaning compared with the existing technology, and provides reliable data guarantee for data analysis; uses a hybrid storage method to reasonably store, allocate and manage data, which improves the efficiency of data storage and management compared with the existing technology, and provides efficient data support for data application; can integrate multi-source data for production plan optimization, equipment failure warning and quality control and optimization, so that production decisions become more scientific, accurate and efficient, reduce the cost of equipment maintenance management, and effectively improve the stability and consistency of product quality. The above technology is adopted to realize the efficient processing and in-depth application of the whole process of production data of the coal preparation plant, provide strong support for the intelligent and refined production management of the coal preparation plant, and can effectively improve the overall operation level and economic benefits of the coal preparation plant.

[0012] 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 are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0013] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 This is a schematic diagram of the structure of a full-process management system for production data of a coal preparation plant according to an embodiment of the present invention; Figure 2 This is a principle block diagram of the overall architecture of a full-process management system for production data of a coal preparation plant in an embodiment of the present invention; Figure 3 This is a block diagram of the working principle of the data acquisition interface in an embodiment of the present invention; Figure 4 This is a schematic diagram of a distributed collection architecture in an embodiment of the present invention; Figure 5 This is a timing diagram of data transmission in an embodiment of the present invention; Figure 6 is a flow chart of outlier detection based on statistical analysis in an embodiment of the present invention; Figure 7 This is a data cleaning flow chart in an embodiment of the present invention; Figure 8 A schematic diagram of a data storage architecture in an embodiment of the present invention; Fig. 9 This is a workflow diagram of a production decision support module in an embodiment of the present invention; Fig.10 This is a block diagram of the working principle of the equipment maintenance warning module in an embodiment of the present invention; Fig.11 This is a workflow diagram of the quality control and optimization module in an embodiment of the present invention; Fig.12 The figure is a flow chart of a method for managing the entire process of production data of a coal preparation plant in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described in combination with the embodiments below. 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.

[0017] To facilitate understanding of this embodiment, firstly, a coal preparation plant production data full process management system disclosed in an embodiment of the present invention is described in detail, see Figure 1 As shown, the coal preparation plant production data full process management system may include a data acquisition subsystem 100, a data transmission subsystem 200, a data cleaning subsystem 300, a data storage subsystem 400 and a data application subsystem 500; the data acquisition subsystem 100 is pre-set with a data acquisition interface 101 for docking with the data transmission subsystem 200; The data acquisition subsystem 100 can be used to collect equipment data and production data from different data sources in each plant through the data acquisition interface 101; The data transmission subsystem 200 can be used to encrypt and transmit the equipment data and production data collected by the data collection subsystem 100 to the data cleaning subsystem 300; The data cleaning subsystem 300 may be used to clean the received equipment data and production data; wherein the cleaning may include outlier detection, outlier processing, missing value detection and missing value processing; The data storage subsystem 400 can be used to store the data cleaned by the data cleaning subsystem 300 in multiple different types of databases using a hybrid storage method; The data application subsystem 500 can be used to obtain target data from the data storage subsystem 400, and perform production plan optimization, equipment failure warning, and quality control and optimization based on the target data and preset data.

[0018] The embodiment of the present invention provides a full-process management system for production data of a coal preparation plant. It uses a unified data acquisition interface to collect multi-source data, which improves the data acquisition efficiency compared with the existing technology; detects and processes abnormal values ​​and missing values ​​in the data cleaning link, which improves the accuracy and effectiveness of data cleaning compared with the existing technology, and provides reliable data guarantee for data analysis; uses a hybrid storage method to reasonably store, allocate and manage data, which improves the efficiency of data storage and management compared with the existing technology, and provides efficient data support for data application; can integrate multi-source data for production plan optimization, equipment failure warning and quality control and optimization, so that production decisions become more scientific, accurate and efficient, reduce the cost of equipment maintenance management, and effectively improve the stability and consistency of product quality. The above-mentioned full-process management system for production data of a coal preparation plant realizes the efficient processing and in-depth application of the whole process of production data of the coal preparation plant, provides strong support for the intelligent and refined production management of the coal preparation plant, and can effectively improve the overall operation level and economic benefits of the coal preparation plant.

[0019] In actual application, see Figure 1 and Figure 2As shown, the overall architecture of the full-process management system of coal preparation plant production data mainly includes data acquisition layer (i.e. data acquisition subsystem 100), data transmission layer (i.e. data transmission subsystem 200), data cleaning layer (i.e. data cleaning subsystem 300), data storage layer (i.e. data storage subsystem 400) and data application layer (i.e. data application subsystem 500). In this architecture, the functions of each layer are as follows: Data collection layer (DC): responsible for collecting production data from various equipment and data sources of the coal preparation plant through a variety of data collection methods and data collection interfaces to ensure the comprehensiveness and accuracy of data collection.

[0020] Data transmission layer (DT): uses reliable communication protocols and network technologies to safely and quickly transmit the collected data to the data cleaning layer to ensure the stability and integrity of data transmission.

[0021] Data cleaning layer (DCL): Cleans and preprocesses the transmitted data, removes outliers, repairs missing values, improves data quality, and provides a reliable data foundation for subsequent data storage and data applications.

[0022] Data storage layer (DSL): Based on the type and characteristics of the data, a hybrid storage method is used to store the cleaned data in a suitable database to facilitate data query, analysis, and management.

[0023] Data application layer (DAL): Based on the stored data, various data application modules are used to provide targeted services and support for the production decision-making, equipment maintenance, quality control, etc. of the coal preparation plant to maximize the value of data.

[0024] The data collection layer can use various types of sensors and data collection equipment, which are connected to the equipment in various production links of the coal preparation plant. Sensors and data collection equipment include but are not limited to the following: Belt scale: used to measure the transportation volume of raw coal and clean coal on the belt. It uses high-precision weighing sensors and speed sensors, and is connected to the data acquisition terminal through the RS485 interface to collect material flow data in real time and transmit the flow data to the data acquisition terminal.

[0025] Equipment operation status sensors: such as temperature sensors (PT100), pressure sensors, vibration sensors, etc., are installed on key equipment such as crushers, centrifuges, and jigs, and transmit the equipment's operation status data to the data acquisition terminal through the analog input interface.

[0026] Coal quality analyzer: Analyzes the quality indicators of raw coal and clean coal, such as ash analyzer, moisture analyzer, etc., and sends the coal quality data to the data acquisition terminal through the Ethernet interface.

[0027] In actual application, see Figures 1 to 3 As shown, in order to be compatible with different types of data sources, the data acquisition layer designs a unified data acquisition interface 101. The data acquisition interface 101 may include a protocol identification module, a driver adaptation module and a data cache module; the protocol identification module may be used to identify the communication protocol of each data source; the driver adaptation module may be used to call the driver program corresponding to the communication protocol of each data source to perform data acquisition operations; the data cache module may be used to temporarily store the data collected by the driver adaptation module.

[0028] See also Figures 1 to 3 As shown, the working principle of the data acquisition interface 101 is as follows: Protocol Recognition Module (PRM): When a data source is connected, this module first automatically identifies the communication protocol of the data source, such as whether it is Modbus, OPC-UA or other custom protocols, to provide a basis for subsequent driver adaptation.

[0029] Driver Adaptation Module (DAM): Based on the result of the protocol identification module, select the corresponding driver to establish a connection with the data source and read the data in a specific protocol format. For example, for a Modbus protocol data source, call the Modbus driver to perform data acquisition operations.

[0030] Data Cache Module (DCM): The collected data is first stored in the data cache module for temporary storage and preliminary sorting, so that it can be subsequently transferred to the data transmission layer in batches, which is conducive to improving the efficiency and stability of data transmission.

[0031] In actual application, see Figures 1 to 4 As shown in the figure, in order to improve the efficiency and quality of data collection, the data collection layer can adopt a distributed collection architecture as a data collection optimization strategy: the coal preparation plant can be divided into multiple areas (areas can be divided according to the plant, and each plant is regarded as a corresponding area), and a data collection node is set in each area. Each data collection node is independently responsible for the collection of equipment data and production data in the area, avoiding the performance bottleneck and data loss risk caused by single-point data collection. Data collection nodes are connected through industrial Ethernet to realize data aggregation and transmission.

[0032] See also Figures 1 to 4 As shown, the data acquisition subsystem 100 may include a data acquisition node pre-set for each factory building; each data acquisition node can be used to collect equipment data and production data from different data sources of the corresponding factory building through the data acquisition interface 101, and send the collected equipment data and production data to the data transmission subsystem 200.

[0033] As a possible implementation, see Figure 1 and Figure 2 As shown, the device data may include the operating status and data changes of the corresponding device, and each device is pre-set with an importance level and a data change threshold; based on this, the data acquisition interface 101 can be used to adjust the frequency of the data acquisition operation corresponding to each device based on the operating status, data changes, importance level and data change threshold of each device.

[0034] In actual application, see Figure 2 As shown, in order to improve the efficiency and quality of data collection, the data collection layer can dynamically adjust the collection cycle: according to the importance of the equipment and the data update frequency, the collection cycle of each data collection task is dynamically adjusted, and the corresponding collection cycle is set for each device separately. For the data of key equipment (such as the main selection equipment, etc.), the collection cycle is shortened to seconds or even milliseconds to ensure the real-time nature of the data; for the data of auxiliary equipment (such as ventilation equipment, etc.), the collection cycle is appropriately extended to reduce the occupation of system resources. The specific adjustment strategy is achieved by presetting the importance level of the equipment and the data update frequency threshold; the data collection layer monitors the operating status and data changes of the equipment in real time. When the data update frequency of the equipment exceeds the corresponding data update frequency threshold or the equipment is in its critical operating state, the equipment collection cycle can be automatically shortened (that is, the equipment collection cycle is set to be less than the standard collection cycle). On the contrary (that is, when the data update frequency of the equipment does not exceed the corresponding data update frequency threshold and the equipment is not in its critical operating state), the equipment collection cycle is appropriately extended (that is, the equipment collection cycle is set to be greater than the standard collection cycle).

[0035] As a possible implementation, see Figure 1 and Figure 2 As shown, the data transmission subsystem 200 can also be used to encrypt the equipment data and production data collected by each data acquisition node using the SSL / TLS protocol, and use the TCP / IP protocol to transmit the encrypted data corresponding to each data acquisition node to the data cleaning subsystem 300; the data cleaning subsystem 300 can also be used to decrypt the received encrypted data.

[0036] In actual application, see Figure 1 and Figure 2 As shown in the figure, the data transmission layer can use a combination of industrial Ethernet and wireless communication technology to transmit the data collected by the data collection layer to the data cleaning layer. During the data transmission process, the data transmission layer uses the following technical means to ensure the security and stability of data transmission: Data encryption transmission: SSL / TLS protocol can be used to encrypt the transmitted data to prevent the data from being stolen and tampered with during the transmission process. A secure encryption channel is established between the data collection nodes of the data collection layer and the data cleaning servers of the data cleaning layer to ensure the confidentiality and integrity of the data during the data transmission process.

[0037] Reliable transmission protocol: TCP / IP protocol is used for data transmission, and reliable data transmission is guaranteed through three-way handshake and timeout retransmission mechanism. When the receiver finds that the data is lost or damaged, it automatically sends a retransmission request to the sender to ensure the integrity and accuracy of the data. Figure 5 The data transmission timing between the data collection node and the data cleaning server is shown. Figure 5 In the process, the data collection node first initiates a connection request (SYN) to the data cleaning server, the data cleaning server sends a confirmation connection request (SYN + ACK) to the data collection node, and then the data collection node sends a signal (ACK) to confirm the connection to the data cleaning server. After the connection is established, the data collection node sends the encrypted data to the data cleaning server. After receiving the data, the data cleaning server sends a signal (ACK) to confirm the receipt of the data to the data collection node. Figure 5 The handshake phase (with three-way handshake) and the data transmission phase are shown in FIG. 1 , and a timeout retransmission mechanism is adopted in both the handshake phase and the data transmission phase.

[0038] As a possible implementation, see Figure 1 and Figure 2 As shown, the data cleaning subsystem 300 can pre-establish multiple different types of outlier detection models; based on this, the data cleaning subsystem 300 can also be used to: use multiple outlier detection models to perform outlier detection on the received equipment data and production data; perform missing value detection on the received equipment data and production data; interpolate continuous data with detected missing values, and perform completion processing based on data correlation on data with detected missing values ​​and correlated with other data.

[0039] In actual application, see Figure 2 and Figure 7 As shown in the figure, the working process of the data cleaning layer mainly includes outlier detection and processing as well as missing value detection and processing.

[0040] For outlier detection, the data cleaning layer has established a variety of outlier detection models based on the characteristics of coal preparation plant data. There are two main ways to implement outlier detection using a variety of outlier detection models: outlier detection based on statistical analysis and outlier detection based on machine learning.

[0041] Outlier detection based on statistical analysis: For some data (such as the flow data of belt scales, etc.), obtain historical data and calculate the mean and standard deviation of historical data, set a reasonable threshold range (such as mean ± 3 times standard deviation, etc.), read real-time data and determine whether the data points in the real-time data exceed the set threshold range, mark the data points that exceed the threshold range as outliers and determine that the data that does not exceed the threshold range is normal; for other data (such as the temperature data of the equipment, etc.), it is necessary to set a reasonable threshold range based on the statistical characteristics of the historical data (i.e., mean and standard deviation) combined with the normal working data range of the equipment to determine whether the data is abnormal. For example, for the temperature data of the equipment, it is necessary to set the temperature threshold range in combination with the season, historical temperature data and the normal working temperature range of the equipment, so that when the real-time temperature data is obtained, the data points that exceed the temperature threshold range can be marked as outliers, that is, how much lower or higher the temperature is than the normal working temperature range of the equipment before it is marked as an outlier, which needs to be judged in combination with the historical temperature of a specific season.

[0042] Based on this, see Figure 6 As shown in the figure, the outlier detection process based on statistical analysis mainly includes: obtaining historical data, calculating the mean and standard deviation, setting the threshold range, reading real-time data, and judging whether the data point exceeds the set threshold range. If so (that is, the data point exceeds the threshold range), it is marked as an outlier; if not (that is, the data point does not exceed the threshold range), the data is judged to be normal.

[0043] Outlier detection based on machine learning: Use machine learning algorithms such as support vector machines (SVM) to model the operating parameters of the equipment. By learning a large amount of normal operating data of the equipment, a normal operating mode model of the equipment is constructed. When new data is input into the model, the model determines the degree of deviation between the data and the normal operating mode data. If the deviation exceeds the set threshold, the data is marked as an outlier.

[0044] For outlier processing, common data processing methods such as deleting outliers can be used, and the outlier processing method is not limited here.

[0045] For missing value detection, missing data points in the data are identified as missing values.

[0046] For missing value processing, the detected missing values ​​can be processed in the following two ways: Interpolation method: For continuous data (such as flow data of belt scales, etc.), in the time series, if a data point is missing, the size of the missing value can be calculated based on the two adjacent non-missing data points before and after the data point through interpolation algorithms (such as linear interpolation or other interpolation algorithms). For example, in the case of using linear interpolation for processing, assuming that the two adjacent data points before and after a missing data point are ( t 1, y 1) and ( t 2, y 2), the time of the missing data point is t , then the missing value y The calculation formula is: .

[0047] Completion method based on data correlation: For some data that are known to have strong correlation with other data, data correlation can be used to establish a multiple regression model (such as a multiple linear regression model or a multiple nonlinear regression model) so that the size of the missing value can be predicted by the multiple regression model using the available relevant data. For example, there is a certain correlation between the ash content of clean coal and the ash content, moisture content and other data of raw coal. A multiple linear regression model can be established based on data correlation. When the ash content data of clean coal is missing, the missing clean coal ash content data can be predicted by the established multiple linear regression model using the relevant data of raw coal and other available production data.

[0048] See also Figure 7 As shown in the figure, the overall workflow of the data cleaning layer mainly includes: first, the data cleaning layer receives the original data from the data transmission layer, and then performs outlier detection, outlier processing, data missing check, and missing value processing in sequence, and finally outputs the cleaned data to the data storage layer.

[0049] As a possible implementation, see Figure 1 and Figure 8 As shown, the data storage subsystem 400 may include a relational database, a time series database and an in-memory database; based on this, the data storage subsystem 400 may also be used to: store the structured data cleaned by the data cleaning subsystem 300 into the relational database; store the time series data cleaned by the data cleaning subsystem 300 into the time series database; store the real-time data cleaned by the data cleaning subsystem 300 into the in-memory database.

[0050] In actual application, see Figure 2 and Figure 8As shown, the data storage layer can adopt a hybrid storage architecture, including the relational database MySQL, the time series database TDengine and the memory database Redis. The hybrid storage architecture can realize the reasonable storage of data according to the type and characteristics of the data.

[0051] Relational database (MySQL): used to store the business data of the coal preparation plant (such as equipment information, personnel information, production plans, sales orders and other structured data), to facilitate complex queries and transaction processing. MySQL achieves efficient data storage and management by designing a reasonable database table structure and establishing relationships between tables.

[0052] Time series database (TDengine): mainly used to store a large amount of historical production data (such as historical records of equipment operating parameters, historical data of coal quality indicators, and other data with time series characteristics). TDengine uses its own special storage structure and indexing method to quickly insert, query, and analyze time series data to meet the needs of retrospective analysis and trend prediction of historical data.

[0053] In-memory database (Redis): used to store data with high real-time requirements (such as the current operating status of the device, real-time alarm information, etc.). Redis stores data in memory, has extremely high read and write speeds, and can quickly respond to real-time data requests from the data application layer, ensuring the timeliness and accuracy of data reading and writing.

[0054] exist Figure 8 The data storage architecture shown in this architecture, the cleaned data output by the data cleaning layer flows into the corresponding database for storage according to its type and characteristics, providing efficient data support for the data application layer.

[0055] As a possible implementation, see Figure 1 and Figure 2 As shown, the data application subsystem may include a production plan optimization module, an equipment failure warning module, and a quality control and optimization module.

[0056] See also Figure 1 and Figure 2As shown, the production plan optimization module can be used to: obtain first target data from the data storage subsystem 400, and establish a production optimization model based on the first target data, and then iteratively optimize the production plan based on the production optimization model; wherein, each time the production plan is optimized, the production optimization model is used to generate the current production plan, and the economic benefits and product quality of the current production plan are evaluated based on preset economic benefit standards and preset product quality standards, and then the parameters of the production optimization model are adjusted according to the economic benefits and product quality of the current production plan, so that the production optimization model is used to generate a production plan corresponding to the next optimized production plan when the production plan is optimized next time.

[0057] See also Figure 1 and Figure 2 As shown, the equipment fault warning module can be used to: obtain the second target data from the data storage subsystem 400, and perform feature extraction on the second target data to obtain corresponding feature data, and then perform fault prediction on the feature data through a pre-trained fault prediction model to generate fault warning information, and send the fault warning information to the terminal device corresponding to the equipment maintenance personnel.

[0058] See also Figure 1 and Figure 2 As shown, the quality control and optimization module can be used to: obtain the third target data from the data storage subsystem 400, and compare the third target data with the preset product quality standard to determine whether the third target data represents abnormal product quality based on the comparison result; if the third target data represents abnormal product quality, use the preset data analysis model to use the third target data to analyze the key factors affecting product quality, and generate a quality control strategy based on the key factors; adjust the process parameters based on the preset feedback control algorithm and quality control strategy.

[0059] See also Figure 1 and Figure 2 As shown, the quality control and optimization module can also be used to: generate first information based on key factors, and send the first information to the equipment failure warning module, so that the equipment failure warning module uses the first information as the fourth target data; adjust the quality control strategy based on the data generated by the production plan optimization module during the iterative optimization of the production plan.

[0060] In actual application, see Figure 1 and Figure 2 As shown, the data application layer may include a production decision support module (i.e., a production plan optimization module), an equipment maintenance warning module (i.e., an equipment failure warning module), and a quality control and optimization module.

[0061] The production decision support module integrates multi-source data in the data storage layer and uses optimization algorithms (such as gradient descent, Newton's method, conjugate gradient method, genetic algorithm, particle swarm optimization algorithm, etc.) and data analysis models (such as linear programming model, integer programming model, machine learning model, etc.) to provide decision support for production managers. Fig. 9 As shown in the figure, the workflow of the production decision support module is as follows: First, the module obtains the property data of raw coal (such as ash content, moisture, sulfur content, calorific value, etc.), the operating status data of the equipment (such as the load rate and availability of each equipment), and the market demand data (such as the quality standard and price trend of clean coal, etc.) from the data storage layer. Then, the module establishes a production optimization model based on the data obtained from the data storage layer, and iteratively optimizes the production plan based on the established production optimization model, such as using a linear programming model to determine the optimal production plan (including the raw coal blending plan and the process parameter adjustment strategy of each production link, etc.) to maximize economic benefits and meet product quality standards; specifically, each time the module optimizes the production plan, it uses the production optimization model to generate the production plan and evaluates the economic benefits and product quality of the production plan, and then adjusts the parameters of the production optimization model according to the evaluation results. Then, the module evaluates the economic benefits and product quality of the production plan, and presents the evaluation results to the production management personnel in an intuitive form (such as charts, reports, etc.). The management personnel confirm the production plan according to the actual situation. If the production plan is confirmed to be passed, the production plan will be issued to the production site for execution. If it is not confirmed to be passed, the optimization calculation of the production plan will be re-performed until a satisfactory production plan is generated.

[0062] The equipment maintenance early warning module collects and analyzes equipment operation data in real time, and predicts equipment failures in advance by building a fault prediction model. Fig.10 As shown, the equipment maintenance warning module can mainly include a data acquisition interface, a feature extraction module, a fault prediction model and a warning notification module. The working principle of the equipment maintenance warning module is as follows: Data Communication Interface (DCI): Obtain the real-time operating data of the equipment (including temperature, pressure, vibration, current and other parameters) from the data storage layer.

[0063] Feature extraction module (FEM): extracts features from the acquired data (for example, calculating statistical features such as the mean, variance, and slope of the data, and extracting the frequency characteristics of the signal through methods such as Fourier transform), and uses the extracted feature data as the input of the fault prediction model.

[0064] Fault prediction model (also known as FPM): A fault prediction model is constructed using machine learning models (such as neural network models, decision tree models, etc.). By learning a large amount of historical equipment operation data and fault cases, the fault prediction model can identify the normal operation mode and fault characteristic mode of the equipment. When the input characteristic data matches the fault characteristic mode, the fault prediction model predicts the possible fault type, fault time, and fault location of the equipment as fault warning information and outputs the fault warning information.

[0065] Early warning notification module (ANM): Once the fault prediction model issues a fault warning message, the early warning notification module immediately sends the fault warning message to the equipment maintenance personnel through SMS, email, system pop-up window, etc., and provides detailed fault diagnosis reports and maintenance suggestions so that the maintenance personnel can take timely measures for preventive maintenance.

[0066] The quality control and optimization module uses real-time online monitoring data (such as ash content, moisture, particle size and other indicators of clean coal) and preset data analysis models (such as multivariate linear regression model, principal component analysis model, etc.) to achieve closed-loop control and continuous optimization of product quality. Fig.11 As shown in the figure, the workflow of the quality control and optimization module is as follows: First, the quality data of clean coal (such as ash content, moisture, etc.) are collected in real time through sensors installed on the production line to realize real-time online monitoring of product quality indicators, and the collected data are compared and analyzed with the preset quality standards to determine whether the quality indicators are abnormal. If the quality indicators are found to be abnormal, the quality analysis program is immediately started. The quality analysis program combines other relevant data in the production process (such as the properties of raw coal, the process parameters of the equipment, the operation records of the operators, etc.) and uses data analysis models (such as multivariate linear regression, principal component analysis, etc.) to analyze to find out the key factors affecting product quality. Then, according to the analysis results, the feedback control algorithm (such as PID control, fuzzy logic control, neural network control, adaptive control, optimal control, etc.) is used to automatically adjust the relevant process parameters in the production process (such as the sorting density of the heavy medium separator, the amount of reagent added to the flotation machine, etc.) to correct the quality deviation and ensure that the product quality is stable within the qualified range. If no quality indicators are found to be abnormal, the product quality indicators will continue to be monitored. At the same time, new production data is continuously collected and analyzed, and the quality control model (i.e., the preset data analysis model) is continuously optimized. In the process of continuously optimizing the quality control model, an online learning algorithm can be used to enable the quality control model to update and adjust itself in real time according to new data samples. For example, using the incremental learning SVM algorithm, whenever a new batch of production data is generated, the quality control model can quickly learn and integrate the newly added data features, and continuously optimize its own decision boundaries, thereby more accurately predicting and controlling product quality.

[0067] In addition, the quality control and optimization module can also have the function of quality data visualization, which can intuitively display the real-time data, historical trends and process parameter adjustments of clean coal quality indicators to quality control personnel through data visualization charts (such as line charts, bar charts, scatter charts, etc.). In this way, the staff can understand the dynamic changes of product quality at a glance and can quickly evaluate the effects of process parameter adjustments so as to make further decisions and adjustments in a timely manner.

[0068] In order to further improve the effect of quality control, the quality control and optimization module also introduces a quality early warning mechanism. When the quality index is close to but has not exceeded the preset unqualified range, the quality control and optimization module will issue an early warning message in advance to remind the operator to pay close attention to the production process and make preventive fine-tuning of process parameters according to the early warning message. This helps to avoid further deterioration of quality problems, reduce the production of unqualified products, and improve the stability and reliability of the production process.

[0069] In actual application, the quality control and optimization module realizes data sharing and collaborative work with the production decision support module and the equipment maintenance warning module. For example, when the quality control and optimization module detects that the product quality problem may be caused by equipment failure or performance degradation, the quality control and optimization module will promptly feedback the relevant information to the equipment maintenance warning module so that the equipment maintenance warning module can conduct more in-depth inspection and diagnosis of the equipment; for another example, when the production decision support module adjusts the production plan or raw material ratio, the quality control and optimization module will also adjust the quality control strategy accordingly to ensure that product quality can be effectively guaranteed under different production conditions.

[0070] Through the synergistic effect of the various functions and measures of the above quality control and optimization modules, the coal preparation plant can achieve refined management and real-time dynamic optimization of product quality, effectively improve the stability and consistency of product quality, enhance product market competitiveness, and at the same time reduce economic losses and resource waste caused by quality problems, providing strong support for the sustainable development of the enterprise.

[0071] In the entire data application layer, each module collaborates with each other and shares data to form an organic whole. The production decision support module provides scientific and reasonable planning and guidance for the company's production operations; the equipment maintenance early warning module ensures the stable operation of the equipment, reduces downtime and maintenance costs; the quality control and optimization module ensures that product quality meets market demand and standards. Through the coordinated operation of these three main modules (i.e., the production decision support module, the equipment maintenance early warning module, and the quality control and optimization module), the coal preparation plant can fully tap the potential value of production data and realize intelligent, efficient and refined management of the production process, thereby occupying an advantageous position in the fierce market competition and promoting the sustainable development and progress of the enterprise.

[0072] Compared with the existing technology, the beneficial effects of the above-mentioned coal preparation plant production data full-process management system are mainly reflected in the following aspects: 1) Adopt a unified and adaptive data acquisition interface and architecture: It can automatically identify and be compatible with communication protocols of multiple data sources (such as Modbus, OPC-UA, MQTT, etc.), and realize efficient collection of various equipment data in coal preparation plants (including belt scales, equipment operation status sensors, coal quality analyzers and other data sources). Adopt a distributed acquisition architecture, divide the coal preparation plant into multiple areas, set up independent data acquisition nodes in each area, dynamically adjust the acquisition cycle and resource allocation according to the importance of the equipment and the frequency of data update, ensure the real-time performance of key data and the overall acquisition performance of the system, and effectively solve the problem of difficult data collection in coal preparation plants, which is different from the traditional fixed acquisition method and the lack of optimized acquisition architecture.

[0073] 2) Accurate data cleaning based on coal preparation plant data characteristics: Establish a variety of accurate outlier detection models for coal preparation plant-specific data such as belt scale flow and equipment temperature. For example, statistical analysis and machine learning are combined to accurately identify abnormal data to avoid misjudgment and missed judgment, and improve the accuracy and effectiveness of data cleaning. A variety of advanced repair and completion methods (such as linear interpolation, correlation-based completion, etc.) are used for outliers and missing values ​​to restore the authenticity and integrity of the data to the greatest extent. At the same time, professional filtering algorithms can be used to effectively remove noise interference and provide reliable data guarantee for data analysis, which is in sharp contrast to the traditional simple threshold judgment and rough data processing methods.

[0074] 3) Adopt hybrid storage architecture: Combining the advantages of relational databases (such as MySQL), time series databases (such as TDengine), and in-memory databases (such as Redis), reasonable storage allocation and management are carried out according to the type, characteristics and frequency of use of data. For example, for a large amount of historical production data, time series databases are used for efficient storage and fast query to facilitate trend analysis and long-term data mining; for equipment status data with high real-time requirements and key indicator data in the production process, they are stored in the in-memory database to ensure fast data reading and writing and real-time response; and data related to business logic is stored in the relational database to facilitate complex data association queries and transaction processing. Through this hybrid storage method, the efficiency of data storage and management is greatly improved, providing an efficient data support platform for data applications.

[0075] 4) Providing production decision support functions at the data application level: The production decision support module integrates multi-source data such as raw coal properties, equipment status, market demand, etc., and uses optimization algorithms and data analysis models to formulate optimal production plans (including raw coal blending plans, process parameter adjustments, etc.) to maximize economic benefits and meet product quality standards, changing the traditional way of making decisions based on manual experience.

[0076] 5) Provide equipment failure warning function at the data application level: The equipment maintenance warning module collects equipment operation data in real time, uses machine learning technology to build a fault prediction model, learns the normal operation mode and fault characteristic mode of the equipment based on the equipment's historical operation data and fault cases, accurately predicts equipment failures in advance, provides detailed fault diagnosis reports and maintenance suggestions, realizes preventive maintenance, reduces equipment downtime and maintenance costs, which is different from the traditional regular inspection and post-maintenance model.

[0077] 6) Provide quality control and optimization functions at the data application level: The quality control and optimization module uses sensors to realize real-time online monitoring of clean coal quality indicators, combines statistical process control (SPC) and data analysis models (such as multivariate linear regression model, principal component analysis model, etc.) to compare quality data with standards in real time, and automatically adjusts production process parameters to achieve closed-loop quality control and continuous optimization. At the same time, it has quality data visualization function, introduces quality early warning mechanism, and realizes data sharing and collaborative work with other modules, effectively improving product quality stability and consistency, which is significantly different from the existing lagging manual sampling detection and manual process adjustment methods.

[0078] In summary, the existing data processing technology of coal preparation plants has obvious deficiencies in data collection, data storage and data application, and cannot meet the needs of modern coal preparation industry for intelligent and refined production management. The above-mentioned coal preparation plant production data full-process management system comprehensively solves these problems through innovative data collection interfaces, efficient hybrid storage architecture and targeted data application methods, and realizes efficient processing and in-depth application of the entire process of coal preparation plant production data. It has significant technical advantages and application value, can bring significant economic and social benefits to coal preparation plants, and has obvious progress and innovation compared with existing technologies.

[0079] The embodiment of the present invention also provides a method for managing the entire process of production data of a coal preparation plant. The method for managing the entire process of production data of a coal preparation plant can be applied to the above-mentioned management system for the entire process of production data of a coal preparation plant. Figure 1 and Fig.12 As shown in the figure, the whole process management method of coal preparation plant production data can include: Step S1202 , the data acquisition subsystem 100 acquires equipment data and production data from different data sources in each plant through the data acquisition interface 101 .

[0080] Step S1204 , the data transmission subsystem 200 encrypts and transmits the equipment data and production data collected by the data collection subsystem 100 to the data cleaning subsystem 300 .

[0081] Step S1206, the data cleaning subsystem 300 cleans the received equipment data and production data; wherein the cleaning includes outlier detection, outlier processing, missing value detection and missing value processing.

[0082] In step S1208, the data storage subsystem 400 stores the data cleaned by the data cleaning subsystem 300 in a plurality of different types of databases using a hybrid storage method.

[0083] Step S1210: the data application subsystem 500 obtains target data from the data storage subsystem 400, and performs production plan optimization, equipment failure warning, and quality control and optimization based on the target data and preset data.

[0084] The above-mentioned full-process management method of coal preparation plant production data has achieved efficient processing and in-depth application of the entire process of coal preparation plant production data, provided strong support for the intelligent and refined production management of the coal preparation plant, and can effectively improve the overall operation level and economic benefits of the coal preparation plant.

[0085] As a possible implementation, see Figure 1 and Fig.12 As shown, the data acquisition subsystem 100 may include a data acquisition node pre-set for each plant; based on this, the step S1202 (i.e., the data acquisition subsystem 100 collects equipment data and production data from different data sources of each plant through the data acquisition interface 101) may include: Each of the data collection nodes performs the following operations: collects the equipment data and production data from different data sources of the corresponding factory through the data collection interface 101 , and sends the collected equipment data and production data to the data transmission subsystem 200 .

[0086] As a possible implementation, see Figure 1 and Fig.12 As shown, the data acquisition interface 101 may include a protocol identification module, a driver adaptation module and a data cache module; based on this, the step S1202 (i.e., the data acquisition subsystem 100 collects equipment data and production data from different data sources in each plant through the data acquisition interface 101) may include: The data acquisition subsystem 100 performs the following operations: identifying the communication protocol of each data source through the protocol identification module; calling the driver corresponding to the communication protocol of each data source to perform data acquisition operations through the driver adapter module; and temporarily storing the data collected by the driver adapter module through the data cache module.

[0087] As a possible implementation, see Figure 1 and Fig.12 As shown, the above-mentioned equipment data may include the operating status and data changes of the corresponding equipment, and each equipment is pre-set with an importance level and a data change threshold; based on this, the above-mentioned step S1202 (that is, the data acquisition subsystem 100 collects equipment data and production data from different data sources in each factory through the data acquisition interface 101) may also include: the data acquisition subsystem 100 adjusts the frequency of the data acquisition operation corresponding to each equipment based on the operating status, data changes, importance level and data change threshold of each equipment through the data acquisition interface 101.

[0088] As a possible implementation, see Figure 1 and Fig.12 As shown, the above step S1204 (i.e. the data transmission subsystem 200 encrypts and transmits the device data and production data collected by the data collection subsystem 100 to the data cleaning subsystem 300) may include: The data transmission subsystem 200 performs the following operations: uses the SSL / TLS protocol to encrypt the equipment data and production data collected by each data collection node, and uses the TCP / IP protocol to transmit the encrypted data corresponding to each data collection node to the data cleaning subsystem 300, so that the data cleaning subsystem 300 decrypts the received encrypted data.

[0089] As a possible implementation, see Figure 1 and Fig.12 As shown, the data cleaning subsystem 300 has pre-established a plurality of different types of outlier detection models; based on this, the step S1206 (i.e., the data cleaning subsystem 300 cleans the received device data and production data) may include: The data cleaning subsystem 300 performs the following operations: uses a plurality of the outlier detection models to perform outlier detection on the received equipment data and production data; performs missing value detection on the received equipment data and production data; performs interpolation processing on continuous data with detected missing values, and performs completion processing based on data correlation on data with detected missing values ​​and having correlation with other data.

[0090] As a possible implementation, see Figure 1 and Fig.12 As shown, the data storage subsystem 400 may include a relational database, a time series database, and a memory database; based on this, the step S1208 (i.e., the data storage subsystem 400 uses a hybrid storage method to store the data cleaned by the data cleaning subsystem 300 in multiple different types of databases) may include: The data storage subsystem 400 performs the following operations: storing the structured data cleaned by the data cleaning subsystem 300 into a relational database; storing the time series data cleaned by the data cleaning subsystem 300 into a time series database; and storing the real-time data cleaned by the data cleaning subsystem 300 into a memory database.

[0091] As a possible implementation, see Figure 1 and Fig.12 As shown, the data application subsystem 500 may include a production plan optimization module, an equipment failure warning module, and a quality control and optimization module; based on this, the step S1210 (i.e., the data application subsystem 500 obtains target data from the data storage subsystem 400, and performs production plan optimization, equipment failure warning, and quality control and optimization based on the target data and preset data) may include: The production plan optimization module performs the following operations: obtaining first target data from the data storage subsystem, establishing a production optimization model based on the first target data, and then iteratively optimizing the production plan based on the production optimization model; wherein, each time the production plan is optimized, the production optimization model is used to generate the current production plan, and the economic benefit and product quality of the current production plan are evaluated based on preset economic benefit standards and preset product quality standards, and then the parameters of the production optimization model are adjusted according to the economic benefit and product quality of the current production plan, so that the production optimization model is used to generate the production plan corresponding to the next optimized production plan when the production plan is optimized next time; The equipment fault warning module performs the following operations: obtaining second target data from the data storage subsystem, performing feature extraction on the second target data to obtain corresponding feature data, then performing fault prediction on the feature data using a pre-trained fault prediction model to generate fault warning information, and sending the fault warning information to a terminal device corresponding to the equipment maintenance personnel; The quality control and optimization module performs the following operations: obtaining third target data from the data storage subsystem, and comparing the third target data with a preset product quality standard, so as to determine whether the third target data indicates abnormal product quality based on the comparison result; if the third target data indicates abnormal product quality, using the third target data with a preset data analysis model to analyze key factors affecting product quality, and generating a quality control strategy based on the key factors; and adjusting process parameters based on a preset feedback control algorithm and the quality control strategy.

[0092] As a possible implementation, see Figure 1 and Fig.12 As shown, the above-mentioned coal preparation plant production data full process management method may also include: The quality control and optimization module performs the following operations: generates first information based on the key factors, and sends the first information to the equipment failure warning module so that the equipment failure warning module uses the first information as fourth target data; adjusts the quality control strategy based on the data generated by the production plan optimization module during the iterative optimization of the production plan.

[0093] An embodiment of the present invention provides a method for managing the entire process of production data of a coal preparation plant, and its implementation principle and the technical effects produced are the same as those of the aforementioned embodiment of the full-process management system for production data of a coal preparation plant. For the sake of brief description, for matters not mentioned in the embodiment of the method for managing the entire process of production data of a coal preparation plant, reference may be made to the corresponding contents in the aforementioned embodiment of the full-process management system for production data of a coal preparation plant.

[0094] Unless otherwise specifically stated, the relative steps, numerical expressions and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0095] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that can be executed by a processor. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., various media that can store program codes.

[0096] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0097] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A full-process management system for production data of a coal preparation plant, characterized in that: include: A data acquisition subsystem, a data transmission subsystem, a data cleaning subsystem, a data storage subsystem and a data application subsystem; the data acquisition subsystem is pre-set with a data acquisition interface for docking with the data transmission subsystem; The data acquisition subsystem is used to collect equipment data and production data from different data sources in each plant through the data acquisition interface; The data transmission subsystem is used to encrypt and transmit the equipment data and production data collected by the data collection subsystem to the data cleaning subsystem; The data cleaning subsystem is used to clean the received equipment data and production data; wherein the cleaning includes outlier detection, outlier processing, missing value detection and missing value processing; The data storage subsystem is used to store the data cleaned by the data cleaning subsystem into multiple different types of databases in a hybrid storage manner; The data application subsystem is used to obtain target data from the data storage subsystem, and perform production plan optimization, equipment failure warning, and quality control and optimization based on the target data and preset data.

2. The coal plant production data full process management system according to claim 1 is characterized in that: The data acquisition subsystem includes a data acquisition node pre-set for each factory building; each of the data acquisition nodes is used to collect equipment data and production data from different data sources of the corresponding factory building through the data acquisition interface, and send the collected equipment data and production data to the data transmission subsystem.

3. The coal plant production data full-process management system according to claim 1 is characterized in that: The data acquisition interface includes a protocol identification module, a driver adapter module and a data cache module; the protocol identification module is used to identify the communication protocol of each data source; the driver adapter module is used to call the driver program corresponding to the communication protocol of each data source to perform data acquisition operations; the data cache module is used to temporarily store the data collected by the driver adapter module.

4. The coal plant production data full process management system according to claim 3 is characterized in that: The device data includes the operating status and data changes of the corresponding device, and each device is pre-set with an importance level and a data change threshold; the data acquisition interface is used to adjust the frequency of data acquisition operations corresponding to each device based on the operating status, data changes, importance level and data change threshold of each device.

5. The coal plant production data full-process management system according to claim 2 is characterized in that: The data transmission subsystem is also used to encrypt the equipment data and production data collected by each data acquisition node using the SSL / TLS protocol, and to transmit the encrypted data corresponding to each data acquisition node to the data cleaning subsystem using the TCP / IP protocol; the data cleaning subsystem is also used to decrypt the received encrypted data.

6. The coal plant production data full-process management system according to claim 1 is characterized in that: The data cleaning subsystem pre-establishes multiple different types of outlier detection models; the data cleaning subsystem is also used to: use multiple of the outlier detection models to perform outlier detection on the received equipment data and production data; perform missing value detection on the received equipment data and production data; interpolate continuous data with detected missing values, and perform completion processing based on data correlation on data with detected missing values ​​and correlation with other data.

7. The coal plant production data full-process management system according to claim 1 is characterized in that: The data storage subsystem includes a relational database, a time series database and a memory database; the data storage subsystem is also used to: store the structured data cleaned by the data cleaning subsystem in the relational database; store the time series data cleaned by the data cleaning subsystem in the time series database; store the real-time data cleaned by the data cleaning subsystem in the memory database.

8. The coal plant production data full-process management system according to claim 1 is characterized in that: The data application subsystem includes a production plan optimization module, an equipment failure early warning module, and a quality control and optimization module; The production plan optimization module is used to: obtain first target data from the data storage subsystem, establish a production optimization model based on the first target data, and then iteratively optimize the production plan based on the production optimization model; wherein, each time the production plan is optimized, the production optimization model is used to generate the current production plan, and the economic benefit and product quality of the current production plan are evaluated based on preset economic benefit standards and preset product quality standards, and then the parameters of the production optimization model are adjusted according to the economic benefit and product quality of the current production plan, so that the production optimization model is used to generate the production plan corresponding to the next optimized production plan when the production plan is optimized next time; The equipment fault warning module is used to: obtain second target data from the data storage subsystem, perform feature extraction on the second target data to obtain corresponding feature data, then perform fault prediction on the feature data through a pre-trained fault prediction model to generate fault warning information, and send the fault warning information to a terminal device corresponding to the equipment maintenance personnel; The quality control and optimization module is used to: obtain third target data from the data storage subsystem, and compare the third target data with the preset product quality standard to determine whether the third target data represents abnormal product quality based on the comparison result; if the third target data represents abnormal product quality, use the third target data to analyze the key factors affecting product quality using the preset data analysis model, and generate a quality control strategy based on the key factors; adjust the process parameters based on the preset feedback control algorithm and the quality control strategy.

9. The coal plant production data full-process management system according to claim 8, characterized in that: The quality control and optimization module is further used to: generate first information based on the key factors, and send the first information to the equipment failure warning module, so that the equipment failure warning module uses the first information as fourth target data; The quality control strategy is adjusted based on the data generated by the production plan optimization module during the iterative optimization of the production plan.

10. A method for managing the entire production process of a coal preparation plant, characterized in that: The full-process management system for production data of a coal preparation plant applied to any one of claims 1 to 9 comprises: The data acquisition subsystem collects equipment data and production data from different data sources in each plant through the data acquisition interface; The data transmission subsystem encrypts and transmits the equipment data and production data collected by the data collection subsystem to the data cleaning subsystem; The data cleaning subsystem cleans the received equipment data and production data; wherein the cleaning includes outlier detection, outlier processing, missing value detection and missing value processing; The data storage subsystem uses a hybrid storage method to store the data cleaned by the data cleaning subsystem into multiple databases of different types; The data application subsystem obtains target data from the data storage subsystem, and performs production plan optimization, equipment failure warning, and quality control and optimization based on the target data and preset data.

Images