Data decision-making method and device based on intelligent coal preparation plant

Abstract

This invention discloses a data-driven decision-making method and apparatus for intelligent coal preparation plants, relating to the field of coal mining industry technology. Its main objective is to address the problem of poor accuracy in existing data monitoring based on coal preparation data. The method includes: acquiring coal preparation operation data to be decided, and classifying it according to data fluctuation states to obtain a first type of coal preparation data and a second type of coal preparation data; determining a first data failure offset and a second data failure offset; if the first data failure offset is less than the second data failure offset, then selecting target data from the first and second types of coal preparation data based on the difference between the first and second data failure offsets; otherwise, selecting target data from the first and second types of coal preparation data based on the sum of the first and second data failure offsets; and clearing the target data if it fails the coal preparation data probability verification.

Description

Technical Field

[0001] This invention relates to the field of coal mining industry technology, and in particular to a data decision-making method and device based on an intelligent coal preparation plant. Background Technology

[0002] With the introduction and use of intelligent control equipment in the coal mining industry, coal preparation plants have also upgraded and optimized their intelligent equipment. In the data processing of various equipment within the coal preparation plant, a large amount of coal preparation data is filtered and processed to achieve accurate data monitoring.

[0003] Currently, existing methods for filtering coal preparation data only clean the data based on the overflow range, removing a small amount of distorted data. However, since there is still a lot of invalid data in the undistorted data, data monitoring based on this coal preparation data will greatly affect the monitoring effect if it is not removed. Therefore, there is an urgent need for a data-driven decision-making method based on intelligent coal preparation plants to solve the above problems. Summary of the Invention

[0004] In view of this, the present invention provides a data decision-making method and device based on intelligent coal preparation plants, the main purpose of which is to solve the problem of poor accuracy of existing data monitoring based on coal preparation data.

[0005] According to one aspect of the present invention, a data-driven decision-making method based on an intelligent coal preparation plant is provided, comprising:

[0006] The coal preparation operation data to be decided is obtained, and the coal preparation operation data is classified according to the data fluctuation state to obtain the first type of coal preparation data and the second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges.

[0007] The first data failure offset of the first type of coal preparation data is determined based on the first screening weight of the first type of coal preparation data, and the second data failure offset of the second type of coal preparation data is determined based on the second screening weight of the second type of coal preparation data.

[0008] If the first data failure offset is less than the second data failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset.

[0009] If the first data failure offset is greater than or equal to the second failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset.

[0010] The target data is subjected to coal preparation data probability verification according to the collection nodes in the coal preparation operation. If the target data fails the coal preparation data probability verification, the target data is cleared, so that the first type of coal preparation data and the second type of coal preparation data after clearing the target data are determined as coal preparation data to be monitored.

[0011] Furthermore, the classification of the coal preparation operation data according to the data fluctuation state to obtain the first type of coal preparation data and the second type of coal preparation data includes:

[0012] The equipment operating parameters of the coal preparation equipment are retrieved, and the data fluctuation status matching the equipment operating parameters is obtained. The data fluctuation status includes stable fluctuation status and offset fluctuation status.

[0013] If the coal preparation operation data matches the first fluctuation threshold range corresponding to the stable fluctuation state, then the coal preparation operation data is determined as the first type of coal preparation data;

[0014] If the coal preparation operation data matches the second fluctuation threshold range corresponding to the offset fluctuation state, then the coal preparation operation data is determined as the second type of coal preparation data;

[0015] Wherein, the first fluctuation threshold range is smaller than the second fluctuation threshold range.

[0016] Further, determining the first data failure offset of the first type of coal preparation data based on the first screening weight of the first type of coal preparation data includes:

[0017] Extract the first screening weight that matches the first type of coal preparation data, where the sum of the weight coefficients of the first screening weight is 1;

[0018] The first data failure offset is obtained by performing a weighted summation operation based on the first screening weight and the first type of coal preparation data.

[0019] The determination of the second data failure offset of the second type of coal preparation data based on the second screening weight of the second type of coal preparation data includes:

[0020] Extract the second screening weights that match the second type of coal preparation data, where the offset weight coefficients in the second screening weights are equal to the sum of the non-offset weight coefficients;

[0021] The second data failure offset is obtained by performing a weighted summation operation based on the second screening weight and the second type of coal preparation data.

[0022] Further, the step of filtering target data from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset includes:

[0023] Calculate the difference between the first data failure offset and the second data failure offset to obtain the failure offset difference value;

[0024] The coal preparation data from the first type of coal preparation data and the second type of coal preparation data that are greater than the failure offset difference are identified as target data; or...

[0025] Target data is extracted from the first type of coal preparation data and the second type of coal preparation data according to the preset offset difference multiple of the failure offset difference.

[0026] Further, the step of filtering target data from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset includes:

[0027] Calculate the sum of the first data failure offset and the second data failure offset to obtain the total failure offset value;

[0028] The coal preparation data exceeding the total failure offset value in the first type of coal preparation data and the second type of coal preparation data are identified as target data; or...

[0029] Target data is extracted from the first type of coal preparation data and the second type of coal preparation data according to a preset offset multiple of the total failure offset.

[0030] Furthermore, the step of verifying the probability of coal preparation data for the target data according to the acquisition nodes in the coal preparation operation includes:

[0031] Based on the acquisition node selected from the coal preparation operation, the acquisition node is used to characterize the execution node corresponding to the coal preparation equipment for coal preparation operation data acquisition when the coal preparation operation is performed;

[0032] The verification object of the acquisition node is matched, and the verification threshold matching the verification object is obtained, so as to determine whether the target data passes the coal preparation data probability verification through the verification threshold. The verification object is used to characterize the data object for the probability verification of the target data. The data object includes at least one of data size, data type, data path, and data proportion.

[0033] Furthermore, the method also includes:

[0034] Under the condition that the target data passes the probability verification of the coal preparation data, the trained data decision classification model is retrieved to perform decision classification processing on the target data to obtain the data decision result. The data decision classification model is obtained by training a binary tree model based on the data decision sample set.

[0035] If the data decision result is an abnormal decision classification, the first fluctuation threshold range of the stable fluctuation state and the second fluctuation threshold range of the offset fluctuation state are adjusted respectively to re-make the data decision.

[0036] According to another aspect of the present invention, a data decision-making device based on an intelligent coal preparation plant is provided, comprising:

[0037] The acquisition module is used to acquire coal preparation operation data to be decided, and classify the coal preparation operation data according to the data fluctuation state to obtain a first type of coal preparation data and a second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges.

[0038] The determination module is used to determine the first data failure offset of the first type of coal preparation data based on the first screening weight of the first type of coal preparation data, and to determine the second data failure offset of the second type of coal preparation data based on the second screening weight of the second type of coal preparation data.

[0039] The first filtering module is used to filter target data from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset if the first data failure offset is less than the second data failure offset.

[0040] The second filtering module is used to filter target data from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset if the first data failure offset is greater than or equal to the second failure offset.

[0041] The clearing module is used to perform coal preparation data probability verification on the target data according to the acquisition nodes in the coal preparation operation, and clear the target data if the target data fails the coal preparation data probability verification, so as to determine the first type of coal preparation data and the second type of coal preparation data after clearing the target data as coal preparation data to be monitored.

[0042] Furthermore, the acquisition module includes:

[0043] The retrieval unit is used to retrieve the equipment operating parameters of the coal preparation equipment and obtain the data fluctuation status that matches the equipment operating parameters. The data fluctuation status includes a stable fluctuation status and an offset fluctuation status.

[0044] The determining unit is used to determine the coal preparation operation data as the first type of coal preparation data if the coal preparation operation data matches the first fluctuation threshold range corresponding to the stable fluctuation state.

[0045] The determining unit is used to determine the coal preparation operation data as the second type of coal preparation data if the coal preparation operation data matches the second fluctuation threshold range corresponding to the offset fluctuation state.

[0046] Wherein, the first fluctuation threshold range is smaller than the second fluctuation threshold range.

[0047] Furthermore, the determining module includes:

[0048] The first extraction unit is used to extract the first screening weight that matches the first type of coal preparation data, and the sum of the weight coefficients of the first screening weight is 1.

[0049] The first calculation unit is used to perform a weight summation operation based on the first screening weight and the first type of coal preparation data to obtain the first data failure offset.

[0050] The second extraction unit is used to extract the second screening weights that match the second type of coal preparation data. The offset weight coefficients in the second screening weights are equal to the sum of the non-offset weight coefficients.

[0051] The second calculation unit is used to perform a weighted summation operation based on the second screening weight and the second type of coal preparation data to obtain the second data failure offset.

[0052] Furthermore, the first screening module includes:

[0053] The first calculation unit is used to calculate the difference between the first data failure offset and the second data failure offset to obtain the failure offset difference value;

[0054] The first determining unit is configured to determine the coal preparation data in the first type of coal preparation data and the second type of coal preparation data that are greater than the failure offset difference as target data; or...

[0055] The first extraction unit is used to extract target data from the first type of coal preparation data and the second type of coal preparation data according to a preset offset difference multiple of the failure offset difference.

[0056] Furthermore, the second screening module includes:

[0057] The second calculation unit is used to calculate the sum of the first data failure offset and the second data failure offset to obtain the total failure offset value;

[0058] The second determining unit is used to determine the coal preparation data in the first type of coal preparation data and the second type of coal preparation data that are greater than the total failure offset value as target data; or...

[0059] The second extraction unit is used to extract target data from the first type of coal preparation data and the second type of coal preparation data according to a preset offset multiple of the total failure offset value.

[0060] Furthermore, the clearing module includes:

[0061] The selected unit is used to select a data acquisition node from the coal preparation operation, wherein the data acquisition node is used to characterize the execution node corresponding to the coal preparation equipment for collecting coal preparation operation data when the coal preparation operation is performed during the coal preparation operation;

[0062] The matching unit is used to match the verification object of the acquisition node and obtain the verification threshold that matches the verification object, so as to determine whether the target data passes the coal preparation data probability verification through the verification threshold. The verification object is used to characterize the data object for the probability verification of the target data. The data object includes at least one of data size, data type, data path, and data proportion.

[0063] Furthermore, the device also includes:

[0064] The retrieval module is used to retrieve a pre-trained data decision classification model to perform decision classification processing on the target data under the condition that the target data passes the coal preparation data probability verification, and to obtain the data decision result. The data decision classification model is obtained by training a binary tree model based on the data decision sample set.

[0065] The adjustment module is used to adjust the first fluctuation threshold range of the stable fluctuation state and the second fluctuation threshold range of the offset fluctuation state respectively if the data decision result is an abnormal decision classification, so as to re-make the data decision.

[0066] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored therein, the executable instruction causing a processor to perform operations corresponding to the data decision-making method based on the intelligent coal preparation plant described above.

[0067] According to another aspect of the present invention, a terminal is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0068] The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the data decision-making method based on the intelligent coal preparation plant described above.

[0069] By employing the above-described technical solutions, the technical solutions provided by the embodiments of the present invention have at least the following advantages:

[0070] This invention provides a data-driven decision-making method and apparatus for intelligent coal preparation plants. Compared with existing technologies, this invention acquires coal preparation operation data to be decided and classifies the data according to its fluctuation state to obtain a first type of coal preparation data and a second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges. A first data failure offset is determined based on a first screening weight of the first type of coal preparation data, and a second data failure offset is determined based on a second screening weight of the second type of coal preparation data. If the first data failure offset is less than the second data failure offset, the difference between the first and second data failure offsets is used to select the first type of coal preparation data and the second type of coal preparation data from the second type of coal preparation data. Target data is selected from the coal preparation data. If the first data failure offset is greater than or equal to the second data failure offset, target data is selected from the first and second types of coal preparation data based on the sum of the first and second data failure offsets. The target data is then subjected to coal preparation data probability verification according to the acquisition nodes in the coal preparation operation. If the target data fails the coal preparation data probability verification, the target data is cleared. The first and second types of coal preparation data after clearing the target data are identified as coal preparation data to be monitored. This enables data filtering decisions for coal preparation data, accurately identifies invalid coal preparation data, improves the accuracy of data processing based on coal preparation data, and achieves effective decision-making on coal preparation data.

[0071] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0072] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0073] Figure 1 This invention provides a flowchart of a data-driven decision-making method for intelligent coal preparation plants.

[0074] Figure 2 A schematic diagram of an intelligent coal preparation plant system architecture based on ISA95 provided by an embodiment of the present invention is shown.

[0075] Figure 3 This invention illustrates an intelligent coal preparation plant big data decision-making system provided by an embodiment of the present invention;

[0076] Figure 4 This diagram illustrates a device composition block diagram provided by an embodiment of the present invention;

[0077] Figure 5 A schematic diagram of the structure of a terminal provided in an embodiment of the present invention is shown. Detailed Implementation

[0078] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0079] In the current practice of filtering coal preparation data, cleaning is often based solely on data overflow, removing only a small amount of distorted data. However, since many invalid data points remain within the undistorted data set, monitoring based on this coal preparation data without removing them would significantly impact the effectiveness of the monitoring. This invention provides a data-driven decision-making method for intelligent coal preparation plants, such as… Figure 1 As shown, the method includes:

[0080] 101. Obtain the coal preparation operation data to be decided, and classify the coal preparation operation data according to the data fluctuation status to obtain the first type of coal preparation data and the second type of coal preparation data.

[0081] In this embodiment of the invention, the current execution end is the system server of an intelligent coal preparation plant. Data is acquired through data connections with various coal preparation subsystems. The coal preparation equipment includes, but is not limited to, raw coal subsystems, clean coal subsystems, and transportation equipment. At this time, the coal preparation operation data to be decided is the operation data collected from the coal preparation equipment, which is to be processed through data monitoring, including but not limited to raw coal ash content data, raw coal moisture content data, raw coal calorific value data, raw coal transmission speed, and clean coal transmission speed, etc., so as to make data-driven decisions based on the coal preparation operation data. After acquiring the coal preparation operation data, it is classified according to the data fluctuation state. Here, the data fluctuation state is used to characterize the state of the data within different data fluctuation ranges. The data fluctuation state includes a stable fluctuation state and a deviated fluctuation state, so as to classify the coal preparation operation data into a first category of coal preparation data and a second category of coal preparation data.

[0082] It should be noted that in a specific coal preparation enterprise system server implementation scenario, the coal preparation plant server system, as a 5G+industrial internet platform, can currently complete the interconnection between the intelligent process control system and various subsystems, thereby completing the interconnection between the three levels of systems: "intelligent process control system", "production execution system" and "enterprise resource planning system". After data collection and fusion through full network connectivity technology, the database system is used to execute the current operation.

[0083] 102. Determine the first data failure offset of the first type of coal preparation data based on the first screening weight of the first type of coal preparation data, and determine the second data failure offset of the second type of coal preparation data based on the second screening weight of the second type of coal preparation data.

[0084] In this embodiment of the invention, after classifying coal preparation data into a first category and a second category based on stable fluctuation states and offset fluctuation states, in order to determine whether there is a data offset between the first and second categories of coal preparation data for filtering usable data, a data failure offset is calculated based on the filtering weights to determine whether the data has an offset, i.e., whether it affects the accuracy of monitoring and processing. The first filtering weight is the weight corresponding to the first category of coal preparation data used to calculate the first data failure offset, and the second filtering weight is the weight corresponding to the second category of coal preparation data used to calculate the second data failure offset. In this embodiment of the invention, the first filtering weight and the second filtering weight are stored separately, and no specific limitations are imposed on this embodiment.

[0085] It should be noted that since the first and second types of coal preparation data are obtained by classifying coal preparation operation data, the number of weight items in the screening weight is the same as the number of data types in the coal preparation operation data, and can be configured in advance. For example, if the first type of coal preparation data includes raw coal ash content data and raw coal moisture content data, then the first screening weight includes raw coal ash content screening weight and raw coal moisture content screening weight. This embodiment of the invention does not make specific limitations.

[0086] 103. If the first data failure offset is less than the second data failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset.

[0087] In this embodiment of the invention, since the data failure offset is used to characterize the deviation between the data and the data that can be used for coal preparation, after calculating the first data failure offset and the second data failure offset, they are compared. When the first failure offset is less than the second failure offset, in order to screen out the target data to be cleaned from the coal preparation data, the target data is screened out from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset, respectively, as the target data to be cleared for coal preparation data probability verification.

[0088] 104. If the first data failure offset is greater than or equal to the second failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset.

[0089] In this embodiment of the invention, since the data failure offset is used to characterize the deviation between the data and the data that can be used for coal preparation, after calculating the first data failure offset and the second data failure offset, they are compared. When the first failure offset is greater than or equal to the second failure offset, in order to screen out the target data to be cleaned from the coal preparation data, the target data is screened out from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset, respectively, as the target data to be cleaned for coal preparation data probability verification.

[0090] 105. Perform coal preparation data probability verification on the target data according to the acquisition nodes in the coal preparation operation, and clear the target data if the target data fails the coal preparation data probability verification, so as to determine the first type of coal preparation data and the second type of coal preparation data after clearing the target data as coal preparation data to be monitored.

[0091] In this embodiment of the invention, since the data processing platform of a coal preparation enterprise system requires different data at each process node during coal preparation operations, for each acquisition node in the coal preparation operation, the target data is first subjected to a probability verification, that is, to determine whether the data to be cleaned exists within the normal data acquisition range of the coal preparation operation. When the probability verification of coal preparation data fails, it indicates that this target data is truly abnormal in the data acquisition of the coal preparation operation, and therefore, the target data is cleared. At this time, the cleared target data can be one or multiple, and can be from different categories of data. Thus, the first category of coal preparation data and the second category of coal preparation data after clearing the target data are determined as the coal preparation data to be monitored, and data monitoring processing is performed.

[0092] It should be noted that in the ISA95-based intelligent coal preparation plant system architecture scenario, the architecture conforms to the ISA international standard. The integration of the coal preparation plant control system with the enterprise system is divided into three levels: "Intelligent Process Control System (IPCS)," "Production Execution System (MES)," and "Enterprise Resource Planning System (ERP)." At this point, the system deeply analyzes and positions the business function modules of the intelligent coal preparation plant distributed across these three levels, providing visual guidance for the modular and standardized construction of the intelligent coal preparation plant, making the construction and implementation of the intelligent coal preparation plant more convenient and clear. First, a 5G+Industrial Internet platform is built to complete the interconnection between the various subsystems of the underlying "Intelligent Process Control System" (IPCS). Then, the interconnection between the three-level systems—"Intelligent Process Control System" (IPCS), "Production Execution System" (MES), and "Enterprise Resource Planning System" (ERP)—is completed. Data is then collected and integrated using technologies such as "full network connectivity" and stored in a system such as... Figure 2 The database system is shown. Next, a hybrid cloud platform server supporting both public and private clouds is built, deploying various database system software, including real-time historical databases, relational databases, multimedia databases, and other types of databases. Then, a web portal is deployed, containing all the business function modules of the "Production Execution System (MES)" and "Enterprise Resource Planning System (ERP)," as well as the monitoring and configuration software SCADA from the "Intelligent Process Control System (IPCS)." The ERP business function modules mainly include market analysis and business management, procurement, transportation and sales management, financial management, cost management, human resource management, and safety management. The MES business function modules mainly include output analysis, quality analysis, performance analysis, downtime analysis, energy consumption analysis, a comprehensive diagnostic platform, implementation status, production scheduling, production planning, intelligent video surveillance, technical management, video linkage, coal quality management, transportation and sales management, material management, spare parts management, access control, and OA office automation and mobile collaborative management.

[0093] In another embodiment of the invention, for further explanation and limitation, the step of classifying the coal preparation operation data according to the data fluctuation state to obtain a first type of coal preparation data and a second type of coal preparation data includes:

[0094] Retrieve the equipment operating parameters of the coal preparation equipment and obtain the data fluctuation status that matches the equipment operating parameters;

[0095] If the coal preparation operation data matches the first fluctuation threshold range corresponding to the stable fluctuation state, then the coal preparation operation data is determined as the first type of coal preparation data;

[0096] If the coal preparation operation data matches the second fluctuation threshold range corresponding to the offset fluctuation state, then the coal preparation operation data is determined as the second type of coal preparation data.

[0097] In this embodiment of the invention, to accurately classify coal preparation operation data into two data content ranges for data filtering, so as to meet the needs of detecting data anomalies in coal preparation data within different data fluctuation ranges, specifically, when classifying coal preparation operation data based on data fluctuation status, the equipment operating parameters of each coal preparation device are first retrieved. At this time, the equipment operating parameters include, but are not limited to, the coal preparation quantity of raw coal preparation equipment, the coal preparation quantity of clean coal preparation equipment, equipment performance energy consumption, etc. For coal preparation operations, different coal preparation devices generate different data that needs to be monitored. Therefore, in this embodiment of the invention, data fluctuation status corresponding to different equipment operating parameters is pre-configured to classify coal preparation operation data according to different data fluctuation statuses. The data fluctuation states include stable fluctuation states and offset fluctuation states. The stable fluctuation state is used to characterize the collected data as belonging to the cosine or sine fluctuation state and being within a stable amplitude. The offset fluctuation state is used to characterize the collected data as belonging to the cosine or sine fluctuation state and occasionally exhibiting abnormal increases or decreases. Both the stable fluctuation state and the offset fluctuation state correspond to a fluctuation threshold range. Moreover, when the data is not in a stable fluctuation state, it must belong to an offset fluctuation state. Therefore, the corresponding configuration of the first fluctuation threshold range is smaller than the second fluctuation threshold range. This embodiment of the invention does not make specific limitations.

[0098] It should be noted that during classification, if the coal preparation operation data falls within the first fluctuation threshold range of a stable fluctuation state, it indicates that the coal preparation operation data conforms to a stable fluctuation state, and therefore, this coal preparation operation data is identified as the first type of coal preparation data. If the coal preparation operation data falls within the second fluctuation threshold range of a deviated fluctuation state, it indicates that the coal preparation operation data conforms to a deviated fluctuation state, and therefore, this coal preparation operation data is identified as the second type of coal preparation data. The setting of the first and second fluctuation threshold ranges can be configured according to the specific data content of the various collected data. For example, the first fluctuation threshold range for ash content data is a to a+10, and the second fluctuation threshold range is a+11 to a+30, where a is a positive real number. The first fluctuation threshold range for moisture content data is b to b+10, and the second fluctuation threshold range is b+11 to b+30, where b is a positive real number. Since a and b are not the same, this embodiment of the invention does not impose specific limitations.

[0099] In another embodiment of the invention, for further explanation and limitation, the step of determining the first data failure offset of the first type of coal preparation data based on the first screening weight of the first type of coal preparation data includes:

[0100] Extract the first screening weight that matches the first type of coal preparation data;

[0101] The first data failure offset is obtained by performing a weighted summation operation based on the first screening weight and the first type of coal preparation data.

[0102] In this embodiment of the invention, to accurately determine whether the classified coal preparation data has become invalid due to excessive offset, specifically, when calculating the first data failure offset based on the first screening weight of the first type of coal preparation data, the first screening weight matching the first type of coal preparation data is first extracted. Then, a weighted summation calculation is performed based on this first screening weight and the first type of coal preparation data to obtain the first data failure offset. Since the first type of coal preparation data is classified according to a stable fluctuation state, the differences between the data are small and the fluctuations are also small. Therefore, the sum of the weight coefficients of the first screening weight is configured to be 1. Thus, by summing, the specific value of data in the first type of coal preparation data that has become invalid due to excessive offset is calculated, thereby improving the calculation stability of the data failure offset.

[0103] In another embodiment of the invention, for further explanation and limitation, the step of determining the second data failure offset of the second type of coal preparation data based on the second screening weight of the second type of coal preparation data includes:

[0104] Extract the second screening weight that matches the second type of coal preparation data;

[0105] The second data failure offset is obtained by performing a weighted summation operation based on the second screening weight and the second type of coal preparation data.

[0106] In this embodiment of the invention, to accurately determine whether the classified coal preparation data has become invalid due to excessive offset, specifically, based on the second screening weight matched with the second type of coal preparation data, a weighted summation calculation is performed between this second screening weight and the second type of coal preparation data to obtain the second data failure offset. Since the second type of coal preparation data is classified according to offset fluctuation states, the differences between data are large and the fluctuations are also significant. Therefore, the offset weight coefficient in the second screening weight is equal to the sum of the non-offset weight coefficients. Thus, by summing, the specific value of data in the second type of coal preparation data that has become invalid due to large offset is calculated, thereby improving the stability of the data failure offset calculation. It should be noted that when configuring the second screening weight, the offset weight is the data item whose data offset exceeds the abnormal range, and the non-offset weight is the data item whose data offset is within the normal range. In this case, their weight coefficients are generally configured to a large value, such that the offset weight coefficient equals the sum of the non-offset weight coefficients. This embodiment of the invention does not impose specific limitations.

[0107] In another embodiment of the invention, for further explanation and limitation, the step of filtering target data from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset includes:

[0108] Calculate the difference between the first data failure offset and the second data failure offset to obtain the failure offset difference value;

[0109] The coal preparation data from the first type of coal preparation data and the second type of coal preparation data that are greater than the failure offset difference are identified as target data; or...

[0110] Target data is extracted from the first type of coal preparation data and the second type of coal preparation data according to the preset offset difference multiple of the failure offset difference.

[0111] In this embodiment of the invention, to accurately filter target data based on the failure offset to determine whether data cleaning is necessary, the filtering process based on the difference between the first and second data failure offsets specifically involves first calculating the difference between them as the failure offset difference. After determining the failure offset difference, it can be directly compared with the first and second types of coal preparation data. Coal preparation data in the first and second types of coal preparation data that exceeds this failure offset difference is identified as target data. Alternatively, a preset offset difference multiple, such as two times or three times, can be pre-configured. After determining the failure offset difference, a value is calculated based on the preset offset difference multiple, and data in the first and second types of coal preparation data exceeding this value is identified as target data. This embodiment of the invention does not impose specific limitations.

[0112] In another embodiment of the invention, for further explanation and limitation, the step of filtering target data from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset includes:

[0113] Calculate the sum of the first data failure offset and the second data failure offset to obtain the total failure offset value;

[0114] The coal preparation data exceeding the total failure offset value in the first type of coal preparation data and the second type of coal preparation data are identified as target data; or...

[0115] Target data is extracted from the first type of coal preparation data and the second type of coal preparation data according to a preset offset multiple of the total failure offset.

[0116] In this embodiment of the invention, to accurately filter target data based on the failure offset to determine whether data cleaning is necessary, the filtering process based on the sum of a first data failure offset and a second data failure offset specifically involves first calculating the sum of the first and second data failure offsets as the total failure offset value. After determining the total failure offset value, it can be directly compared with the first and second types of coal preparation data. Coal preparation data in the first and second types of coal preparation data that exceeds this total failure offset value is identified as target data. Alternatively, a preset offset multiple, such as two times or three times, can be pre-configured. After determining the total failure offset value, a value is calculated based on the preset offset multiple, and data in the first and second types of coal preparation data exceeding this value is identified as target data. This embodiment of the invention does not impose specific limitations.

[0117] In another embodiment of the invention, for further explanation and limitation, the step of verifying the probability of coal preparation data of the target data according to the acquisition nodes in the coal preparation operation includes:

[0118] Based on the acquisition node selected from the coal preparation operation;

[0119] Match the verification object of the collection node and obtain the verification threshold that matches the verification object.

[0120] In this embodiment of the invention, to further verify the invalid offset data identified as needing to be cleared and to avoid accidental clearing, verification is performed on specific acquisition nodes in the coal preparation operation. Specifically, firstly, an execution node is selected from the coal preparation operation to characterize the data acquisition of coal preparation equipment operation data during the coal preparation process. That is, the acquisition node is used to characterize the execution node corresponding to the data acquisition of coal preparation equipment operation data during the coal preparation process. Then, a verification threshold is obtained based on the verification object matched by this acquisition node, and the target data is used to determine whether it passes the coal preparation data probability verification. The verification object is used to characterize the data object for the probability verification of the target data. The data object includes at least one of data size, data type, data path, and data proportion, so as to further determine whether the target data is possible based on the verification threshold corresponding to at least one of the data size, data type, data path, and data proportion in the target data. At this point, the verification threshold is a value that may exist in the coal preparation scenario for the data size, data type, data path, and data proportion. If the verification threshold is matched, it means that the target data will not be cleared in the end. If the verification threshold is not matched, the target data will be cleared. This embodiment of the invention does not make specific limitations.

[0121] In another embodiment of the invention, for further explanation and limitation, the steps further include:

[0122] If the target data passes the probability verification of the coal preparation data, then the trained data decision classification model is retrieved to perform decision classification processing on the target data to obtain the data decision result;

[0123] If the data decision result is an abnormal decision classification, the first fluctuation threshold range of the stable fluctuation state and the second fluctuation threshold range of the offset fluctuation state are adjusted respectively to re-make the data decision.

[0124] To meet the data processing requirements for target data that has passed the probability verification of coal preparation data, a data decision classification model with completed training data is retrieved to perform decision classification processing on the target data. This determines whether steps 101-104 performed on the target data are normal, without clearing the target data. In this case, the data decision classification model is obtained by training a binary tree model based on a data decision sample set. If the data decision result is an abnormal decision classification, it indicates that the first fluctuation threshold range and the second fluctuation threshold range are no longer applicable for filtering the target data when performing steps 101-104. Therefore, the first fluctuation threshold range for stable fluctuation states and the second fluctuation threshold range for offset fluctuation states are adjusted respectively to re-perform data decision-making. The adjustment of the first and second fluctuation threshold ranges can be based on manual adjustment or automatic adjustment by increasing or decreasing the factor; this embodiment of the invention does not impose specific limitations.

[0125] It should be noted that when training the binary tree model based on the data decision sample set, the data decision sample set contains multiple target data and labels marked as abnormal or normal decision categories for training the binary tree model. This embodiment of the invention does not impose specific limitations.

[0126] In this embodiment of the invention, when the intelligent coal preparation plant executes the data decision-making method of this embodiment, firstly, a business theme data warehouse of the intelligent coal preparation plant is established as the data target. A batch collection of Kettle database connections is initiated, selecting the connection type as Hadoop Hive 2 or MS SQL Server, entering the hostname, and then filling in the business theme database name. The data tables of the coal preparation plant's relational database are used as input tables for configuration transformation or job setup, achieving the synchronization or creation of batch collection of coal preparation plant business data. Then, real-time collection from Kafka is performed, transmitting event data streams from producers to consumers. Kafka event streams are organized into topics. Producers select topics to send given events, and consumers select topics to extract events. Kafka topics are divided into partitions. All events for a given topic in one or more partitions are processed. The coal preparation plant's real-time historical database is used as the Kafka producer, and the production time-series database MongoDB is used as the Kafka consumer. Received events are converted into BSON documents and then stored in the database, achieving the synchronization or creation of real-time collection of coal preparation plant production data. Figure 3As shown, an intelligent coal preparation plant business-themed data warehouse and a production time-series database, MongoDB, were built using batch data acquisition with Kettle and real-time data acquisition with Kafka. Through data warehouse OLAP technology, decision-makers can interactively analyze multidimensional data on coal preparation plant business themes from multiple perspectives. This includes rolling up market data to sales analysis to understand sales trends, drilling down product-categorized sales data to view individual product sales data, slicing OLAP cubes to represent specific data sets, and viewing the same sales data from different perspectives (e.g., supplier, date, customer, product, or region). Here, multidimensional data includes raw coal ash content, raw coal moisture content, raw coal calorific value, raw coal quantity, raw coal price, equipment runtime, energy consumption (affecting yield), medium consumption (affecting yield), water consumption (affecting yield), product coal ash content, product coal moisture content, product coal calorific value, product coal output, and product coal price. Finally, decision modeling is performed. Based on the user needs of intelligent coal preparation plants, a decision model with the objective function of maximizing economic benefits is established. Various factors affecting the economic benefits of intelligent coal preparation plants, such as production efficiency, product quality, safety, and production costs, are used as constraints. Python algorithm libraries (TensorFlow, Scikit-Learn, Numpy, Keras, PyTorch, LightGBM, Eli5, SciPy, Theano, Pandas) are used to select corresponding data mining algorithms (classification, estimation, prediction, association, clustering, etc.) to analyze the data and extract valuable information. Based on the data decision objectives (production status analysis, operational status analysis, production indicator prediction, product structure optimization, and economic benefit prediction), the optimal parameters of the model are determined. The decision modeling is evaluated and deployed, such as by executing the data decision method in the embodiments of this invention. The embodiments of this invention do not impose specific limitations.

[0127] This invention provides a data-driven decision-making method for intelligent coal preparation plants. Compared with existing technologies, this invention acquires coal preparation operation data to be decided upon and classifies the data according to its fluctuation state to obtain a first type of coal preparation data and a second type of coal preparation data. The data fluctuation state characterizes the state of the data within different data fluctuation ranges. A first data failure offset is determined based on a first screening weight of the first type of coal preparation data, and a second data failure offset is determined based on a second screening weight of the second type of coal preparation data. If the first data failure offset is less than the second data failure offset, the difference between the first and second data failure offsets is used to select the first type of coal preparation data and the second type of coal preparation data from the second type of coal preparation data. Target data is selected from the coal preparation data. If the first data failure offset is greater than or equal to the second data failure offset, target data is selected from the first and second types of coal preparation data based on the sum of the first and second data failure offsets. The target data is then subjected to coal preparation data probability verification according to the acquisition nodes in the coal preparation operation. If the target data fails the coal preparation data probability verification, the target data is cleared. The first and second types of coal preparation data after clearing the target data are identified as coal preparation data to be monitored. This enables data filtering decisions for coal preparation data, accurately identifies invalid coal preparation data, improves the accuracy of data processing based on coal preparation data, and achieves effective decision-making on coal preparation data.

[0128] Furthermore, as a response to the above Figure 1 The implementation of the method shown in this invention provides a data decision-making device based on an intelligent coal preparation plant, such as... Figure 4 As shown, the device includes:

[0129] The acquisition module 21 is used to acquire coal preparation operation data to be decided, and classify the coal preparation operation data according to the data fluctuation state to obtain a first type of coal preparation data and a second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges.

[0130] The determining module 22 is used to determine the first data failure offset of the first type of coal preparation data based on the first screening weight of the first type of coal preparation data, and to determine the second data failure offset of the second type of coal preparation data based on the second screening weight of the second type of coal preparation data;

[0131] The first screening module 23 is used to select target data from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset if the first data failure offset is less than the second data failure offset.

[0132] The second filtering module 34 is used to filter target data from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset if the first data failure offset is greater than or equal to the second failure offset.

[0133] The clearing module 25 is used to perform coal preparation data probability verification on the target data according to the acquisition nodes in the coal preparation operation, and clear the target data if the target data fails the coal preparation data probability verification, so as to determine the first type of coal preparation data and the second type of coal preparation data after clearing the target data as coal preparation data to be monitored.

[0134] Furthermore, the acquisition module includes:

[0135] The retrieval unit is used to retrieve the equipment operating parameters of the coal preparation equipment and obtain the data fluctuation status that matches the equipment operating parameters. The data fluctuation status includes a stable fluctuation status and an offset fluctuation status.

[0136] The determining unit is used to determine the coal preparation operation data as the first type of coal preparation data if the coal preparation operation data matches the first fluctuation threshold range corresponding to the stable fluctuation state.

[0137] The determining unit is used to determine the coal preparation operation data as the second type of coal preparation data if the coal preparation operation data matches the second fluctuation threshold range corresponding to the offset fluctuation state.

[0138] Wherein, the first fluctuation threshold range is smaller than the second fluctuation threshold range.

[0139] Furthermore, the determining module includes:

[0140] The first extraction unit is used to extract the first screening weight that matches the first type of coal preparation data, and the sum of the weight coefficients of the first screening weight is 1.

[0141] The first calculation unit is used to perform a weight summation operation based on the first screening weight and the first type of coal preparation data to obtain the first data failure offset.

[0142] The second extraction unit is used to extract the second screening weights that match the second type of coal preparation data. The offset weight coefficients in the second screening weights are equal to the sum of the non-offset weight coefficients.

[0143] The second calculation unit is used to perform a weighted summation operation based on the second screening weight and the second type of coal preparation data to obtain the second data failure offset.

[0144] Furthermore, the first screening module includes:

[0145] The first calculation unit is used to calculate the difference between the first data failure offset and the second data failure offset to obtain the failure offset difference value;

[0146] The first determining unit is configured to determine the coal preparation data in the first type of coal preparation data and the second type of coal preparation data that are greater than the failure offset difference as target data; or...

[0147] The first extraction unit is used to extract target data from the first type of coal preparation data and the second type of coal preparation data according to a preset offset difference multiple of the failure offset difference.

[0148] Furthermore, the second screening module includes:

[0149] The second calculation unit is used to calculate the sum of the first data failure offset and the second data failure offset to obtain the total failure offset value;

[0150] The second determining unit is used to determine the coal preparation data in the first type of coal preparation data and the second type of coal preparation data that are greater than the total failure offset value as target data; or...

[0151] The second extraction unit is used to extract target data from the first type of coal preparation data and the second type of coal preparation data according to a preset offset multiple of the total failure offset value.

[0152] Furthermore, the clearing module includes:

[0153] The selected unit is used to select a data acquisition node from the coal preparation operation, wherein the data acquisition node is used to characterize the execution node corresponding to the coal preparation equipment for collecting coal preparation operation data when the coal preparation operation is performed during the coal preparation operation;

[0154] The matching unit is used to match the verification object of the acquisition node and obtain the verification threshold that matches the verification object, so as to determine whether the target data passes the coal preparation data probability verification through the verification threshold. The verification object is used to characterize the data object for the probability verification of the target data. The data object includes at least one of data size, data type, data path, and data proportion.

[0155] Furthermore, the device also includes:

[0156] The retrieval module is used to retrieve a pre-trained data decision classification model to perform decision classification processing on the target data under the condition that the target data passes the coal preparation data probability verification, and to obtain the data decision result. The data decision classification model is obtained by training a binary tree model based on the data decision sample set.

[0157] The adjustment module is used to adjust the first fluctuation threshold range of the stable fluctuation state and the second fluctuation threshold range of the offset fluctuation state respectively if the data decision result is an abnormal decision classification, so as to re-make the data decision.

[0158] This invention provides a data decision-making device based on an intelligent coal preparation plant. Compared with existing technologies, this invention acquires coal preparation operation data to be decided and classifies the data according to its fluctuation state to obtain a first type of coal preparation data and a second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges. A first data failure offset is determined based on a first screening weight of the first type of coal preparation data, and a second data failure offset is determined based on a second screening weight of the second type of coal preparation data. If the first data failure offset is less than the second data failure offset, the difference between the first and second data failure offsets is used to select the first type of coal preparation data and the second type of coal preparation data from the second type of coal preparation data. Target data is selected from the coal preparation data. If the first data failure offset is greater than or equal to the second data failure offset, target data is selected from the first and second types of coal preparation data based on the sum of the first and second data failure offsets. The target data is then subjected to coal preparation data probability verification according to the acquisition nodes in the coal preparation operation. If the target data fails the coal preparation data probability verification, the target data is cleared. The first and second types of coal preparation data after clearing the target data are identified as coal preparation data to be monitored. This enables data filtering decisions for coal preparation data, accurately identifies invalid coal preparation data, improves the accuracy of data processing based on coal preparation data, and achieves effective decision-making on coal preparation data.

[0159] According to one embodiment of the present invention, a storage medium is provided, the storage medium storing at least one executable instruction, the computer-executable instruction being able to execute the data decision-making method based on an intelligent coal preparation plant in any of the above method embodiments.

[0160] Figure 5 The diagram shows a structural schematic of a terminal according to an embodiment of the present invention. The specific implementation of the terminal is not limited by the specific embodiments of the present invention.

[0161] like Figure 5 As shown, the terminal may include: a processor 302, a communications interface 304, a memory 306, and a communications bus 308.

[0162] The processor 302, communication interface 304, and memory 306 communicate with each other via communication bus 308.

[0163] Communication interface 304 is used to communicate with other network elements such as clients or other servers.

[0164] The processor 302 is used to execute program 310, specifically the relevant steps in the above-described data decision-making method embodiment based on intelligent coal preparation plant.

[0165] Specifically, program 310 may include program code that includes computer operation instructions.

[0166] Processor 302 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The terminal may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.

[0167] Memory 306 is used to store program 310. Memory 306 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0168] Specifically, program 310 can be used to cause processor 302 to perform the following operations:

[0169] The coal preparation operation data to be decided is obtained, and the coal preparation operation data is classified according to the data fluctuation state to obtain the first type of coal preparation data and the second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges.

[0170] The first data failure offset of the first type of coal preparation data is determined based on the first screening weight of the first type of coal preparation data, and the second data failure offset of the second type of coal preparation data is determined based on the second screening weight of the second type of coal preparation data.

[0171] If the first data failure offset is less than the second data failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset.

[0172] If the first data failure offset is greater than or equal to the second failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset.

[0173] The target data is subjected to coal preparation data probability verification according to the collection nodes in the coal preparation operation. If the target data fails the coal preparation data probability verification, the target data is cleared, so that the first type of coal preparation data and the second type of coal preparation data after clearing the target data are determined as coal preparation data to be monitored.

[0174] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0175] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A data-driven decision-making method based on an intelligent coal preparation plant, characterized in that, include: The coal preparation operation data to be decided is obtained, and the coal preparation operation data is classified according to the data fluctuation state to obtain the first type of coal preparation data and the second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges. The first data failure offset of the first type of coal preparation data is determined based on the first screening weight of the first type of coal preparation data, and the second data failure offset of the second type of coal preparation data is determined based on the second screening weight of the second type of coal preparation data. If the first data failure offset is less than the second data failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset. If the first data failure offset is greater than or equal to the second data failure offset, then target data is selected from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset. According to the collection nodes in the coal preparation operation, the target data is subjected to coal preparation data probability verification. If the target data fails the coal preparation data probability verification, the target data is cleared, so that the first type of coal preparation data and the second type of coal preparation data after clearing the target data are determined as coal preparation data to be monitored. The classification of the coal preparation operation data according to the data fluctuation state to obtain the first type of coal preparation data and the second type of coal preparation data includes: The equipment operating parameters of the coal preparation equipment are retrieved, and the data fluctuation status matching the equipment operating parameters is obtained. The data fluctuation status includes stable fluctuation status and offset fluctuation status. If the coal preparation operation data matches the first fluctuation threshold range corresponding to the stable fluctuation state, then the coal preparation operation data is determined as the first type of coal preparation data; If the coal preparation operation data matches the second fluctuation threshold range corresponding to the offset fluctuation state, then the coal preparation operation data is determined as the second type of coal preparation data; Wherein, the first fluctuation threshold range is smaller than the second fluctuation threshold range.

2. The method according to claim 1, characterized in that, The determination of the first data failure offset of the first type of coal preparation data based on the first screening weight of the first type of coal preparation data includes: Extract the first screening weight that matches the first type of coal preparation data, where the sum of the weight coefficients of the first screening weight is 1; The first data failure offset is obtained by performing a weighted summation operation based on the first screening weight and the first type of coal preparation data. The determination of the second data failure offset of the second type of coal preparation data based on the second screening weight of the second type of coal preparation data includes: Extract the second screening weights that match the second type of coal preparation data, where the offset weight coefficients in the second screening weights are equal to the sum of the non-offset weight coefficients; The second data failure offset is obtained by performing a weighted summation operation based on the second screening weight and the second type of coal preparation data.

3. The method according to claim 2, characterized in that, The step of filtering target data from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset includes: Calculate the difference between the first data failure offset and the second data failure offset to obtain the failure offset difference value; The coal preparation data from the first type of coal preparation data and the second type of coal preparation data that are greater than the failure offset difference are identified as target data; or... Target data is extracted from the first type of coal preparation data and the second type of coal preparation data according to the preset offset difference multiple of the failure offset difference.

4. The method according to claim 3, characterized in that, The step of filtering target data from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset includes: Calculate the sum of the first data failure offset and the second data failure offset to obtain the total failure offset value; The coal preparation data exceeding the total failure offset value in the first type of coal preparation data and the second type of coal preparation data are identified as target data; or... Target data is extracted from the first type of coal preparation data and the second type of coal preparation data according to a preset offset multiple of the total failure offset.

5. The method according to claim 4, characterized in that, The step of verifying the probability of coal preparation data for the target data according to the acquisition nodes in the coal preparation operation includes: Based on the acquisition node selected from the coal preparation operation, the acquisition node is used to characterize the execution node corresponding to the coal preparation equipment for coal preparation operation data acquisition when the coal preparation operation is performed; The verification object of the acquisition node is matched, and the verification threshold matching the verification object is obtained. The verification threshold is used to determine whether the target data passes the coal preparation data probability verification. The verification object is used to characterize the data object for the probability verification of the target data. The data object includes at least one of the following: data size, data type, data path, and data proportion.

6. The method according to claim 5, characterized in that, The method further includes: Under the condition that the target data passes the probability verification of the coal preparation data, the trained data decision classification model is retrieved to perform decision classification processing on the target data to obtain the data decision result. The data decision classification model is obtained by training a binary tree model based on the data decision sample set. If the data decision result is an abnormal decision classification, the first fluctuation threshold range of the stable fluctuation state and the second fluctuation threshold range of the offset fluctuation state are adjusted respectively to re-make the data decision.

7. A data decision-making device based on an intelligent coal preparation plant, characterized in that, include: The acquisition module is used to acquire coal preparation operation data to be decided, and classify the coal preparation operation data according to the data fluctuation state to obtain a first type of coal preparation data and a second type of coal preparation data. The data fluctuation state is used to characterize the state of the data in different data fluctuation ranges. The determination module is used to determine the first data failure offset of the first type of coal preparation data based on the first screening weight of the first type of coal preparation data, and to determine the second data failure offset of the second type of coal preparation data based on the second screening weight of the second type of coal preparation data. The first filtering module is used to filter target data from the first type of coal preparation data and the second type of coal preparation data based on the difference between the first data failure offset and the second data failure offset if the first data failure offset is less than the second data failure offset. The second filtering module is used to filter target data from the first type of coal preparation data and the second type of coal preparation data based on the sum of the first data failure offset and the second data failure offset if the first data failure offset is greater than or equal to the second data failure offset. The clearing module is used to perform coal preparation data probability verification on the target data according to the acquisition nodes in the coal preparation operation, and clear the target data if the target data fails the coal preparation data probability verification, so as to determine the first type of coal preparation data and the second type of coal preparation data after clearing the target data as coal preparation data to be monitored. The acquisition module includes: The retrieval unit is used to retrieve the equipment operating parameters of the coal preparation equipment and obtain the data fluctuation status that matches the equipment operating parameters. The data fluctuation status includes a stable fluctuation status and an offset fluctuation status. The determining unit is used to determine the coal preparation operation data as the first type of coal preparation data if the coal preparation operation data matches the first fluctuation threshold range corresponding to the stable fluctuation state. The determining unit is used to determine the coal preparation operation data as the second type of coal preparation data if the coal preparation operation data matches the second fluctuation threshold range corresponding to the offset fluctuation state. Wherein, the first fluctuation threshold range is smaller than the second fluctuation threshold range.

8. A storage medium storing at least one executable instruction that causes a processor to perform an operation corresponding to the data decision-making method based on an intelligent coal preparation plant as described in any one of claims 1-6.

9. A terminal, comprising: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the data decision-making method based on an intelligent coal preparation plant as described in any one of claims 1-6.

Images