Decision generation method, device, equipment, medium and program product of coal power system

Abstract

This application relates to a decision generation method, apparatus, equipment, medium, and program product for coal-fired power systems, comprising: acquiring full-chain data of the coal-fired power system; determining the weights of the full-chain data according to a dynamic weighting model; performing weighted fusion processing on the full-chain data based on the weights, and using a preset collaborative analysis algorithm to perform collaborative analysis on the weighted fused full-chain data to obtain target decision data; the preset collaborative analysis algorithm employs at least one of association rule analysis and clustering algorithms; the target decision data satisfies a preset balance objective; and processing the target decision data based on the target decision system to obtain the target decision result. This application calculates the weights corresponding to the data in real time through a dynamic weighting model to adapt to data changes in each link of the coal-fired power system, and mines the correlation between each link through a collaborative algorithm, which can effectively improve the accuracy of data processing, enable the decision result to achieve balance across multiple evaluation indicators, and provide more accurate decision results for the coal-fired power system.

Description

Technical Field

[0001] This application relates to the field of coal-fired power data management technology, and in particular to a decision generation method, apparatus, equipment, medium, and program product for a coal-fired power system. Background Technology

[0002] The coal-fired power industry encompasses the entire chain, including coal mining, transportation, power generation, and transmission, generating massive amounts of heterogeneous data from multiple sources. Existing decision-making solutions related to coal-fired power systems rely on multiple data sources and use fixed weights for data processing, failing to adapt to dynamic changes in coal quality, equipment, and costs. When making decisions and managing coal-fired power systems, the collected multi-source data is often processed manually. However, manual data processing suffers from poor timeliness, is prone to loss, and is susceptible to errors and storage chaos, leading to unreliable decision-making results. Summary of the Invention

[0003] Therefore, it is necessary to provide a method for dynamically adjusting weights and achieving collaborative data management across the entire chain to address the aforementioned technical problems, thereby improving the accuracy of data processing and the scientific nature of decision-making in coal-fired power systems. This method includes decision generation methods, devices, equipment, media, and program products.

[0004] Firstly, this application provides a decision generation method for a coal-fired power system, including:

[0005] Acquire full-chain data from the coal-fired power system;

[0006] The weights of the entire data chain are determined based on a dynamic weighting model;

[0007] The entire chain of data is weighted and fused based on weights, and then a pre-defined collaborative analysis algorithm is used to perform collaborative analysis on the weighted and fused data to obtain target decision data. The pre-defined collaborative analysis algorithm uses at least one of association rule analysis and clustering algorithm. The target decision data satisfies a pre-defined balance objective. The pre-defined balance objective is a balance objective on multiple evaluation indicators. The evaluation indicators include economy and security.

[0008] The target decision system processes target decision data to obtain target decision results; the target decision system is used to generate at least one target decision result among production scheduling plan, equipment maintenance plan and cost control plan.

[0009] In one embodiment, the dynamic weighting model is constructed based on a long short-term memory convolutional neural network; the weights of the entire chain of data are determined according to the dynamic weighting model, including:

[0010] Identify the influencing factors for the entire data chain; the influencing factors include at least one of the link type and coal type;

[0011] The impact factor weight coefficients in the dynamic weight model are adjusted according to the impact factors to obtain the target dynamic weight model that matches the full-chain data.

[0012] The weights of the entire data chain are determined based on the target dynamic weight model.

[0013] In one embodiment, the dynamic weighting model is represented as:

[0014]

[0015] in, For the first The weight of each data point Due to data latency, For hyperparameters, For sensor error rate, These are the weighting coefficients of the impact factors. , , The coefficients are dynamically updated based on a Long Short-Term Memory (LST) convolutional neural network, and the constraint adjustment is as follows: .

[0016] In one embodiment, acquiring full-chain data of the coal-fired power system includes:

[0017] Obtain the decision-making objectives of the target decision-making system; the decision-making objectives include at least one of production scheduling, equipment maintenance, and cost control.

[0018] Obtain the entire chain of data related to the coal-fired power system and the target to be decided based on the target to be decided.

[0019] In one embodiment, weighted fusion processing of the entire chain of data is performed based on weights, including:

[0020] The entire data chain is preprocessed to obtain standardized data in a unified format; the preprocessing includes data cleaning, data transformation and data normalization.

[0021] Standardized data is weighted and fused based on weights.

[0022] In one embodiment, the decision generation method for a coal-fired power system further includes:

[0023] Real-time monitoring system collects data from the entire coal-fired power system.

[0024] When the target parameters of the entire data chain are updated, the dynamic weight model is optimized based on the updated target parameters.

[0025] The optimized dynamic weight model is used to calculate the weights of the entire chain of data.

[0026] Secondly, this application also provides a decision generation device for a coal-fired power system, comprising:

[0027] The acquisition module is used to acquire data across the entire coal-fired power system.

[0028] The determination module is used to determine the weights of the entire chain of data based on the dynamic weight model;

[0029] The fusion module is used to perform weighted fusion processing on the entire chain of data based on weights, and to perform collaborative analysis on the weighted fused data using a preset collaborative analysis algorithm to obtain target decision data. The preset collaborative analysis algorithm employs at least one of association rule analysis and clustering algorithms. The target decision data satisfies a preset balance objective. The preset balance objective is a balance objective across multiple evaluation indicators, including economic efficiency and security.

[0030] The decision module is used to process target decision data based on the target decision system to obtain target decision results.

[0031] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the decision generation method for the coal-fired power system described in the first aspect.

[0032] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the decision generation method for the coal-fired power system described in the first aspect.

[0033] Fifthly, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the decision generation method for a coal-fired power system described in the first aspect.

[0034] In summary, this application proposes a decision generation method, apparatus, equipment, medium, and program product for coal-fired power systems, comprising: acquiring full-chain data of the coal-fired power system; determining the weights of the full-chain data according to a dynamic weighting model; performing weighted fusion processing on the full-chain data based on the weights, and using a preset collaborative analysis algorithm to perform collaborative analysis on the weighted fused full-chain data to obtain target decision data; wherein the preset collaborative analysis algorithm employs at least one of association rule analysis and clustering algorithms; the target decision data satisfies a preset balance objective; and processing the target decision data based on a target decision system to obtain the target decision result. This application calculates the weights corresponding to the data in real time through a dynamic weighting model to adapt to data changes in each link of the coal-fired power system, and mines the correlation between each link through a collaborative algorithm, which can effectively improve data processing accuracy, strengthen data collaboration, and enable the decision result to achieve balance across multiple evaluation indicators, providing more accurate decision results for the coal-fired power system. Attached Figure Description

[0035] Figure 1 This is an application environment diagram of a decision generation method for a coal-fired power system in one embodiment.

[0036] Figure 2 This is a flowchart illustrating a decision generation method for a coal-fired power system in one embodiment;

[0037] Figure 3 This is a flowchart illustrating the steps for determining the weights of the entire data chain in one embodiment;

[0038] Figure 4 This is a flowchart illustrating the steps involved in obtaining full-chain data in one embodiment.

[0039] Figure 5 This is a flowchart illustrating the decision generation method for a coal-fired power system in another embodiment;

[0040] Figure 6 This is a structural block diagram of a decision generation device for a coal-fired power system in one embodiment.

[0041] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0043] In related technologies, the coal-fired power industry chain mainly involves multiple stages such as coal mining, transportation, power generation, and power transmission, each of which generates a large amount of data. However, power plants face significant management challenges regarding the data generated at each stage, primarily including:

[0044] First, the diverse data sources and inconsistent formats across different stages hinder effective data integration and utilization. Second, existing data management methods often employ fixed weighting for data processing, which is ill-suited to the constantly changing data across the entire coal-fired power generation chain, such as fluctuations in the quality and price of incoming / furnace coal, and changes in the status of power generation / transmission equipment. Third, current power plant data export and entry require manual processing, resulting in poor timeliness and a high risk of data loss. Fourth, actual power plant production requires a large amount of real-time and historical data; however, data acquisition errors, data loss, and disorganized data storage make it difficult to obtain accurate and effective information during data analysis, thus affecting judgments and decisions regarding the power plant's production status. Fifth, for power plants burning special types of coal, both incoming coal quality data and production operation data have a significant impact on boiler production stability and safety; operators must make adjustments to boilers burning special coal types based on data analysis.

[0045] Based on the aforementioned data management issues, existing methods for intelligent decision-making based on multi-source heterogeneous data from various stages of the coal-fired power industry chain suffer from problems with decision accuracy and adaptability. They cannot effectively adapt to changes in the production environment, coal quality, and electricity / coal price fluctuations, leading to a decline in the management efficiency and accuracy of the coal-fired power system. To improve the management capabilities of the coal-fired power system and effectively enhance the accuracy and adaptability of intelligent decision-making, this embodiment provides a decision generation method for the coal-fired power system. Through the application of a dynamic weighting model and collaborative analysis algorithms, it effectively solves the problems of ineffective management of multi-source data, inability to adapt to the continuous changes in data across various stages of the coal-fired power industry chain, and low efficiency of manual processing in related technologies.

[0046] The decision generation method for coal-fired power systems provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or located in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0047] In one embodiment, such as Figure 2 As shown, a decision generation method for coal-fired power systems is provided, which is then applied to... Figure 1 Taking the server in the example, the following steps are included:

[0048] Step 202: Obtain full-chain data of the coal-fired power system.

[0049] In this embodiment, the entire chain data of the coal-fired power system can also be referred to as multi-source data generated by multiple links in the coal-fired power industry chain. Specifically, the coal-fired power industry chain mainly includes coal mining, coal transportation, power plant generation, and power transmission. Data generated from the coal mining environment includes coal production output and coal quality data. Data generated from the coal transportation link includes transportation methods, transportation costs, transportation distances, and transportation times. Data generated from the power plant generation link includes operating parameters of main and auxiliary equipment and equipment status parameters, such as the power plant's distributed control system (DCS). Data from the power transmission link includes current, voltage, and losses during the transmission process. It should be noted that coal production output can be collected through weighing equipment in the coal mining link, such as electronic belt scales. Coal quality data can be collected through coal quality testing equipment in the coal mining link, such as online laser coal quality analyzers.

[0050] It should be noted that the method for acquiring data across the entire chain of a coal-fired power system can be adaptively designed according to the needs of the actual application scenario.

[0051] In one embodiment, after collecting data from the entire coal-fired power system, the collected data needs to be preprocessed. Preprocessing includes data cleaning, data transformation, and data normalization. Specifically, data cleaning removes duplicate, erroneous, and missing data. For example, it removes data from periods of sensor network outage (outage time > 30 seconds). Data transformation converts data from different formats into a unified format. Data normalization maps all data to a specific interval, typically using linear normalization. The expression for linear normalization is:

[0052]

[0053] in, For the collected data, The alarm lower limit value set for each parameter, Alarm upper limit values ​​for each parameter setting.

[0054] Step 204: Determine the weights of the entire chain of data based on the dynamic weight model.

[0055] In this embodiment, the dynamic weight model is constructed based on a Long Short-Term Memory Convolutional Neural Network (CNN-LSTM) model, and the function representation of the dynamic weight model is as follows:

[0056]

[0057] in, For the first The weight of each data point Due to data latency, For hyperparameters, For sensor error rate, These are the weighting coefficients of the impact factors. , , The coefficients are dynamically updated based on a Long Short-Term Memory (LST) convolutional neural network, and the constraint adjustment is as follows: Hyperparameters It can be fixed at 0.1.

[0058] In this embodiment, the dynamic weight model implements the weight coefficients using a CNN-LSTM hybrid neural network. , , Dynamic updates, combined with physical constraints This leads to the formation of an adaptive control strategy. Specifically, the core architecture of the CNN-LSTM model includes CNN layers, LSTM layers, fully connected layers, and a softmax layer. The CNN layers are used to extract spatial features from the input data. Spatial features include time-series patterns of sensor data and the spatial distribution of equipment states. For example, convolutional feature extraction can be used to extract local hot spots or temperature gradient anomalies from a boiler temperature field. The LSTM layers are used to model temporal dependencies and handle data latency. (second-level) and sensor error rate The dynamic changes are controlled by gating mechanisms (input gate, forget gate, output gate) to prevent gradient vanishing / exploding. Fully connected layers and softmax layers are used to map the output of the LSTM layer to... , , And force satisfaction through the Softmax activation function. The constraints. It should be noted that a Softmax constraint is applied to the output layer of the CNN-LSTM to ensure... For example, if the model predicts... , ,but It will automatically adjust to 0.3.

[0059] In this embodiment, the input features of the dynamic weighting model include time dimension features, error dimension features, and influence factor dimension features. The time dimension feature can be data latency. And / or historical state sequences, such as pressure / temperature changes over the past hour. Error dimension features could include sensor error rate. For example, the percentage deviation between pressure sensor measurements and standard values. Influencing factor dimensional characteristics. It can adapt to different aspects of the coal-fired power generation process or the composition of the coal. For example, in the coal mining process, In the coal transportation process, In the power generation stage of a power plant, In the power transmission process, If the coal is high-alkali coal, If the coal is conventional coal, It is important to know that... The actual value can be configured according to the needs of the actual application scenario.

[0060] It should be noted that, based on the functional expression of the above dynamic weight model, it is clear that... It belongs to the exponentially decaying term. This indicates that the impact of data latency on weights decays exponentially over time. Belongs to the logical function term, when When the value is increased, the weight of this item is reduced to suppress the influence of high-error sensors. As an influencing factor weighting item, the basic weight can be adjusted according to the type of process or coal type to strengthen the control priority of key processes.

[0061] Based on the above steps, by comprehensively considering factors such as data latency, sensor error, process type, and coal type, weights can be calculated in real time. Dynamic weights can be used to adapt to data changes in each link of the entire coal-fired power generation chain, greatly improving the processing accuracy of coal-fired power generation data.

[0062] Step 206: Perform weighted fusion processing on the entire chain of data based on weights, and use a preset collaborative analysis algorithm to perform collaborative analysis on the weighted fusion of the entire chain of data to obtain target decision data.

[0063] In this embodiment, weighted fusion processing obtains a comprehensive data representation by linearly combining data from each link of the coal-fired power generation chain according to their weights. Weighted fusion highlights the influence of important data, suppresses interference from low-quality data, and improves the comprehensive representation capability of the data. Important data includes data with high real-time performance, low error, and key links. Specifically, the function representation of the weighted fused data is as follows:

[0064]

[0065] in, For the collected data items, This refers to the weight corresponding to each data item. Specifically, For the first The feature vector of a data source, such as sensor readings or economic indicators. Must meet .

[0066] In this embodiment, collaborative algorithms (such as association rule mining and cluster analysis) are used to perform collaborative analysis and mining on the weighted and fused data. Specifically, the preset collaborative analysis algorithm adopts at least one of association rule analysis and clustering algorithms. Specifically, the association rule analysis algorithm can be the Apriori algorithm, used to mine implicit association rules in the entire chain of data to assist decision-making. The clustering algorithm can be K-Means or DBSCAN clustering algorithms, used to classify the weighted and fused data, identify abnormal patterns, or optimize grouping.

[0067] Specifically, the steps of the association rule analysis algorithm include: First, statistically analyzing the data features of high-frequency co-occurrence to mine frequent itemsets. Second, calculating the support and confidence of each frequent itemset to filter out strong association rules. The steps of the clustering algorithm include: First, setting key indicators, such as limited economic cost or safety risk level, to classify power generation conditions into types such as high-efficiency and low-risk or low-efficiency and high-risk, to guide operation and maintenance strategies. Second, running the clustering algorithm to obtain different clusters. In practical applications, the K-Means clustering algorithm requires pre-determining the number of clusters k and is suitable for spherically distributed data. The DBSCAN clustering algorithm is based on density clustering and can automatically identify outliers.

[0068] In practical applications, if there is a clear objective for association rule mining, association rule analysis algorithms can be used directly. If a combination of classification and association analysis is required, clustering algorithms can be used to divide the data into subsets, and then association rule analysis algorithms can be used to mine association rules for each subset to refine the decision rules.

[0069] In this embodiment, the target decision data satisfies a preset balance objective. The preset balance objective is a balance objective across multiple evaluation indicators. These evaluation indicators include economy and safety.

[0070] Specifically, economic evaluation indicators include fuel costs, operation and maintenance costs, and electricity price revenue. Safety evaluation indicators include failure rate, number of safety threshold violations, and emergency response time. In practical applications, the preset balancing objective can be to ensure that the decision data simultaneously satisfies the following within a multi-objective optimization framework:

[0071]

[0072] in, The weighting coefficients for economic indicators, The weighting coefficients for safety indicators As an economic indicator, For safety indicators.

[0073] It should be noted that the preset balance target can be to meet the lowest safety risk while minimizing coal / transportation / power generation costs. The preset balance target can also be set to any target that meets the power generation needs of the coal-fired power system, depending on the actual application scenario. This embodiment does not limit the actual target of the preset collaborative analysis algorithm for the coal-fired power system.

[0074] Based on the above steps, by weighted fusion processing and collaborative analysis of the data, the scientific nature, accuracy, and effectiveness of the decision-making results can be effectively improved, ensuring that the target decision-making results can provide reliable suggestions to managers in all aspects of coal power.

[0075] Step 208: Process the target decision data based on the target decision system to obtain the target decision results.

[0076] In this embodiment, the target decision system integrates an artificial intelligence model that combines data-driven and machine learning models, enabling it to automatically process target decision data and obtain the final target decision result. The target decision system is used to generate at least one target decision result among production scheduling plans, equipment maintenance plans, and cost control plans.

[0077] Specifically, the decision-making system primarily focuses on production scheduling, equipment maintenance, and cost control for enterprises. For example, in production scheduling, the system can rationally arrange coal mining and transportation plans based on coal production output, transportation conditions, and power generation demand. Regarding equipment maintenance, the system can predict equipment failure risks and formulate maintenance plans based on the operating and status parameters of power generation equipment. In cost control, the system can identify key cost control points and develop cost control strategies by analyzing data such as coal mining costs, transportation costs, power generation costs, and transmission losses.

[0078] In summary, this embodiment provides a decision-making generation method for coal-fired power systems. By deploying a dynamic weighting model, it enables efficient collaborative management of data across the entire coal-fired power supply chain, improving the accuracy of data processing and the scientific nature of decision-making. Employing collaborative algorithms to process data allows for the uncovering of potential correlations between data from different stages, achieving efficient collaborative data management and providing more comprehensive and in-depth information for enterprise decision-making. This provides reliable data support for coal-fired power enterprises' decision-making, helping them make more scientific and rational decisions based on actual conditions, and optimizing production scheduling and resource allocation.

[0079] In one specific embodiment, this embodiment provides a detailed description of the coal mining, coal transportation, and power plant generation processes.

[0080] Coal mining is the initial stage of the coal-fired power industry chain, and its core task is to safely and efficiently extract underground coal resources. In this embodiment, key data such as geological environment data, equipment operation data, and production management data can be collected by deploying sensor networks, intelligent devices, and monitoring systems in the coal mining stage. Geological environment data includes geological exploration data and underground environment data. Geological exploration data, such as coal seam thickness, geological structure, and gas content, can be used to guide mining planning. Underground environment data, such as gas concentration, carbon monoxide concentration, temperature, humidity, and wind speed, can be used to ensure safe production. Equipment operation data includes the operation data of mining equipment, transportation equipment, and auxiliary equipment. Mining equipment operation data includes the operating status, working parameters (such as current, voltage, temperature, and vibration), and location information of equipment such as coal mining machines and tunneling machines. Transportation equipment operation data includes the operating speed, load, and fault alarm data of underground belt conveyors and mine cars. Auxiliary equipment operation data includes the start / stop status, operating efficiency, and energy consumption data of equipment such as ventilation fans, drainage pumps, and air compressors. Production management data includes output data, personnel location data, and video surveillance data. Output data includes real-time output, cumulative output, and shift output. Personnel location data includes the real-time location and activity trajectory of underground personnel, used for safety management and emergency dispatch. Video surveillance data includes real-time video streams of critical work areas, used for remote monitoring and safety supervision.

[0081] The coal transportation segment is responsible for efficiently and cleanly transporting mined coal from the mining area to the power plant. In practical applications, data collection in the coal transportation segment is primarily achieved through monitoring and statistical analysis of logistics and transportation vehicles. Data in the coal transportation segment includes logistics and transportation data as well as environmental energy consumption data. Specifically, logistics and transportation data includes vehicle / ship location data, transportation status data, and warehousing data. Environmental energy consumption data includes transportation energy consumption and environmental impact data.

[0082] The power generation stage of a power plant is the core process of converting the chemical energy of coal into electrical energy. A thermal power plant is a highly complex system with extremely dense internal data, typically divided into three main systems: the combustion system, the steam-water system, and the electrical system. This embodiment uses sensor devices and controllers installed within the power plant's DCS system to collect the operating and status parameters of equipment in each system.

[0083] In one embodiment, such as Figure 3 As shown, the dynamic weighting model is constructed based on a long short-term memory convolutional neural network. The weights of the entire data chain are determined according to the dynamic weighting model, including:

[0084] Step 301: Determine the influencing factors for the entire chain of data. Influencing factors include at least one of the following: stage type and coal type.

[0085] Step 302: Adjust the influence factor weight coefficients in the dynamic weight model according to the influence factors to obtain the target dynamic weight model that matches the whole chain data.

[0086] Step 303: Determine the weights of the entire chain of data based on the target dynamic weight model.

[0087] In this embodiment, determining the weights of the entire data chain requires first determining the influencing factors of the current entire data chain. After determining the influencing factors of the current entire data chain, the weight coefficients of the influencing factors are determined based on these factors. For example, if the current entire data chain involves the coal transportation process and the coal is conventional coal, then the weight coefficients of the influencing factors should be set to... If the current end-to-end data represents the power plant generation stage, then the weighting coefficient of the influencing factors should be set to... If the current coal is high-alkali coal, then the weighting coefficient of the influencing factor should be set to... .

[0088] After determining the weight coefficients of the influencing factors, the values ​​of each coefficient and hyperparameter are adaptively adjusted using the characteristics of CNN-LSTM to update the target dynamic weight model. Finally, the target dynamic weight model is used to determine the weights of the entire data chain. It should be noted that the specific implementation method for determining the weights of the entire data chain based on the target dynamic weight model can be found in the previous embodiments, and will not be repeated here.

[0089] In one embodiment, such as Figure 4 As shown, the entire chain of data for the coal-fired power system is obtained, including:

[0090] Step 401: Obtain the decision-making objectives of the target decision-making system. The decision-making objectives include at least one of production scheduling, equipment maintenance, and cost control.

[0091] Step 402: Obtain the full-chain data of the coal-fired power system related to the target to be decided based on the target to be decided.

[0092] In this embodiment, during the process of acquiring full-chain data of the coal-fired power system, the full-chain data to be collected can be determined based on the decision-making objectives of the management system or decision-making system, namely at least one of production scheduling, equipment maintenance and cost control.

[0093] For example, if the decision-making objective is cost control, then only cost-related data from each stage of the entire coal-fired power system chain needs to be collected. More specifically, if the decision-making objective is to optimize coal costs, then only coal cost-related data from each stage of the entire coal-fired power system chain needs to be collected, without needing to collect electricity price cost-related data.

[0094] It should be noted that in practical application scenarios, the acquisition of full-chain data can also refer to the settings of the preset balance target, which will not be listed here.

[0095] Based on the above steps, the accuracy of data acquisition across the entire chain can be optimized, and the need to call up massive amounts of data in every decision-making process can be avoided, which can effectively save computing resources, increase decision-making speed, and reduce data processing difficulty.

[0096] In one embodiment, weighted fusion processing of the entire chain of data is performed based on weights, including:

[0097] The entire data chain is preprocessed to obtain standardized data in a unified format. Preprocessing includes data cleaning, data transformation, and data normalization. The standardized data is then weighted and fused based on its weights.

[0098] In this embodiment, the raw data of each link in the entire coal-fired power system obtained in step 202 are usually multi-source, heterogeneous and noisy, and must be standardized and preprocessed before they can be used for subsequent analysis and modeling.

[0099] Specifically, data cleaning includes denoising, outlier removal, and missing value imputation. Denoising can use filtering algorithms (such as median filtering and Gaussian filtering) to remove random noise from sensor signals. Outlier removal can use statistical methods or machine learning algorithms (such as Isolation Forest) to identify and process abnormal data points caused by equipment failure or transmission errors. Missing value imputation can fill in missing data caused by network interruptions or other reasons using interpolation methods (such as linear interpolation and multiple interpolation) or models based on historical data. It should be noted that the specific data cleaning methods can be determined according to the needs of the actual application scenario.

[0100] Data conversion includes format unification and data integration. Format unification converts equipment data from different manufacturers and using different protocols into a unified format and encoding standard, such as a unified timestamp format and data units. Data integration links and integrates data from different sources (such as DCS system data, Safety Instrumented System (SIS) data, and management information system data) to form a complete data view.

[0101] Data normalization is used to scale data of different physical meanings and orders of magnitude to a specific range, such as scaling temperature (°C), pressure (MPa), and rotational speed (r / min) to between 0 and 1, in order to eliminate the influence of dimensional differences on the data analysis model and ensure that each feature has equal importance in model training.

[0102] In one embodiment, such as Figure 5 As shown, the decision generation method for coal-fired power systems also includes:

[0103] Step 501: Collect real-time data from the entire coal-fired power system through a real-time monitoring system.

[0104] Step 502: With the target parameters of the entire chain of data updated, optimize the dynamic weight model based on the updated target parameters.

[0105] Step 503: Apply the optimized dynamic weight model to calculate the weights of the entire chain of data.

[0106] In this embodiment, the real-time monitoring system includes sensors and monitoring devices installed at each stage. This embodiment configures the real-time monitoring system to acquire data from the entire chain in real time, compares the target parameters of the entire chain data with preset thresholds, and when abnormal changes in the data are detected, promptly feeds the updated target parameters back to the dynamic weight model for data weight optimization.

[0107] In this embodiment, the target parameters include electricity price information, coal price information, coal composition information, and transportation cost information. It should be noted that the target parameters are related to the weight coefficients of the influencing factors in the dynamic weight model of the actual application scenario; the more influencing factors, the more target parameters there are.

[0108] It should be noted that the actual implementation of step 503 can refer to the specific implementation of step 204 in the foregoing embodiments, and will not be repeated here.

[0109] In summary, this embodiment provides a decision generation method for coal-fired power systems. It collects multi-source data from coal mining, transportation, power generation, and transmission, performs standardized preprocessing, constructs a dynamic weight model based on a long short-time memory convolutional neural network, and calculates weights in real time by considering data latency, error rate, and stage type. After weighted fusion of the data, it analyzes the data using a collaborative algorithm and outputs decision data to the enterprise decision-making system, generating production scheduling, equipment maintenance, and cost control results. This method can dynamically adapt to data changes, improving data collaboration and the scientific nature of decision-making. It is suitable for coal-fired boilers burning high-alkali coal and effectively solves problems such as traditional fixed weights, difficulty in data integration, and decision lag. Furthermore, by monitoring the entire coal-fired power system's data in real time, it can promptly detect data anomalies and changes in operating status, and dynamically adjust data weights, enabling the decision generation method to adapt to the constantly changing realities throughout the entire coal-fired power chain.

[0110] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0111] Based on the same inventive concept, this application also provides a decision generation device for a coal-fired power system to implement the decision generation method for the coal-fired power system described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the decision generation device for a coal-fired power system provided below can be found in the limitations of the decision generation method for coal-fired power systems described above, and will not be repeated here.

[0112] In one embodiment, such as Figure 6 As shown, a decision generation device 600 for a coal-fired power system is provided, comprising: an acquisition module 610, a determination module 620, a fusion module 630, and a decision module 640, wherein:

[0113] The acquisition module 610 is used to acquire data from the entire chain of the coal-fired power system;

[0114] The determination module 620 is used to determine the weights of the entire chain of data based on the dynamic weight model;

[0115] The fusion module 630 is used to perform weighted fusion processing on the full-chain data based on the weights, and to perform collaborative analysis on the weighted fused full-chain data using a preset collaborative analysis algorithm to obtain target decision data; wherein, the preset collaborative analysis algorithm adopts at least one of association rule analysis and clustering algorithm; the target decision data satisfies a preset balance target; the preset balance target is a balance target on multiple evaluation indicators; the evaluation indicators include economy and security;

[0116] The decision module 640 is used to process the target decision data based on the target decision system to obtain the target decision result; wherein, the target decision system is used to generate at least one target decision result among the production scheduling plan, equipment maintenance plan and cost control plan.

[0117] In one embodiment, the determining module 620 is further configured to determine the influence factors of the full-chain data; the influence factors include at least one of link type and coal type; adjust the influence factor weight coefficients in the dynamic weight model according to the influence factors to obtain a target dynamic weight model that matches the full-chain data; and determine the weight of the full-chain data according to the target dynamic weight model.

[0118] In one embodiment, the acquisition module 610 is further configured to acquire the target to be decided by the target decision system; the target to be decided includes at least one of production scheduling, equipment maintenance and cost control; and acquire full-chain data of the coal-fired power system related to the target to be decided based on the target to be decided.

[0119] In one embodiment, the fusion module 630 is further configured to preprocess the entire chain of data to obtain standardized data in a unified format; wherein the preprocessing includes data cleaning, data transformation and data normalization; and to perform weighted fusion processing on the standardized data based on the weights.

[0120] In one embodiment, the determining module 620 is further configured to collect the entire chain data of the coal-fired power system in real time through a real-time monitoring system; optimize the dynamic weight model according to the updated target parameters when the target parameters of the entire chain data are updated; and calculate the weight of the entire chain data by applying the optimized dynamic weight model.

[0121] In summary, this embodiment provides a decision generation device for a coal-fired power system. By dynamically adjusting data weights, it fully considers the real-time nature, accuracy, and importance of data, enabling the data processing results to more accurately reflect the actual operation of the entire coal-fired power chain. Employing collaborative algorithms to process data allows for the discovery of potential correlations between data from different stages, achieving efficient collaborative data management and providing more comprehensive and in-depth information for enterprise decision-making. It provides reliable data support for coal-fired power enterprises' decision-making, helping them make more scientific and rational decisions based on actual conditions, optimizing production scheduling and resource allocation, and improving the enterprise's economic efficiency and competitiveness. The real-time monitoring system can promptly detect data anomalies and changes in operating status, and dynamically adjust data weights, enabling the data management method to adapt to the constantly changing actual conditions throughout the entire coal-fired power chain.

[0122] Each module in the decision generation device of the aforementioned coal-fired power system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0123] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a decision-making method for a coal-fired power system. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0124] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0125] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0126] Acquire full-chain data from the coal-fired power system;

[0127] The weights of the entire data chain are determined based on a dynamic weighting model;

[0128] The entire chain of data is weighted and fused based on weights, and then a pre-defined collaborative analysis algorithm is used to perform collaborative analysis on the weighted and fused data to obtain target decision data. The pre-defined collaborative analysis algorithm uses at least one of association rule analysis and clustering algorithm. The target decision data satisfies a pre-defined balance objective. The pre-defined balance objective is a balance objective on multiple evaluation indicators. The evaluation indicators include economy and security.

[0129] The target decision system processes target decision data to obtain target decision results; the target decision system is used to generate at least one target decision result among production scheduling plan, equipment maintenance plan and cost control plan.

[0130] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0131] Acquire full-chain data from the coal-fired power system;

[0132] The weights of the entire data chain are determined based on a dynamic weighting model;

[0133] The entire chain of data is weighted and fused based on weights, and then a pre-defined collaborative analysis algorithm is used to perform collaborative analysis on the weighted and fused data to obtain target decision data. The pre-defined collaborative analysis algorithm uses at least one of association rule analysis and clustering algorithm. The target decision data satisfies a pre-defined balance objective. The pre-defined balance objective is a balance objective on multiple evaluation indicators. The evaluation indicators include economy and security.

[0134] The target decision system processes target decision data to obtain target decision results; the target decision system is used to generate at least one target decision result among production scheduling plan, equipment maintenance plan and cost control plan.

[0135] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0136] Acquire full-chain data from the coal-fired power system;

[0137] The weights of the entire data chain are determined based on a dynamic weighting model;

[0138] The entire chain of data is weighted and fused based on weights, and then a pre-defined collaborative analysis algorithm is used to perform collaborative analysis on the weighted and fused data to obtain target decision data. The pre-defined collaborative analysis algorithm uses at least one of association rule analysis and clustering algorithm. The target decision data satisfies a pre-defined balance objective. The pre-defined balance objective is a balance objective on multiple evaluation indicators. The evaluation indicators include economy and security.

[0139] The target decision system processes target decision data to obtain target decision results; the target decision system is used to generate at least one target decision result among production scheduling plan, equipment maintenance plan and cost control plan.

[0140] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0141] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0142] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0143] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A decision generation method for a coal-fired power system, characterized in that, include: Acquire full-chain data from the coal-fired power system; The weights of the entire chain of data are determined based on a dynamic weighting model; The entire chain of data is weighted and fused based on the weights, and a preset collaborative analysis algorithm is used to perform collaborative analysis on the weighted and fused data to obtain target decision data. The preset collaborative analysis algorithm employs at least one of association rule analysis and clustering algorithms. The target decision data satisfies a preset balance objective. The preset balance objective is a balance objective across multiple evaluation indicators, including economic efficiency and security. The target decision system processes the target decision data to obtain target decision results; wherein, the target decision system is used to generate at least one target decision result among production scheduling plan, equipment maintenance plan and cost control plan.

2. The method according to claim 1, characterized in that, The dynamic weighting model is constructed based on a long short-term memory convolutional neural network; determining the weights of the entire chain of data according to the dynamic weighting model includes: Determine the influencing factors of the entire chain data; the influencing factors include at least one of the link type and coal type; The influence factor weight coefficients in the dynamic weight model are adjusted according to the influence factors to obtain a target dynamic weight model that matches the full-chain data. The weights of the entire chain of data are determined based on the target dynamic weight model.

3. The method according to claim 2, characterized in that, The dynamic weight model is expressed as follows: in, For the first The weight of each data point Due to data delay, For hyperparameters, For sensor error rate, These are the weighting coefficients of the impact factors. , , The coefficients are dynamically updated based on a Long Short-Term Memory (LST) convolutional neural network, and the constraint adjustment is as follows: .

4. The method according to claim 1, characterized in that, The acquisition of full-chain data from the coal-fired power system includes: Obtain the decision-making target of the target decision system; the decision-making target includes at least one of production scheduling, equipment maintenance and cost control. Based on the target to be decided, obtain the full-chain data of the coal-fired power system related to the target to be decided.

5. The method according to claim 1, characterized in that, The weighted fusion processing of the entire chain of data based on the weights includes: The entire data chain is preprocessed to obtain standardized data in a unified format; wherein, the preprocessing includes data cleaning, data transformation and data normalization; The standardized data is then weighted and fused based on the weights.

6. The method according to claim 1, characterized in that, The method further includes: The entire chain of data of the coal-fired power system is collected in real time through a real-time monitoring system; When the target parameters of the entire chain of data are updated, the dynamic weight model is optimized based on the updated target parameters; The optimized dynamic weight model is used to calculate the weights of the entire chain of data.

7. A decision generation device for a coal-fired power system, characterized in that, The device includes: The acquisition module is used to acquire data across the entire coal-fired power system. The determination module is used to determine the weights of the entire chain of data based on the dynamic weight model; The fusion module is used to perform weighted fusion processing on the entire chain of data based on the weights, and to perform collaborative analysis on the weighted fused entire chain of data using a preset collaborative analysis algorithm to obtain target decision data; wherein, the preset collaborative analysis algorithm adopts at least one of association rule analysis and clustering algorithm; the target decision data satisfies a preset balance target; the preset balance target is a balance target on multiple evaluation indicators; the evaluation indicators include economy and security; The decision module is used to process the target decision data based on the target decision system to obtain the target decision result.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the decision generation method for the coal-fired power system as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the decision generation method for the coal-fired power system as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the decision generation method for the coal-fired power system as described in any one of claims 1 to 6.

Images