Data processing system suitable for fuel whole-process management of thermal power plant

Abstract

The invention discloses a data processing system suitable for thermal power plant fuel whole process management, and belongs to the technical field of electric power industry automation. The system adopts a five-layer architecture and comprises a data acquisition layer used for acquiring real-time data of each link of a fuel whole process through a plurality of data sources; the data storage layer adopts a mixed storage architecture comprising a real-time library, a business library, a history library and a data lake for classified storage; the data processing layer is used for cleaning, integrating, calculating and contrastively analyzing the stored data and constructing a digital twinborn model; the application service layer provides monitoring, optimization, tracing and early warning services; and the display layer is used for data visual display and man-machine interaction. According to the method, the whole-process data chain and the dynamic optimization model are constructed, so that refined and intelligent management of the whole life cycle from fuel purchase to ash disposal is realized, and the fuel management efficiency, economy and safety of the thermal power plant are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of thermal power generation technology, and more specifically to a data processing system applicable to the entire process management of fuel in thermal power plants. Background Technology

[0002] Fuel management in thermal power plants is a core element in ensuring the safe, stable, and economical operation of generating units, involving multiple complex processes such as procurement, transportation, acceptance, storage, blending, combustion, and ash disposal.

[0003] Currently, most thermal power plants still use traditional or rudimentary fuel management systems, which suffer from the following technical shortcomings: First, data from each stage is scattered across independent subsystems (such as procurement, weighing, coal inventory, and testing), creating severe data silos. This lack of unified standards and effective integration makes it difficult to achieve end-to-end data connectivity and traceability. Second, the system's data processing capabilities are lagging, with poor real-time performance, making it impossible to dynamically monitor and provide timely warnings for key parameters such as fuel consumption and coal quality changes. Summary of the Invention

[0004] Therefore, the present invention provides a data processing system applicable to the entire process management of fuel in thermal power plants, in order to solve the problems in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A data processing system applicable to the entire fuel management process of thermal power plants includes a data acquisition layer, a data storage layer, a data processing layer, an application service layer, and a presentation layer;

[0007] The data acquisition layer collects raw data in real time and accurately from each key node of the entire fuel process; after performing preliminary edge preprocessing on the raw data, the data acquisition layer classifies it, timestamps it, and continuously injects it into the corresponding database in the data storage layer.

[0008] The data storage layer provides storage and management for the raw data;

[0009] The data processing layer processes and refines the stored raw data, transforming it into valuable information and knowledge.

[0010] The data processing layer actively pulls data from the data storage layer; the data processing layer first cleans and integrates the data, processing the messy original data into a high-quality, interconnected end-to-end data chain; then it performs calculations and analysis, and outputs the processed valuable information; the data processing layer stores the valuable information back to the data storage layer, and simultaneously pushes it to the application service layer;

[0011] The application service layer encapsulates the valuable information generated by the data processing layer into specific, reusable business applications and services.

[0012] The presentation layer provides an interface for human-computer interaction, presenting complex data and analysis results to users with different roles in an intuitive and easy-to-understand form;

[0013] The presentation layer transmits user instructions to the decision module in the data processing layer; the decision module generates specific control instructions and sends them to the underlying control system connected to the data acquisition layer through the system interface.

[0014] Furthermore, the valuable information includes cleaned and integrated standard data, calculated indicators, analytical conclusions, early warning signals, and digital twin simulation results.

[0015] Furthermore, once a new control command is executed, the data acquisition layer will continuously monitor new fuel consumption, boiler efficiency, and environmental emission data; the system will compare and analyze the new real-time data with the expected optimization results to verify the correctness of the decision.

[0016] Furthermore: the data acquisition layer includes a procurement management system, a weighing system, a coal inventory system, a boiler control system, a testing system, and an ash and slag treatment system;

[0017] Procurement Management System: Collects data on procurement plan number, planned coal type, planned procurement quantity, planned procurement time, and supplier information; supports multiple data interface protocols.

[0018] Weighing system: integrates metering equipment to collect real-time information on vehicle number, arrival time, coal type, and coal weight of coal entering the plant;

[0019] Coal inventory system: It adopts laser scanning and 3D modeling technology to dynamically collect spatial distribution data of various coal types, and supports both automatic and manual coal inventory modes;

[0020] Boiler control system: Real-time monitoring of operating parameters of each type of coal per unit time, and interface with DCS system through standard protocol;

[0021] Laboratory system: Collects key quality indicators of incoming coal through automated testing equipment, and supports automatic data entry and intelligent review of test data;

[0022] Ash and slag treatment system: Accurately collects environmentally relevant data on ash and slag after combustion by deploying weighing sensors and flow metering equipment.

[0023] Furthermore, the data storage layer adopts an innovative hybrid storage architecture, including a real-time database, a business database, a historical database, and a data lake;

[0024] Real-time database: It adopts the core technology of time series database, which is specially designed to handle high-frequency and high-concurrency sensor data. It supports millisecond-level data writing and query response, and has data compression and caching optimization functions.

[0025] Business database: Based on a relational database architecture, it adopts a distributed transaction processing mechanism to provide secure and reliable transactional storage for the core business data of fuel management;

[0026] Historical database: It adopts distributed column storage technology and data compression algorithm to support efficient storage and fast query of petabyte-level historical data, and provides data archiving and lifecycle management functions;

[0027] Data lake: It adopts an open storage format and supports the raw storage of structured, semi-structured and unstructured data.

[0028] Furthermore: the data processing layer includes a data cleaning module, a data integration module, a data calculation module, a comparative analysis module, a digital twin module, and a decision-making module;

[0029] Data cleaning module: It adopts an anomaly detection algorithm based on deep learning and integrates multiple data repair algorithms, including interpolation methods based on time series prediction, mean imputation methods based on association rules, and collaborative filtering methods that consider spatial correlation, and establishes a multi-dimensional data quality assessment system.

[0030] Data integration module: Construct a unified data model with "fuel unit" as the basic management unit and "coal type" and "time" as the core correlation dimensions. Use graph database technology to build a fuel flow relationship network to achieve deep data integration across systems.

[0031] Furthermore: The data calculation module, based on a professional fuel management algorithm library, calculates key performance indicators in real time, including the qualified rate of incoming coal, coal inventory turnover rate, coal consumption per unit of power generation, and ash and slag output rate, and supports the customization of indicators and flexible adjustment of calculation rules;

[0032] The pass rate of incoming coal is a core indicator for measuring the compliance of purchased coal quality standards, and its calculation formula is as follows:

[0033] ;

[0034] Where Rqual is the qualified rate of incoming coal; Q pass,i Q represents the quantity of qualified coal in the i-th batch; total,j Let n be the total amount of coal delivered to the plant in the j-th batch; n and m be the number of qualified batches and the total number of batches, respectively.

[0035] Coal inventory turnover rate is a dynamic indicator reflecting the efficiency of inventory utilization. Its calculation formula is as follows:

[0036] ;

[0037] Among them, T turn For inventory turnover days; C avg This represents the average daily coal consumption; I begin I end These are the beginning and ending inventory levels, respectively.

[0038] Coal consumption per unit of electricity generated is a key performance indicator characterizing the economic efficiency of coal combustion, and its calculation formula is as follows:

[0039] ;

[0040] Among them, C unit B represents the standard coal consumption for power generation; E represents the total coal consumption during the statistical period; and E represents the power generation during the same period.

[0041] Ash and slag yield can be used to assess the efficiency of post-combustion by-product generation. Its calculation formula is as follows:

[0042] ;

[0043] Among them, Y ash W represents the overall yield of ash and slag. slag W represents the amount of slag produced. flyash B represents the amount of fly ash generated; B represents the total amount of coal fed into the furnace.

[0044] Furthermore: The comparative analysis module provides multiple analysis modes, is specifically designed with a co-firing effect analysis function, and supports multi-dimensional data insights;

[0045] In the co-firing benefit evaluation model established by the comparative analysis module, its multi-objective optimization function is:

[0046] ;

[0047] Its constraints are:

[0048] Calorific value constraint: ;

[0049] Sulfur content limit: ;

[0050] Moisture limit: ;

[0051] in, The overall benefit score; , , All are weighting coefficients; The percentage increase in boiler efficiency; For the benefit of environmental protection and emission reduction; Savings in combustion costs; Let be the proportion of coal used in the i-th batch; Let be the lower heating value of the i-th coal; Let be the sulfur content of the i-th type of coal; Let be the total moisture content of the i-th type of coal; This refers to the maximum permissible total sulfur emissions as stipulated by environmental regulations. The minimum total heat requirement for boiler combustion; The maximum overall moisture threshold allowed by the process;

[0052] Digital twin module: Constructs a virtual model that is highly consistent with the physical fuel system, supporting system behavior simulation and key parameter trend prediction;

[0053] The energy conservation equation for the constructed boiler combustion dynamics model is as follows:

[0054] ;

[0055] The heat transfer coefficient is corrected to:

[0056] ;

[0057] in, For inputting heat; This is due to heat loss; For the increase of vapor enthalpy; Re represents the enthalpy change of the flue gas; Re and Pr are the Reynolds number and Prandtl number, respectively. Dynamic viscosity; Wall viscosity;

[0058] The decision-making module supports continuous learning and parameter tuning, generating scientific management suggestions and optimization solutions; it employs a linear programming solution model with the objective function being:

[0059] ;

[0060] Its constraints are:

[0061] Inventory balance: ;

[0062] Supply capacity: ;

[0063] Storage capacity: ;

[0064] in, Total operating costs; This refers to the unit price of the procurement cost. This refers to the unit price of holding costs. For purchase quantity; This refers to inventory levels; For demand; V represents the maximum supply; V represents the storage capacity. The beginning inventory of period t is the inventory level at the end of the previous period. The ending inventory of period t is the inventory level at the end of the current period.

[0065] Furthermore: the application service layer includes a monitoring service module, an optimization service module, a tracing service module, and an early warning service module;

[0066] Monitoring service module: Supports custom monitoring rules and alarm conditions, and provides real-time status awareness and performance evaluation;

[0067] Optimize service module: Supports manual intervention and scheme adjustment to ensure that optimization results are scientific and practical;

[0068] Traceability Service Module: Utilizing the immutability of blockchain technology, a trustworthy fuel quality traceability system is built, supporting multi-dimensional traceability analysis;

[0069] Early warning service module: It can dynamically adjust the early warning threshold and establish a comprehensive early warning processing and tracking mechanism;

[0070] The determination of the warning threshold is based on anomaly detection statistics;

[0071] The formula for calculating the median absolute deviation (MAD) is as follows:

[0072] ;

[0073] An improved Z-score algorithm is used, namely:

[0074] ;

[0075] Warning triggering conditions: If A value greater than 3.5 indicates a severe abnormality; a value greater than or equal to 3.5 indicates a serious abnormality. A value greater than 2.5 indicates a general abnormality;

[0076] in, Represents the i-th observation in the dataset; median(X) represents the median of dataset X; MAD is the median of all absolute deviations. This indicates the signed deviation, with positive values ​​being higher than the reference and negative values ​​being lower than the reference.

[0077] Furthermore: the presentation layer provides three display methods, namely, a web visual interface, a mobile application, and a large-screen display;

[0078] Web visualization interface: Provides rich interactive functions and detailed data display, and supports multi-browser compatibility;

[0079] Mobile application: Optimized for mobile device usage, it supports offline data viewing and operation, and provides convenient mobile monitoring functions;

[0080] Large screen display: Supports stunning visual effects to help administrators quickly grasp the system's operating status.

[0081] This invention has the following advantages: It establishes a unified data model and standardized interface, connecting the entire data chain from procurement, delivery, and inventory to combustion and emissions, completely eliminating information silos and enabling precise tracking and visualized management of each fuel unit throughout the entire process; it can simulate and comprehensively evaluate blending schemes, transforming fuel proportioning from reliance on manual experience to data-driven scientific decision-making, significantly improving boiler efficiency and environmental performance; and through an intelligent early warning mechanism, it achieves early risk identification, effectively ensuring the safe and stable operation of the unit.

[0082] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. Attached Figure Description

[0083] To more intuitively illustrate the prior art and this application, exemplary drawings are provided below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing this application; for example, based on the technical concept disclosed in this application and the exemplary drawings, those skilled in the art are able to easily make conventional adjustments or further optimizations to the addition / reduction / classification, specific shapes, positional relationships, connection methods, size ratios, etc. of certain units (components).

[0084] Figure 1 This is a system block diagram of a data processing system applicable to the entire fuel management process of a thermal power plant, as described in this application.

[0085] Figure 2 This is a timing diagram of data processing in a data processing system applicable to the entire fuel management process of a thermal power plant, according to the present invention. Detailed Implementation

[0086] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these embodiments are merely for further explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Technical engineers in the field can make some non-essential improvements and adjustments to the present invention based on the above-described content. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0087] Please see Figures 1-2 A data processing system suitable for the entire fuel management process in thermal power plants, comprising a data acquisition layer, a data storage layer, a data processing layer, an application service layer, and a presentation layer.

[0088] The data acquisition layer collects raw data in real time and accurately from all key nodes in the entire fuel process (such as procurement, entry, inventory, consumption, testing, and environmental protection); the data storage layer provides secure, reliable, and efficient storage and management for massive amounts of multi-category raw data; and the data processing layer processes and refines the stored raw data, transforming it into valuable information and knowledge.

[0089] The application service layer encapsulates the information generated by the data processing layer into specific, reusable business applications and services; the presentation layer provides an interface for human-computer interaction, presenting complex data and analysis results to users with different roles in an intuitive and easy-to-understand form.

[0090] After the data acquisition layer performs preliminary edge preprocessing on the raw data (such as format standardization and preliminary filtering), it classifies the data, timestamps it, and continuously injects it into the corresponding database in the data storage layer (such as real-time data into the real-time database and business documents into the business database).

[0091] The data processing layer actively pulls data from the data storage layer; the data processing layer first cleans and integrates the data, processing the messy original data into a high-quality, interconnected end-to-end data chain; then it performs calculations and analysis, producing key indicators such as unit coal consumption and inventory turnover rate, trend analysis reports and forecast results.

[0092] The data processing layer outputs processed valuable information, including cleaned and integrated standard data, calculated indicators, analysis conclusions, early warning signals, and digital twin simulation results.

[0093] The data processing layer stores these results back to the data storage layer (forming data assets) and simultaneously pushes them to the application service layer; the application service layer (such as the monitoring service) then calls these information and pushes them to the presentation layer.

[0094] After receiving the information, the presentation layer performs visualization rendering and finally presents it to the user in the form of charts, curves, 3D models, warning pop-ups, etc.

[0095] The presentation layer transmits user instructions to the decision module in the data processing layer; the decision module generates specific control instructions and sends them to the underlying control system (such as coal feeder and coal blending device) connected to the data acquisition layer through the system interface; at the same time, the decision module can also request more in-depth analysis data from the data processing layer to verify or optimize the solution, and the processing layer responds to the request and provides support.

[0096] Once a new control command is executed, the data acquisition layer continuously monitors new data such as fuel consumption, boiler efficiency, and environmental emissions. The system compares and analyzes the new real-time data with the expected optimization results to verify the correctness of the decision and provide a basis for further optimization. This process repeats itself, forming a data closed loop of "monitoring-analysis-decision-execution-remonitoring," enabling continuous iterative optimization of fuel management.

[0097] The data acquisition layer includes a procurement management system, a weighing system, a coal inventory system, a boiler control system, a testing system, and an ash and slag treatment system.

[0098] The procurement management system collects data such as procurement plan number, planned coal type, planned procurement quantity, planned procurement time, and supplier information. It supports multiple data interface protocols to ensure the integrity and accuracy of procurement data.

[0099] Weighing system: Integrates weighing equipment such as truck scales and belt scales to collect real-time weight information such as vehicle number, arrival time, coal type, and actual purchase quantity of coal entering the plant.

[0100] Coal inventory system: It adopts laser scanning and 3D modeling technology to dynamically collect spatial distribution data such as inventory location, inventory quantity, and inventory update time of each type of coal, and supports both automatic and manual coal inventory modes.

[0101] Boiler control system: Real-time monitoring of operating parameters such as the consumption of each type of coal, limestone consumption, and corresponding power generation load per unit time, and interface with the DCS system through standard protocols such as OPC and Modbus.

[0102] The testing system collects key quality indicators such as moisture, ash, volatile matter, and calorific value of incoming coal through automated testing equipment, and supports automatic data entry and intelligent review of testing data.

[0103] Ash and slag treatment system: Through the deployment of weighing sensors and flow metering equipment, it accurately collects environmentally relevant data such as the amount of ash and slag generated after combustion, the amount of ash and slag transported, and the treatment methods.

[0104] The data storage layer adopts an innovative hybrid storage architecture, including real-time databases, business databases, historical databases, and a data lake.

[0105] Real-time database: Employs time-series database core technology, specifically designed to handle high-frequency, high-concurrency sensor data, supports millisecond-level data writing and query response, and features data compression and caching optimization functions.

[0106] Business Database: Based on a relational database architecture, it adopts a distributed transaction processing mechanism to provide secure and reliable transactional storage for core business data of fuel management, and supports SQL standards and ACID properties.

[0107] Historical Database: Employing distributed columnar storage technology combined with advanced data compression algorithms, it supports efficient storage and fast querying of petabyte-level historical data, and provides data archiving and lifecycle management functions.

[0108] Data lakes employ an open storage format, supporting the raw storage of structured, semi-structured, and unstructured data, providing abundant data resources for data mining and machine learning applications.

[0109] The data processing layer includes a data cleaning module, a data integration module, a data calculation module, a comparative analysis module, a digital twin module, and a decision-making module.

[0110] Data cleaning module: Employs deep learning-based anomaly detection algorithms and integrates multiple data repair algorithms, including time series prediction-based interpolation, association rule-based mean imputation, and collaborative filtering methods that consider spatial correlation, to establish a multi-dimensional data quality assessment system.

[0111] Data integration module: Construct a unified data model with "fuel unit" as the basic management unit and "coal type" and "time" as the core correlation dimensions. Use graph database technology to build a fuel flow relationship network to achieve deep data integration across systems.

[0112] Data Calculation Module: Based on a professional fuel management algorithm library, it calculates key performance indicators such as incoming coal qualification rate, coal inventory turnover rate, coal consumption per unit of power generation, and ash and slag output rate in real time, and supports the customization of indicators and flexible adjustment of calculation rules.

[0113] The pass rate of incoming coal is a core indicator for measuring the compliance of purchased coal quality standards, and its calculation formula is as follows:

[0114] ;

[0115] Where Rqual is the qualified rate of incoming coal (%); Q pass,i Q represents the quantity (tons) of qualified coal in the i-th batch. total,jLet n be the total amount of coal delivered to the plant in tons for the j-th batch; n and m are the number of qualified batches and the total number of batches, respectively.

[0116] Coal inventory turnover rate is a dynamic indicator reflecting the efficiency of inventory utilization. Its calculation formula is as follows:

[0117] ;

[0118] Among them, T turn Inventory turnover days (days); C avg Average daily coal consumption (tons / day); I begin I end These are the beginning and ending inventory levels (in tons).

[0119] Coal consumption per unit of electricity generated is a key performance indicator characterizing the economic efficiency of coal combustion, and its calculation formula is as follows:

[0120] ;

[0121] Among them, C unit B represents the standard coal consumption for power generation (g / kWh); E represents the total coal consumption during the statistical period (tons); and E represents the power generation during the same period (MWh).

[0122] Ash and slag yield can be used to assess the efficiency of post-combustion by-product generation. Its calculation formula is as follows:

[0123] ;

[0124] Among them, Y ash W represents the overall yield of ash and slag (%). slag Slag production (tons); W flyash B represents the amount of fly ash generated (tons); B represents the total amount of coal fed into the furnace (tons).

[0125] Comparative Analysis Module: Provides multiple analysis modes such as horizontal comparison, vertical analysis, and trend prediction. It adopts intelligent pattern recognition algorithm and is specially designed with blending effect analysis function to support multi-dimensional data insight.

[0126] In the established co-firing benefit evaluation model, its multi-objective optimization function is:

[0127] ;

[0128] Its constraints are:

[0129] Calorific value constraint: ;

[0130] Sulfur content limit: ;

[0131] Moisture limit: ;

[0132] in, The overall benefit score; , , All are weighting coefficients; The percentage increase in boiler efficiency; For the benefit of environmental protection and emission reduction; Savings in combustion costs; Let be the proportion of coal used in the i-th batch; Let be the lower heating value of the i-th coal; Let be the sulfur content of the i-th type of coal; Let be the total moisture content of the i-th type of coal; This refers to the maximum permissible total sulfur emissions as stipulated by environmental regulations. The minimum total heat requirement for boiler combustion; This represents the highest possible overall moisture threshold allowed by the process.

[0133] Digital Twin Module: Based on accurate physical models and machine learning algorithms, it constructs a virtual model that is highly consistent with the physical fuel system, supporting system behavior simulation and key parameter trend prediction.

[0134] The energy conservation equation for the constructed boiler combustion dynamics model is as follows:

[0135] ;

[0136] The heat transfer coefficient is corrected to:

[0137] ;

[0138] Among them, among them, Input heat (MW); Heat loss (MW); For the increase in vapor enthalpy (J / kg); Δ is the enthalpy change of flue gas (J / kg); Re and Pr are the Reynolds number and Prandtl number, respectively; Dynamic viscosity; The viscosity is the wall viscosity.

[0139] Decision module: Integrates multiple optimization algorithms and decision models, supports continuous learning and parameter tuning, and generates scientific management suggestions and optimization solutions.

[0140] In this embodiment, a linear programming solution model is used, and its objective function is:

[0141] ;

[0142] Its constraints are:

[0143] Inventory balance: ;

[0144] Supply capacity: ;

[0145] Storage capacity: ;

[0146] in, Total operating costs; This refers to the unit price of the procurement cost. This refers to the unit price of holding costs. For purchase quantity; This refers to inventory levels; For demand; V represents the maximum supply; V represents the storage capacity. The beginning inventory of period t is the inventory level at the end of the previous period. The ending inventory of period t is the inventory level at the end of the current period.

[0147] The application service layer includes a monitoring service module, an optimization service module, a tracing service module, and an early warning service module.

[0148] Monitoring service module: Adopting a distributed monitoring architecture and microservice design concept, it supports custom monitoring rules and alarm conditions, and provides real-time status awareness and performance evaluation.

[0149] Optimization Service Module: Based on multi-objective optimization theory, a complete fuel blending optimization model system is constructed, supporting manual intervention and scheme adjustment to ensure that the optimization results are scientific and practical.

[0150] Traceability Service Module: Utilizing the immutability of blockchain technology, a trustworthy fuel quality traceability system is built, supporting multi-dimensional traceability analysis.

[0151] Early warning service module: Employs machine learning and statistical analysis techniques to dynamically adjust early warning thresholds and establish a comprehensive early warning processing and tracking mechanism.

[0152] The determination of the warning threshold is based on anomaly detection statistics;

[0153] The formula for calculating MAD (Median Absolute Deviation) is as follows:

[0154] ;

[0155] An improved Z-score algorithm is used, namely:

[0156] ;

[0157] Warning triggering conditions: If A value greater than 3.5 indicates a severe abnormality; a value greater than or equal to 3.5 indicates a serious abnormality. A value greater than 2.5 indicates a general abnormality;

[0158] in, Represents the i-th observation in the dataset; median(X) represents the median of dataset X (the middle value after sorting all Xi); MAD is the median of all absolute deviations; This indicates a signed deviation, with positive values ​​being higher than the reference (e.g., higher calorific value) and negative values ​​being lower than the reference (e.g., lower calorific value).

[0159] Taking the detection of the calorific value of coal fed into the boiler as an example; the data acquisition process is to obtain calorific value data X from the boiler control system every hour. i For example, [5180, 5220, 5150, 5300, 4900] kcal / kg.

[0160] During calculation, median(X) = 5180 (the median value after sorting);

[0161] The absolute deviation sequence [0, 40, 30, 120, 280] gives MAD = median(0, 30, 40, 120, 280) = 40;

[0162] The new data Xnew=4800, then =-6.745;

[0163] Warning value determination:

[0164] =6.745>3.5, triggering a critical anomaly alarm;

[0165] Root cause: A sudden drop in calorific value of 280 kcal / kg may cause the boiler to extinguish, requiring emergency intervention.

[0166] The presentation layer offers three display methods: a web-based visual interface, a mobile application, and a large-screen display.

[0167] Web visualization interface: Based on the responsive design concept, it provides rich interactive functions and detailed data display, and supports multi-browser compatibility.

[0168] Mobile application: Optimized for mobile device usage, it supports offline data viewing and operation, and provides convenient mobile monitoring functions.

[0169] Large screen display: Designed with a cockpit concept, it supports stunning visualization effects, helping managers quickly grasp the system's operating status.

[0170] See Figure 2 The data processing timing flow of the system of the present invention is as follows:

[0171] (1) Data acquisition stage: Each data source sends real-time data streams to the acquisition module at a predetermined frequency. The data acquisition module performs edge preprocessing, including data format standardization, basic validity verification, and preliminary filtering of abnormal data. The processed data is then classified and stored in the corresponding database units.

[0172] (2) Data processing stage: The data storage module provides batch data to the data processing module. The data processing module performs data cleaning, data integration, data calculation and comparative analysis in sequence to form a complete data analysis archive and store it back to the data storage module.

[0173] (3) Monitoring and early warning stage: The data storage module pushes real-time monitoring data to the display layer, performs visualization rendering and threshold checks in the display layer, generates early warning information when an anomaly is detected, and initiates the early warning processing flow through the decision module.

[0174] (4) Optimization decision-making stage: The decision-making module requests in-depth analysis data from the data processing module, generates an optimization plan based on the analysis report and trend prediction, combined with the optimization model and business rules, and pushes it to the display layer.

[0175] (5) Feedback and adjustment phase: Managers review and confirm the optimization plan through the presentation layer. The system sends control instructions to relevant data sources through the decision module, continuously monitors the implementation effect of the plan, and collects feedback data.

[0176] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A data processing system suitable for fuel full-cycle management of a thermal power plant, characterized in that, The system comprises a data collection layer, a data storage layer, a data processing layer, an application service layer, and a display layer. The data collection layer collects real-time and accurate original data from each key node of the fuel whole process. After the data collection layer performs preliminary edge preprocessing on the original data, it classifies and timestamps the data and continuously injects the data into the corresponding database in the data storage layer. The data storage layer provides storage and management for the original data. The data processing layer processes and refines the stored original data and converts it into valuable information and knowledge. The data processing layer actively pulls data from the data storage layer. The data processing layer first performs data cleaning and integration to process the disordered original data into high-quality and correlated whole-process data chain. Then, the data processing layer performs calculation and analysis and outputs the processed valuable information. The data processing layer stores the valuable information back to the data storage layer and simultaneously pushes the valuable information to the application service layer. The application service layer encapsulates the valuable information generated by the data processing layer into specific and reusable business applications and services. The display layer provides an interface for human-computer interaction and presents complex data and analysis results to users in an intuitive and easy-to-understand manner. The display layer transmits user instructions to the decision module in the data processing layer. The decision module generates specific control instructions and sends the instructions to the underlying control system connected to the data collection layer through a system interface.

2. The data processing system for fuel whole-process management of thermal power plants according to claim 1, characterized in that, The valuable information includes cleaned and integrated standard data, calculated indicators, analysis conclusions, early warning signals, and digital twin simulation results.

3. The data processing system for fuel total process management of thermal power plant according to claim 1, characterized in that, After the execution of new control instructions, the data collection layer continuously monitors new fuel consumption, boiler efficiency, and environmental emission data. The system compares the new real-time data with the expected optimization effect to verify the correctness of the decision.

4. The data processing system for fuel total process management of thermal power plant according to claim 1, wherein, The data collection layer comprises a procurement management system, a weighing system, a coal inventory system, a boiler control system, a testing system, and an ash disposal system. The procurement management system collects procurement plan number data, planned coal type data, planned purchase quantity data, planned purchase time data, and supplier information data, and supports multiple data interface protocols. The weighing system integrates metering equipment to collect real-time vehicle number, arrival time, coal type, and coal weight information of incoming coal. The coal inventory system uses laser scanning and three-dimensional modeling technology to dynamically collect spatial distribution data of each coal type, supporting automatic coal inventory and manual coal inventory. The boiler control system monitors real-time operating parameters of each coal type per unit time and interfaces with the DCS system through a standard protocol. The testing system collects key quality indicators of incoming coal through automatic testing equipment, supporting automatic entry and intelligent review of testing data. The ash disposal system accurately collects environmental protection-related data of burned ash through deployed weighing sensors and flow metering equipment.

5. The data processing system for fuel total process management of thermal power plant according to claim 1, wherein, The data storage layer adopts an innovative hybrid storage architecture, including a real-time database, a business database, a historical database, and a data lake. The real-time database uses time series database core technology to process high-frequency and high-concurrency sensor data, supports millisecond-level data writing and query response, and has data compression and cache optimization functions. Business database: based on a relational database architecture, using distributed transaction processing mechanism, provides transactional storage guarantee for the core business data of fuel management; History database: using distributed column storage technology, combined with data compression algorithm, supports efficient storage and fast query of PB level historical data, provides data archiving and life cycle management functions; Data lake: using open storage format, supporting structured, semi-structured and unstructured data raw storage.

6. The data processing system for fuel total process management of thermal power plant according to claim 1, wherein, The data processing layer includes data cleaning module, data integration module, data calculation module, comparative analysis module, digital twin module and decision module; Data cleaning module: using deep learning based anomaly detection algorithm, integrating multiple data repair algorithms, including interpolation method based on time series prediction, mean filling method based on association rules, considering spatial correlation collaborative filtering method, establishing multi-dimensional data quality evaluation system; Data integration module: build a unified data model with "fuel unit" as the basic management unit and "coal type" and "time" as the core related dimensions, use graph database technology to build fuel flow relationship network, realize cross-system data deep integration.

7. The data processing system for fuel total process management of thermal power plant according to claim 6, wherein, The data calculation module: based on professional fuel management algorithm library, real-time calculation of key performance indicators including incoming coal qualified rate, coal inventory turnover rate, unit power generation coal consumption, ash slag output rate, support for index customization and flexible adjustment of calculation rules; The qualified rate of incoming coal is the core index to measure the quality of purchased coal, and its calculation formula is: ； Wherein, Rqual is the qualified rate of incoming coal; Q pass,i is the amount of qualified coal of the i th batch; Q total,j is the total amount of coal of the j th batch to the plant; n and m are the number of qualified batches and the total number of batches, respectively. The coal inventory turnover rate is a dynamic index reflecting the utilization efficiency of inventory, and its calculation formula is: ； Wherein, T turn is the inventory turnover days; C avg is the daily average coal consumption; I begin , I end are the initial and final inventory, respectively; The unit power generation coal consumption is a key performance indicator to represent the economy of coal-fired power generation, and its calculation formula is: ； wherein C unit is the standard coal consumption for power supply; B is the total coal consumption in the statistical period; E is the power generation in the same period; The ash slag output rate can evaluate the efficiency of byproduct generation after combustion, and its calculation formula is: ； where Y ash is the overall yield of ash; W slag is the amount of slag produced; W flyash is the amount of fly ash produced; and B is the total amount of coal fed to the furnace.

8. The data processing system for fuel total process management of thermal power plant according to claim 6, wherein, The comparative analysis module provides multiple analysis modes and specially designs the blending effect analysis function to support multi-dimensional data insight; The multi-objective optimization function of the blending benefit evaluation model established by the comparative analysis module is: ； The constraint condition is: Calorific value constraint: ​ Sulfur fractionation: ​ Upper moisture limit: ; wherein, is the comprehensive benefit score; , , are weight coefficients; is the boiler efficiency improvement range; is the environmental protection and emission reduction benefit; is the combustion cost saving amount; is the blending ratio of the i-th coal; is the low calorific value of the i-th coal; is the sulfur content of the i-th coal; is the total moisture content of the i-th coal; is the maximum allowable total sulfur emission amount specified by environmental protection regulations; is the minimum total heat requirement required for boiler combustion; is the highest comprehensive moisture threshold allowed by the process; The digital twin module: builds a virtual model highly consistent with the physical fuel system, supports system behavior simulation and key parameter trend prediction; Among them, the energy conservation equation of the boiler combustion kinetics model is: ； The heat transfer coefficient is corrected as: ； wherein, Qin is the input heat; Qout is the heat loss; Δh is the steam enthalpy increase; Δf is the flue gas enthalpy change; Re, Pr are the Reynolds and Prandtl numbers, respectively; μ is the dynamic viscosity; μw is the wall viscosity; The decision module: supports continuous learning and parameter optimization, generates scientific management recommendations and optimization schemes; uses linear programming solution model, and its objective function is: ； The constraint condition is: Inventory balancing: ​ Supply capacity: ​ Storage capacity: ​ wherein, is the total operating cost; is the purchase cost unit price; is the holding cost unit price; is the purchase quantity; is the inventory quantity; is the demand quantity; is the maximum supply quantity; V is the capacity of the warehouse; is the inventory level at the beginning of the period t, i.e. at the end of the previous cycle; is the inventory level at the end of the period t, i.e. at the end of the current cycle.

9. The data processing system for fuel total process management of thermal power plant according to claim 1, wherein, The application service layer includes monitoring service module, optimization service module, traceability service module and early warning service module; The monitoring service module: supports custom monitoring rules and alarm conditions, provides real-time state awareness and performance evaluation; The optimization service module: supports manual intervention and scheme adjustment to ensure that the optimization results are scientific and practical; The traceability service module: uses the tamper-proof feature of blockchain technology to build a trusted fuel quality traceability system and supports multi-dimensional traceability analysis; The early warning service module: can dynamically adjust the threshold value and establish a perfect early warning processing and tracking mechanism; The determination of the early warning threshold value is based on the anomaly detection statistic quantity; Wherein, the median absolute deviation MAD calculation formula is: ； The improved Z-score algorithm is adopted, namely: ； Warning triggering conditions: If A value greater than 3.5 indicates a severe abnormality; a value greater than or equal to 3.5 indicates a serious abnormality. A value greater than 2.5 indicates a general abnormality; wherein, represents the ith observation in the dataset; median(X) represents the median of the dataset X; MAD is the median of all absolute deviations; represents the signed deviation, with positive values above the baseline and negative values below the baseline.

10. The data processing system for fuel total process management of thermal power plant according to claim 1, wherein, The display layer provides three display modes, which are web visual interface, mobile application and large screen display. Web visual interface: provides rich interactive functions and detailed data display, supports multi-browser compatibility; Mobile application: optimized for mobile device usage characteristics, supports offline data viewing and operation, provides convenient mobile monitoring function; Large screen display: supports stunning visual effect display, helps managers quickly grasp the system running status.

Images