Intelligent monitoring system and method for thermal power generating unit based on artificial intelligence technology

Abstract

The present application relates to the technical field of operation monitoring and intelligent management of thermal power generating units, and discloses an intelligent monitoring system and method for thermal power generating units based on artificial intelligence technology, which comprises a data acquisition and preprocessing module: connected with a DCS system interface, used for real-time acquisition of various parameters in the operation process of the thermal power generating unit, including temperature, pressure, flow, equipment state, etc., and performing cleaning, filtering and standardization processing on the acquired data to remove noise and abnormal data and ensure data quality; an intelligent computing engine: having high real-time performance and high concurrent processing capability, supporting operation of various intelligent algorithms such as machine learning and deep learning, and adopting a computing load balancing technology to dynamically allocate computing resources according to task requirements. Through a parameter early warning model and an equipment health scoring mechanism, important operation parameters of the thermal power generating unit are monitored and intelligently analyzed in real time, potential faults and hidden dangers are found in advance, and the limitations of the traditional post-alarm mode are effectively avoided.

Description

Technical Field

[0001] This invention relates to the field of thermal power unit operation monitoring and intelligent management technology, specifically to an intelligent monitoring system and method for thermal power units based on artificial intelligence technology. Background Technology

[0002] Thermal power units are crucial equipment in power generation, and their operational status directly affects the security and stability of the power supply. During operation, thermal power units generate a large amount of operating parameters and equipment status data, such as temperature, pressure, flow rate, and equipment vibration. This data contains critical information about the unit's operating condition and is of great significance for ensuring the safe and economical operation of the unit.

[0003] Traditional monitoring of thermal power units relies primarily on operators' real-time observation and experience-based judgment of numerous operating parameters and equipment status measurement points in the DCS system. This model requires operators to maintain high concentration for extended periods, resulting in extremely high workload and a high risk of information overload or misjudgment due to fatigue. Furthermore, existing DCS system alarm mechanisms are mostly reactive, lacking early warning capabilities for potential faults. Once parameters exceed limits or equipment trips, the time available for operators is extremely short, easily leading to serious accidents such as equipment damage and unit shutdowns. In addition, the DCS system itself has limited functionality, unable to provide guidance on optimal unit operating modes, assess equipment health status, or enhance system automation. These problems not only affect the operating efficiency and safety of thermal power units but also severely restrict the intelligent development of power plants.

[0004] Therefore, the traditional monitoring mode of thermal power units can no longer meet the needs of safe, economical and intelligent operation of modern thermal power units, and there is obvious room for improvement. A more efficient and intelligent monitoring system and method are needed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent monitoring system and method for thermal power units based on artificial intelligence technology, which solves the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent monitoring system for thermal power units based on artificial intelligence technology, comprising: Data acquisition and preprocessing module: Connected to the DCS system interface, it is used to collect various parameters during the operation of thermal power units in real time, including temperature, pressure, flow rate, equipment status, etc., and to clean, filter and standardize the collected data to remove noise and abnormal data and ensure data quality. Intelligent computing engine: It has high real-time performance and high concurrency processing capabilities, supports the operation of various intelligent algorithms such as machine learning and deep learning, adopts computing load balancing technology, can dynamically allocate computing resources according to task requirements, and supports free configuration of single machines and computing clusters to meet the computing needs of thermal power units of different sizes. Intelligent Analysis Module: Based on big data analysis and artificial intelligence algorithms, it performs in-depth analysis of the collected operating parameters, and realizes functions such as parameter early warning model, equipment health scoring mechanism, energy efficiency and consumption difference analysis model, and intelligent control optimization strategy; Early warning and control module: Based on the output of the intelligent analysis module, it generates corresponding early warning signals and control commands. The early warning signals are promptly notified to the duty officer and maintenance personnel through audible and visual alarms, SMS notifications, etc., while the control commands are directly sent to the DCS system or other automated control systems to realize automatic control of the equipment. Data Interaction and Visualization Module: Enables data interaction between various modules within the system, as well as with external systems, displaying operating parameters, early warning information, equipment health status, and other information in an intuitive way on the user interface.

[0007] Preferably, the parameter early warning model targets important operating parameters of megawatt-class thermal power units, and learns and analyzes historical and real-time data of the parameters through algorithms such as cluster analysis and neural networks to identify the normal fluctuation range and abnormal trends of the parameters, and issues early warning signals in advance to avoid failures caused by parameters exceeding limits.

[0008] Preferably, the equipment health scoring mechanism comprehensively considers the equipment's operating data to generate a real-time health score for each device. By establishing an equipment health status assessment model and using data fusion technology, the system weights and calculates data from different dimensions to obtain the equipment's health score. When the health score is lower than a preset threshold, the system automatically triggers an alarm to remind maintenance personnel to conduct timely inspections and repairs.

[0009] Preferably, the energy efficiency and energy consumption difference analysis model establishes a data-driven energy efficiency analysis model to accurately measure the operating efficiency and energy consumption of the computer group, analyze the energy consumption distribution of each link, identify potential energy-saving space, and provide optimization suggestions.

[0010] Preferably, the intelligent control optimization strategy combines intelligent energy consumption difference analysis and optimization model to form an optimal closed-loop control system. Based on real-time energy consumption data, equipment health status and other information, it automatically adjusts production parameters to achieve intelligent control of the production process and improve production efficiency and energy utilization.

[0011] Preferably, the intelligent computing engine adopts a distributed computing architecture, which can distribute computing tasks to multiple computing nodes for parallel processing.

[0012] Preferably, the data interaction and visualization module supports multi-source data visualization methods, including charts, curves, maps, etc., and users can customize the displayed content and layout as needed.

[0013] The intelligent monitoring method for thermal power units based on artificial intelligence technology includes the following steps: Data acquisition and preprocessing: Real-time acquisition of various parameters during the operation of thermal power units, followed by cleaning, filtering, and standardization. Intelligent computing and analysis: The pre-processed data is input into the intelligent computing engine, and in-depth analysis is performed through parameter early warning models, equipment health scoring mechanisms, energy efficiency and consumption difference analysis models in the intelligent analysis module to generate corresponding early warning signals and optimization suggestions; Early warning and control: Based on the intelligent analysis results, the early warning and control module generates early warning signals and notifies relevant personnel through various means. At the same time, it generates control commands based on optimization suggestions and sends them to the automated control system to realize automatic control of the equipment. Data interaction and visualization: Data interaction is carried out between various modules within the system, and information such as operating parameters, early warning information, and equipment health status are displayed to the duty officer through a visual interface for real-time monitoring and operation; Self-learning and self-adaptation: During operation, the system continuously learns and accumulates operational data, automatically adjusting model parameters and optimization strategies to adapt to changes in the operating conditions of thermal power units.

[0014] Preferably, in the intelligent computing and analysis step, the parameter early warning model uses a long short-term memory network in deep learning algorithms to model and analyze operating parameters with time-series characteristics. The equipment health scoring mechanism uses a Bayesian network to construct an equipment health status assessment model, describes the dependency relationship between various influencing factors through conditional probability distribution, and accurately assesses the health status of the equipment. In the early warning and control step, the generation of control commands uses a fuzzy control algorithm to transform the optimization suggestions of operating parameters into specific control commands, which can handle the uncertainty and fuzziness in the control process.

[0015] This invention provides an intelligent monitoring system and method for thermal power units based on artificial intelligence technology. It has the following beneficial effects: 1. This invention utilizes a parameter early warning model and an equipment health scoring mechanism to monitor and intelligently analyze critical operating parameters of thermal power units in real time, identifying potential faults early and effectively avoiding the limitations of traditional reactive alarm methods. The parameter early warning model employs advanced algorithms such as cluster analysis and neural networks to accurately identify normal fluctuation ranges and abnormal trends in parameters, issuing early warning signals to provide operators with sufficient time to take action and prevent faults caused by parameters exceeding limits. The equipment health scoring mechanism integrates multi-dimensional information such as equipment operating data, historical fault records, and operating time to generate a real-time health score for each piece of equipment. When the score falls below a preset threshold, an alarm is automatically triggered, reminding maintenance personnel to promptly inspect and repair the equipment.

[0016] 2. The energy efficiency and energy consumption difference analysis model of this invention is data-driven, accurately calculating the operating efficiency and energy consumption of computer groups, deeply analyzing the energy consumption distribution of each link, identifying energy-saving potential, and providing optimization suggestions. Simultaneously, the intelligent control optimization strategy, combined with intelligent energy consumption difference analysis and the optimization model, forms an optimal closed-loop control system. Based on real-time energy consumption data and equipment health status, it automatically adjusts production parameters, achieving intelligent regulation and control, effectively reducing coal consumption and improving energy utilization.

[0017] 3. This invention employs machine learning, deep learning, and data analysis algorithms to automatically learn and identify complex patterns and abnormal trends in the operation of thermal power units, replacing tedious manual monitoring and reducing labor intensity. The system possesses self-learning and adaptive capabilities, automatically adjusting model parameters and optimization strategies based on actual unit operating conditions and data changes to ensure long-term stable operation. Simultaneously, the system seamlessly integrates with existing monitoring and control systems such as DCS systems, enabling real-time data linkage and forming a complete intelligent monitoring solution. Through data interaction and visualization modules, operating parameters and early warning information are displayed intuitively, facilitating real-time monitoring by operators. This not only improves monitoring efficiency and quality but also provides strong support for the intelligent management of power plants, promoting the development of power plants towards intelligence and automation. Attached Figure Description

[0018] Figure 1 This is a flowchart of the intelligent monitoring system of the present invention. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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. Example

[0020] Please see the appendix Figure 1This invention provides an intelligent monitoring system for thermal power units based on artificial intelligence technology, comprising: Data acquisition and preprocessing module: Connected to the DCS system interface, it is used to collect various parameters during the operation of thermal power units in real time, including temperature, pressure, flow rate, equipment status, etc., and to clean, filter and standardize the collected data to remove noise and abnormal data and ensure data quality. Specifically, this module is primarily responsible for interface connection with the DCS (Distributed Control System) of the thermal power unit, enabling real-time acquisition of various key parameters during unit operation. These parameters cover multiple aspects such as temperature, pressure, flow rate, and equipment status, comprehensively reflecting the operating status of the thermal power unit. After acquiring this raw data, the module performs a series of preprocessing operations. Specifically, this includes data cleaning to remove noise and outliers, ensuring data purity; data filtering to select the parts of data that are of significant value to the monitoring system; and data standardization to convert data of different types and dimensions into a standardized format, thereby ensuring the reliability and accuracy of data quality in subsequent analysis and processing, laying a solid foundation for the efficient operation of the entire intelligent monitoring system.

[0021] Intelligent computing engine: It has high real-time performance and high concurrency processing capabilities, supports the operation of various intelligent algorithms such as machine learning and deep learning, adopts computing load balancing technology, can dynamically allocate computing resources according to task requirements, and supports free configuration of single machines and computing clusters to meet the computing needs of thermal power units of different sizes. Specifically, the intelligent computing engine, as the core power component of the intelligent monitoring system for thermal power units, possesses high real-time performance and high concurrency processing capabilities, meeting the demands for rapid data analysis and processing during the operation of thermal power units. It supports the operation of various intelligent algorithms, including but not limited to advanced algorithms such as machine learning and deep learning. This enables the system to perform in-depth mining and intelligent analysis of massive amounts of collected operational data, thereby extracting valuable information. Furthermore, the engine employs computing load balancing technology, which flexibly and dynamically allocates computing resources according to different task requirements, providing excellent adaptation and support for both single-machine operation and computing cluster environments.

[0022] Intelligent Analysis Module: Based on big data analysis and artificial intelligence algorithms, it performs in-depth analysis of the collected operating parameters, realizing functions such as parameter early warning model, equipment health scoring mechanism, energy efficiency and consumption difference analysis model, and intelligent control optimization strategy; The parameter early warning model targets the important operating parameters of million-kilowatt thermal power units, and learns and analyzes historical and real-time data of parameters through cluster analysis, neural network and other algorithms, identifies the normal fluctuation range and abnormal trends of parameters, and issues early warning signals in advance to avoid failures caused by parameters exceeding limits.

[0023] Specifically, the intelligent analysis module is the core component of the intelligent monitoring system for thermal power units. Based on big data analytics and artificial intelligence algorithms, it performs in-depth analysis of pre-processed operating parameters. This module enables several key functions, including parameter early warning models, equipment health scoring mechanisms, energy efficiency and consumption difference analysis models, and intelligent control optimization strategies. Through these functions, the intelligent analysis module can comprehensively monitor the operating status of thermal power units, identify potential problems in advance, and provide decision support for optimizing unit operation. The parameter early warning model is a crucial component of the intelligent analysis module, focusing on monitoring and analyzing important operating parameters of million-kilowatt thermal power units. This model utilizes advanced algorithms such as cluster analysis and neural networks to learn and analyze historical and real-time parameter data. In this way, it can identify the normal fluctuation range and abnormal trends of parameters, thereby issuing early warning signals.

[0024] The equipment health scoring mechanism is also a key function of the intelligent analysis module. It comprehensively considers multi-dimensional information such as current, vibration, and temperature from various equipment operating data to generate a real-time health score for each device. By establishing an equipment health status assessment model and employing data fusion technology to weight and calculate data from different dimensions, the system obtains the equipment's health score. When the health score falls below a preset threshold, the system automatically triggers an alarm, promptly reminding maintenance personnel to inspect and repair the equipment, effectively reducing equipment failure rates and improving equipment reliability and operating efficiency.

[0025] The energy efficiency and energy consumption difference analysis model primarily focuses on the operating efficiency and energy consumption of thermal power units. Based on a data-driven approach, this model establishes an energy efficiency analysis model that accurately calculates the unit's operating efficiency and analyzes the energy consumption distribution at each stage. In this way, it identifies potential energy-saving opportunities and provides optimization suggestions to operators, thereby helping thermal power units achieve their energy conservation and emission reduction goals, improve energy utilization efficiency, and reduce production costs.

[0026] The intelligent control optimization strategy combines intelligent energy consumption analysis and optimization models to form an optimal closed-loop control system. Based on real-time energy consumption data and equipment health status, it automatically adjusts production parameters, such as boiler combustion parameters and turbine regulation parameters. In this way, intelligent control of the production process is achieved, improving production efficiency and energy utilization, and ensuring that thermal power units operate under optimal conditions.

[0027] The closed-loop control system includes the following formulas: ; in: This represents the control input, such as boiler combustion parameters and turbine regulation parameters. K represents the controller's gain matrix. e(t) represents the deviation between the controlled variable and the setpoint.

[0028] Dynamic programming solves optimal control problems recursively, and its core is the Bellman optimality equation. Its formula is as follows: ; in, Let represent the expected reward of executing the optimal policy in state S, 'a' represent the action (control variable) that can be taken in state S, 'a' represent the probability of transitioning to state s^' after taking action a in state S, and 'g' represent the immediate reward obtained after taking action a in state S and transitioning to state s^'. γ represents the discount factor, used to balance the weights of immediate and future rewards. Using a dynamic programming algorithm, the optimal action in each state can be solved, thereby achieving optimal control of the system.

[0029] The equipment health scoring mechanism comprehensively considers equipment operating data to generate a real-time health score for each device. By establishing an equipment health status assessment model and employing data fusion technology, it weights and calculates data from different dimensions to obtain the equipment's health score. When the health score falls below a preset threshold, the system automatically triggers an alarm, reminding maintenance personnel to promptly inspect and repair the equipment. The energy efficiency and energy consumption difference analysis model, based on a data-driven approach, accurately calculates the operating efficiency and energy consumption of the equipment group. It can analyze the energy consumption distribution at each stage, identify potential energy-saving opportunities, and provide optimization suggestions.

[0030] Early warning and control module: Based on the output of the intelligent analysis module, corresponding early warning signals and control commands are generated. Early warning signals are promptly notified to duty personnel and maintenance personnel through audible and visual alarms, SMS notifications, etc., while control commands are directly sent to the DCS system or other automated control systems to achieve automatic control of the equipment. The intelligent control optimization strategy, combined with intelligent energy consumption difference analysis and optimization model, forms an optimal closed-loop control system. Based on real-time energy consumption data, equipment health status and other information, production parameters are automatically adjusted to achieve intelligent regulation of the production process, thereby improving production efficiency and energy utilization.

[0031] Data Interaction and Visualization Module: Enables data interaction between various modules within the system, as well as with external systems, displaying operating parameters, early warning information, equipment health status, and other information in an intuitive way on the user interface.

[0032] The intelligent computing engine employs a distributed computing architecture, enabling the allocation of computing tasks across multiple computing nodes for parallel processing. The data interaction and visualization module supports various data visualization methods, including charts, curves, and maps, allowing users to customize the displayed content and layout as needed.

[0033] The intelligent monitoring method for thermal power units based on artificial intelligence technology includes the following steps: Data acquisition and preprocessing: Real-time acquisition of various parameters during the operation of thermal power units, followed by cleaning, filtering, and standardization. Intelligent computing and analysis: The pre-processed data is input into the intelligent computing engine, and in-depth analysis is performed through parameter early warning models, equipment health scoring mechanisms, energy efficiency and consumption difference analysis models in the intelligent analysis module to generate corresponding early warning signals and optimization suggestions; In the intelligent computing and analysis step, the parameter early warning model uses the long short-term memory network in the deep learning algorithm to model and analyze the operating parameters with time-series characteristics. The equipment health scoring mechanism uses a Bayesian network to build an equipment health status assessment model, describes the dependency relationship between various influencing factors through conditional probability distribution, and accurately assesses the health status of the equipment. In the early warning and control step, the generation of control commands adopts a fuzzy control algorithm, which transforms the optimization suggestions of operating parameters into specific control commands, and can handle the uncertainty and fuzziness in the control process.

[0034] Specifically, data acquisition and preprocessing involves cleaning, filtering, and standardizing the acquired data to remove noise and outliers, ensuring data purity and accuracy. Simultaneously, the data is converted into a unified standard format, providing a high-quality data foundation for subsequent intelligent analysis.

[0035] Intelligent computing engine: It adopts an intelligent computing engine with high real-time and high concurrency processing capabilities, supports the operation of a variety of intelligent algorithms, and can quickly process and analyze massive amounts of operational data.

[0036] Parameter Early Warning Model: This model utilizes Long Short-Term Memory (LSTM) networks from deep learning to model and analyze operational parameters with time-series characteristics. LSTM, through its unique gating mechanism (input gate, forget gate, and output gate), effectively processes time-series data, accurately capturing long-term dependencies and dynamic trends in parameters. In this way, abnormal parameter fluctuations can be identified in advance, issuing early warning signals to prevent parameter exceeding limits and causing malfunctions.

[0037] Equipment health scoring mechanism: A Bayesian network is used to construct an equipment health status assessment model. The Bayesian network describes the dependencies between various influencing factors through conditional probability distributions, and integrates multi-dimensional information such as equipment operating data and historical fault records to generate a real-time health score for each piece of equipment. When the score falls below a preset threshold, the system automatically triggers an alarm, reminding maintenance personnel to promptly inspect and repair the equipment, effectively reducing the equipment failure rate.

[0038] Based on the output of the intelligent analysis module, corresponding early warning signals are generated. These signals are promptly communicated to duty officers and maintenance personnel via audible and visual alarms, SMS notifications, etc., ensuring a rapid response to potential faults. A fuzzy control algorithm is employed to translate optimized operating parameters into specific control commands. This algorithm handles uncertainties and fuzziness in the control process, generating precise control commands through a fuzzy rule base and fuzzy inference mechanism, achieving automatic equipment control and ensuring the unit operates in optimal condition. Data interaction between internal modules and data sharing with external systems (such as digital maintenance systems and enterprise information management systems) are implemented, ensuring data real-time performance and consistency. Various data visualization methods (such as charts, curves, and maps) are used to intuitively display operating parameters, early warning information, and equipment health status on the user interface, facilitating real-time monitoring and operation by duty officers. Based on the intelligent analysis results, the early warning and control module generates early warning signals and notifies relevant personnel through various means. Simultaneously, it generates control commands based on optimization suggestions and sends them to the automated control system to achieve automatic equipment control.

[0039] During operation, the system continuously learns and accumulates operational data, automatically adjusting model parameters and optimization strategies to adapt to changes in the operating conditions of thermal power units.

[0040] To better demonstrate the application effects and technical advantages of the intelligent monitoring system and method for thermal power units based on artificial intelligence technology, the following are several practical application examples. Specific data comparisons verify the effectiveness of the invention. These examples cover thermal power units of different sizes and integrated applications of multiple units, fully demonstrating the significant achievements of the invention in improving the operational safety, economy, and automation level of thermal power units, as detailed below: Example 2: Application of Intelligent Monitoring System for 300MW Thermal Power Units This intelligent monitoring system is deployed in a 300MW thermal power unit. The data acquisition and preprocessing module connects to the unit's DCS system interface to collect various operating parameters in real time, such as temperature, pressure, flow rate, and equipment status. The intelligent computing engine is configured with high-performance servers and adopts a distributed computing architecture, supporting flexible configuration of single machines and computing clusters to meet the unit's computing needs. The intelligent analysis module, based on big data analysis and artificial intelligence algorithms, implements functions such as parameter early warning models, equipment health scoring mechanisms, energy efficiency and consumption difference analysis models, and intelligent control optimization strategies. The early warning and control module generates early warning signals and control commands, while the data interaction and visualization module enables data interaction and visualization both within and outside the system.

[0041] After the system was put into operation, it significantly impacted the operational safety and economy of thermal power units. Regarding operational safety, the parameter early warning model and equipment health scoring mechanism detected multiple potential faults in advance. For example, the parameter early warning model, through cluster analysis and neural network algorithms, accurately identified abnormal fluctuations in main steam pressure, issuing an early warning signal 15 minutes in advance. Operators made timely adjustments, preventing a potential unit trip due to pressure exceeding limits. The equipment health scoring mechanism generated a health score for a critical piece of equipment that consistently fell below a preset threshold, triggering an alarm. Maintenance personnel then inspected the equipment and discovered early signs of failure, enabling timely repairs and effectively reducing the equipment failure rate. In terms of operational economy, the energy efficiency and consumption difference analysis model and intelligent control optimization strategy optimized the unit's operating parameters. Statistics show that after implementing this system, the unit's coal consumption decreased by 3.2%, and energy utilization significantly improved. Simultaneously, the optimized closed-loop control for soot blowing reduced unnecessary soot blowing frequency, saving a significant amount of steam, and reducing steam consumption by 5.1%.

[0042] By applying this invention to a 300MW thermal power unit, the safety and economy of unit operation have been effectively improved. The parameter early warning model and equipment health scoring mechanism detect potential faults in advance, avoiding potential serious accidents and ensuring continuous and stable unit operation. The energy efficiency and consumption difference analysis model and intelligent control optimization strategy optimize unit operating parameters, significantly reducing coal and steam consumption and bringing considerable economic benefits to the power plant. Simultaneously, the system possesses self-learning and adaptive capabilities, automatically adjusting model parameters and optimization strategies according to changes in unit operating conditions to ensure long-term stable operation.

[0044] Example 3: Application of Intelligent Monitoring Method for 600MW Thermal Power Units This intelligent monitoring method was applied in a 600MW thermal power unit. Following the steps of the method, data acquisition and preprocessing were first performed. Various parameters during unit operation were collected in real time and then cleaned, filtered, and standardized. The preprocessed data was then input into the intelligent computing engine. In-depth analysis was conducted using parameter early warning models, equipment health scoring mechanisms, and energy efficiency and consumption difference analysis models within the intelligent analysis module, generating corresponding early warning signals and optimization suggestions. Based on the intelligent analysis results, the early warning and control module generated early warning signals and notified relevant personnel through various methods. Simultaneously, control commands were generated based on the optimization suggestions and sent to the automation control system to achieve automatic control of the equipment. During operation, the system continuously learns and accumulates operating data, automatically adjusting model parameters and optimization strategies to adapt to changes in unit operating conditions.

[0045] After applying this method, the unit's operational safety and economy were significantly improved. Regarding operational safety, the parameter early warning model uses a long short-term memory network to model and analyze operating parameters with time-series characteristics, identifying multiple potential faults in advance and providing operators with ample time to take action. The equipment health scoring mechanism uses a Bayesian network to accurately assess the equipment's health status, triggering alarms promptly and reminding maintenance personnel to inspect and repair. In terms of operational economy, the intelligent control optimization strategy, combined with intelligent energy consumption difference analysis and an optimization model, forms an optimal closed-loop control system. Based on real-time energy consumption data and equipment health status, it automatically adjusts production parameters, achieving intelligent control of the production process. Statistics show that after applying this method, the unit's coal consumption decreased by 4.1%, energy utilization improved, and the optimized soot blowing closed-loop control saved a significant amount of steam, reducing steam consumption by 6.3%.

[0046] Application in a 600MW thermal power unit demonstrates that the intelligent monitoring method of this invention effectively addresses the shortcomings of traditional monitoring models. Through parameter early warning models and equipment health scoring mechanisms, potential faults are detected in advance, avoiding the limitations of post-event alarms and improving the safety and reliability of unit operation. The intelligent control optimization strategy enables intelligent regulation of the production process, significantly reducing coal and steam consumption, resulting in significant economic benefits. Simultaneously, the system possesses self-learning and adaptive capabilities, automatically adjusting model parameters and optimization strategies according to changes in unit operating conditions, ensuring long-term stable operation. Furthermore, the intelligent monitoring method reduces the workload of operators and improves work efficiency.

[0048] Example 3: Integrated Application of Intelligent Monitoring System for Multiple Thermal Power Units This intelligent monitoring system is integrated into multiple thermal power units in a large power plant. The system achieves seamless integration with the DCS systems of each unit. The data acquisition and preprocessing module collects the operating parameters of each unit in real time and performs unified data cleaning, filtering, and standardization. The intelligent computing engine employs load balancing technology, dynamically allocating computing resources according to the task requirements of each unit, supporting flexible configuration of single machines and computing clusters. The intelligent analysis module, based on big data analysis and artificial intelligence algorithms, performs in-depth analysis of the operating parameters of each unit, realizing functions such as parameter early warning models, equipment health scoring mechanisms, energy efficiency and consumption difference analysis models, and intelligent control optimization strategies. The data interaction and visualization module enables data interaction between modules within the system, as well as data sharing with external systems (such as digital maintenance systems and enterprise information management systems). Through visualization technology, it displays the operating parameters, early warning information, and equipment health status of each unit in an intuitive way on the operating interface.

[0049] After system integration and application, it has had a significant impact on the overall operation management and economic benefits of the power plant. In terms of operation management, through data interaction and visualization modules, operators can monitor the operating status of multiple units in real time on a unified interface, facilitating centralized management and dispatch. Regarding operational safety and economy, the parameter early warning models and equipment health scoring mechanisms for each unit have proactively identified multiple potential faults, effectively preventing the escalation of faults. Energy efficiency and consumption difference analysis models and intelligent control optimization strategies have optimized the operating parameters of each unit, improving energy utilization. Statistics show that the power plant's overall coal consumption has decreased by 3.8%, and energy utilization has significantly improved. Simultaneously, the system's self-learning and adaptive capabilities ensure stable operation under different unit conditions and operating conditions.

[0050] The specific data comparison is shown in Table 3 below: Table 3.

[0051] Integrated application across multiple thermal power units demonstrates that the intelligent monitoring system of this invention effectively improves the overall operation and management level and economic benefits of the power plant. The system enables centralized monitoring and management of multiple units, facilitating operation and scheduling for staff. Parameter early warning models and equipment health scoring mechanisms detect potential faults in advance, preventing escalation and improving the safety and reliability of unit operation. Energy efficiency and consumption difference analysis models and intelligent control optimization strategies optimize the operating parameters of each unit, significantly reducing the power plant's overall coal consumption and resulting in significant economic benefits. Simultaneously, the system's self-learning and adaptive capabilities ensure stable operation under different unit conditions and operating circumstances, adapting to the complex environment of multiple units operating in a power plant.

[0052] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent monitoring system for thermal power units based on artificial intelligence technology, characterized in that, include: Data acquisition and preprocessing module: Connected to the DCS system interface, it is used to collect various parameters during the operation of thermal power units in real time, including temperature, pressure, flow rate, equipment status, etc., and to clean, filter and standardize the collected data to remove noise and abnormal data and ensure data quality. Intelligent computing engine: It has high real-time performance and high concurrency processing capabilities, supports the operation of various intelligent algorithms such as machine learning and deep learning, adopts computing load balancing technology, can dynamically allocate computing resources according to task requirements, and supports free configuration of single machines and computing clusters to meet the computing needs of thermal power units of different sizes. Intelligent Analysis Module: Based on big data analysis and artificial intelligence algorithms, it performs in-depth analysis of the collected operating parameters, and realizes functions such as parameter early warning model, equipment health scoring mechanism, energy efficiency and consumption difference analysis model, and intelligent control optimization strategy; Early warning and control module: Based on the output of the intelligent analysis module, it generates corresponding early warning signals and control commands. The early warning signals are promptly notified to the duty officer and maintenance personnel through audible and visual alarms, SMS notifications, etc., while the control commands are directly sent to the DCS system or other automated control systems to realize automatic control of the equipment. Data Interaction and Visualization Module: Enables data interaction between various modules within the system, as well as with external systems, displaying operating parameters, early warning information, equipment health status, and other information in an intuitive way on the user interface.

2. The intelligent monitoring system for thermal power units based on artificial intelligence technology according to claim 1, characterized in that, The parameter early warning model targets important operating parameters of megawatt-class thermal power units. Through algorithms such as cluster analysis and neural networks, it learns and analyzes historical and real-time data of the parameters to identify normal fluctuation ranges and abnormal trends, and issues early warning signals in advance to avoid faults caused by parameters exceeding limits.

3. The intelligent monitoring system for thermal power units based on artificial intelligence technology according to claim 1, characterized in that, The equipment health scoring mechanism comprehensively considers the equipment's operating data to generate a real-time health score for each device. By establishing an equipment health status assessment model and using data fusion technology, the system weights and calculates data from different dimensions to obtain the equipment's health score. When the health score is lower than a preset threshold, the system automatically triggers an alarm to remind maintenance personnel to check and repair in a timely manner.

4. The intelligent monitoring system for thermal power units based on artificial intelligence technology according to claim 1, characterized in that, The energy efficiency and energy consumption difference analysis model establishes a data-driven energy efficiency analysis model to accurately measure the operating efficiency and energy consumption of computer groups. It can analyze the energy consumption distribution of each link, identify potential energy-saving space, and provide optimization suggestions.

5. The intelligent monitoring system for thermal power units based on artificial intelligence technology according to claim 1, characterized in that, The intelligent control optimization strategy combines intelligent energy consumption difference analysis and optimization model to form an optimal closed-loop control system. Based on real-time energy consumption data, equipment health status and other information, it automatically adjusts production parameters to achieve intelligent control of the production process and improve production efficiency and energy utilization.

6. The intelligent monitoring system for thermal power units based on artificial intelligence technology according to claim 1, characterized in that, The intelligent computing engine adopts a distributed computing architecture, which can distribute computing tasks to multiple computing nodes for parallel processing.

7. The intelligent monitoring system for thermal power units based on artificial intelligence technology according to claim 1, characterized in that, The data interaction and visualization module supports a variety of data visualization methods, including charts, curves, and maps. Users can customize the displayed content and layout as needed.

8. A method for intelligent monitoring of thermal power units based on artificial intelligence technology, comprising the intelligent monitoring system for thermal power units based on artificial intelligence technology according to any one of claims 1-7, characterized in that, Includes the following steps: Data acquisition and preprocessing: Real-time acquisition of various parameters during the operation of thermal power units, followed by cleaning, filtering, and standardization. Intelligent computing and analysis: The pre-processed data is input into the intelligent computing engine, and in-depth analysis is performed through parameter early warning models, equipment health scoring mechanisms, energy efficiency and consumption difference analysis models in the intelligent analysis module to generate corresponding early warning signals and optimization suggestions; Early warning and control: Based on the intelligent analysis results, the early warning and control module generates early warning signals and notifies relevant personnel through various means. At the same time, it generates control commands based on optimization suggestions and sends them to the automated control system to realize automatic control of the equipment. Data interaction and visualization: Data interaction is carried out between various modules within the system, and information such as operating parameters, early warning information, and equipment health status are displayed to the duty officer through a visual interface for real-time monitoring and operation; Self-learning and self-adaptation: During operation, the system continuously learns and accumulates operational data, automatically adjusting model parameters and optimization strategies to adapt to changes in the operating conditions of thermal power units.

9. The intelligent monitoring method for thermal power units based on artificial intelligence technology according to claim 8, characterized in that, In the intelligent computing and analysis step, the parameter early warning model uses a long short-term memory network in deep learning algorithms to model and analyze operating parameters with time-series characteristics. The equipment health scoring mechanism uses a Bayesian network to construct an equipment health status assessment model, describes the dependency relationship between various influencing factors through conditional probability distribution, and accurately assesses the health status of the equipment. In the early warning and control step, the generation of control commands adopts a fuzzy control algorithm, which transforms the optimization suggestions of operating parameters into specific control commands, and can handle the uncertainty and fuzziness in the control process.

Images