Thermal power plant equipment operation state monitoring and early warning system and method based on PLC

Abstract

The embodiment of the invention relates to the field of equipment intelligent monitoring, and provides a thermal power plant equipment operation state monitoring and early warning system and method based on a PLC. Firstly, sensors are installed on key equipment of a thermal power plant to collect data such as boiler temperature, pressure, steam flow and equipment vibration amplitude; the method comprises the following steps of: acquiring data of a thermal power plant, transmitting the acquired data to a PLC (Programmable Logic Controller) through a communication network, transmitting the data to a thermal power plant equipment central monitoring server in batches by virtue of the PLC, analyzing the acquired data on the thermal power plant equipment central monitoring server to obtain an equipment monitoring result, and generating an early warning prompt according to the result. Therefore, effective monitoring and abnormal early warning of the equipment state can be realized, and the reliability and the safety of the thermal power plant equipment are further guaranteed.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent equipment monitoring, and more specifically, to a PLC-based thermal power plant equipment operation status monitoring and early warning system and method. Background Art

[0002] Thermal power plants are a vital component of electricity production, and their stable operation is crucial to the safety and economics of the power grid. However, due to the complex equipment and harsh operating environments (such as high temperatures and high voltages), equipment failures can lead not only to significant economic losses but also to serious safety incidents. Therefore, ensuring equipment reliability and safety has become a top priority in thermal power plant operations and management.

[0003] Patent CN117193087A discloses a method and device for monitoring equipment status in thermal power plants. This method uses multiple sensors to collect equipment operating parameters, then fuses these data using techniques such as Kalman filtering. This generates a fused value reflecting the health of the equipment, employs weighted averaging to improve assessment accuracy, and provides a visual display. When the corrected operating status value exceeds a preset threshold, the system automatically issues a fault warning.

[0004] This patent uses Kalman filtering technology and weighted averaging to calculate the estimated average value of the operating state and performs an early warning based on the comparison between the state deviation between the actual operating state (i.e., the operating state correction value) and the preset threshold. However, the Kalman filtering technology and weighted averaging method rely on preset parameters such as the state transfer matrix and covariance matrix, and assume that the system is linear or approximately linear. For complex nonlinear systems or scenarios with large dynamic changes, its adaptability and prediction accuracy may be limited. At the same time, it mainly focuses on the direct relationship and deviation between existing parameters, and lacks the ability to mine deep features hidden in the data. In addition, relying on preset thresholds to trigger early warnings may cause false alarms or missed alarms due to improper threshold settings, and due to the existence of smoothing effects, the response to sudden failures is not fast enough.

[0005] Therefore, an optimized thermal power plant equipment operation status monitoring and early warning solution is desired. Summary of the Invention

[0006] The present disclosure aims to solve at least one of the problems existing in the prior art and provide a PLC-based thermal power plant equipment operation status monitoring and early warning system and method.

[0007] One aspect of the present disclosure provides a PLC-based method for monitoring and early warning the operating status of thermal power plant equipment, comprising:

[0008] Deploy sensor components on key equipment in thermal power plants, wherein the sensor components are used to collect equipment operating data, including boiler temperature, boiler pressure, steam flow, and equipment vibration amplitude;

[0009] The sensor assembly transmits the device operation data to the PLC controller using a communication network;

[0010] The PLC controller transmits the equipment operation data in batches to the central monitoring server of the thermal power plant equipment;

[0011] At the central monitoring server for equipment in the thermal power plant, the equipment operating data is analyzed to obtain monitoring results, and based on the monitoring results, an early warning prompt is generated, including: performing time series arrangement on the equipment operating data and implicit correlation between parameters to obtain a time queue of implicit correlation feature vectors between equipment operating parameters; performing multiple gated time series feature dynamic propagation on the time queue of implicit correlation feature vectors between equipment operating parameters to obtain a time series significant propagation coding vector of equipment operating status features; obtaining the monitoring results based on the time series significant propagation coding vector of equipment operating status features, and generating the early warning prompt based on the monitoring results.

[0012] Optionally, performing time sequence arrangement on the device operation data and implicit correlation between parameters to obtain a time queue of implicit correlation feature vectors between device operation parameters includes:

[0013] Arranging the device operation data according to a time step dimension to obtain a time queue of the device operation data;

[0014] Each device operation data in the time queue of the device operation data is input into a device operation parameter correlation feature extractor based on a fully connected layer to obtain a time queue of implicit correlation feature vectors between the device operation parameters.

[0015] Optionally, performing multiple gated timing feature dynamic propagation on the time queue of the implicit correlation feature vectors between the device operating parameters to obtain the device operating state feature time series significant propagation coding vector includes:

[0016] Performing Poincare space mapping on each implicit correlation feature vector between device operating parameters in the time queue of implicit correlation feature vectors between device operating parameters to obtain a time queue of implicit correlation heterogeneous modulation coding feature vectors between device operating parameters;

[0017] Performing feature sequence encoding on the time queue of the implicitly associated heterogeneous modulation coding feature vectors between the device operating parameters to obtain a heterogeneous propagation representation vector of the implicitly associated feature between the device operating parameters;

[0018] Calculate the significant propagation representation between the time queue of the implicit correlation feature heterogeneous propagation representation vector between the device operating parameters and the implicit correlation heterogeneous modulation coding feature vector between the device operating parameters to obtain the device operating status feature time series significant propagation coding vector.

[0019] Optionally, the time queue of the implicitly associated heterogeneous modulation coding feature vectors between the device operating parameters is subjected to feature sequence encoding to obtain the heterogeneous propagation representation vector of the implicitly associated features between the device operating parameters, including: forward LSTM sequence encoding of the time queue of the implicitly associated heterogeneous modulation coding feature vectors between the device operating parameters to obtain the heterogeneous propagation representation vector of the implicitly associated features between the device operating parameters.

[0020] Optionally, calculating a significant propagation representation between a time sequence of a heterogeneous propagation representation vector of an implicit correlation feature between the device operating parameters and a heterogeneous modulation coding feature vector of an implicit correlation feature between the device operating parameters to obtain a significant propagation coding vector of a temporal sequence of the device operating state features includes:

[0021] Calculating, based on the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters, the modulation weight value of each heterogeneous modulation and coding feature vector of the implicit correlation feature between the device operating parameters in the time queue of the heterogeneous modulation and coding feature vector of the implicit correlation feature between the device operating parameters to obtain the time queue of the modulation weight of the implicit correlation transmission between the device operating parameters;

[0022] Based on the time queue of the modulation weights of the implicit correlation transmission between the device operating parameters, the time queue of the implicit correlation heterogeneous modulation coding feature vector between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation characteristics between the device operating parameters are weighted compensated and fused to obtain the device operating status feature time series significant propagation coding vector.

[0023] Optionally, based on the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters, calculating the modulation weight value of each heterogeneous modulation and coding feature vector of the implicit correlation feature between the device operating parameters in the time queue of the heterogeneous modulation and coding feature vector of the implicit correlation between the device operating parameters to obtain the time queue of the modulation weight of the implicit correlation transmission between the device operating parameters includes:

[0024] Calculating the semantic correlation between each implicitly associated heterogeneous modulation and coding feature vector between device operating parameters in the time queue of the implicitly associated heterogeneous modulation and coding feature vector between device operating parameters and the heterogeneous propagation representation vector of the implicitly associated feature between device operating parameters to obtain a time queue of the semantic correlation between the implicitly associated device operating parameters;

[0025] Calculating the semantic space jump value of each implicitly associated heterogeneous modulation and coding feature vector between device operating parameters in the time queue of the implicitly associated heterogeneous modulation and coding feature vector between the device operating parameters relative to the heterogeneous propagation representation vector of the implicitly associated feature between the device operating parameters to obtain a time queue of the semantic space jump value of the implicitly associated semantic space between the device operating parameters;

[0026] Multiple gated information transmission modulation is performed on the time queue of the implicit association semantic association degree between the device operation parameters and the time queue of the implicit association semantic space jump value between the device operation parameters to obtain the time queue of the implicit association transmission modulation weight between the device operation parameters.

[0027] Optionally, based on the time queue of the modulation weights of the implicit correlation transfer between the device operating parameters, weighted compensation fusion is performed on the time queue of the implicit correlation heterogeneous modulation coding feature vector between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters to obtain the device operating state feature time series significant propagation coding vector, including:

[0028] Using the time sequence of the modulation weights of the implicit correlation transmission between the device operating parameters as a weight sequence, calculating the position-weighted sum of the time sequence of the heterogeneous modulation coding feature vectors of the implicit correlation between the device operating parameters to obtain a significant propagation representation vector of the implicit correlation feature between the device operating parameters;

[0029] Performing a residual connection on the significant propagation representation vector of the implicit correlation feature between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters to obtain a compensation representation vector of the significant propagation of the heterogeneous feature of the implicit correlation between the device operating parameters;

[0030] The implicit correlation heterogeneous feature significant propagation compensation representation vector between the equipment operating parameters is mapped to the Euclidean space to obtain the equipment operating state feature time series significant propagation coding vector.

[0031] Optionally, obtaining the monitoring result based on the significant propagation coding vector of the characteristic time series of the equipment operation status, and generating the early warning prompt based on the monitoring result, includes:

[0032] Inputting the equipment operating state feature time series significant propagation coding vector into a classifier-based operating state monitoring module to obtain the monitoring result, which is used to indicate whether the operating state of the key equipment of the thermal power plant is abnormal;

[0033] If the monitoring result indicates that the working status of the key equipment of the thermal power plant is abnormal, the early warning prompt is generated.

[0034] Another aspect of the present disclosure provides a PLC-based thermal power plant equipment operation status monitoring and early warning system, comprising:

[0035] An equipment operation data acquisition module is used to deploy sensor components on key equipment in a thermal power plant. The sensor components are used to collect equipment operation data, including boiler temperature, boiler pressure, steam flow, and equipment vibration amplitude values;

[0036] A device operation data transmission module, used for the sensor component to transmit the device operation data to the PLC controller using a communication network;

[0037] The central monitoring server receives data modules, which are used for the PLC controller to transmit the equipment operation data in batches to the central monitoring server of the thermal power plant equipment;

[0038] The equipment operation data analysis module is used to analyze the equipment operation data in the central monitoring server of the thermal power plant equipment to obtain monitoring results, and generate early warning prompts based on the monitoring results. The equipment operation data analysis module includes: an equipment operation data encoding unit, which is used to perform time sequence arrangement and implicit correlation between parameters of the equipment operation data to obtain a time queue of implicit correlation feature vectors between equipment operation parameters; an equipment operation feature dynamic propagation unit, which is used to perform multiple gated time sequence feature dynamic propagation on the time queue of implicit correlation feature vectors between equipment operation parameters to obtain equipment operation status feature time sequence significant propagation encoding vectors; an equipment monitoring result generation unit, which is used to obtain the monitoring result based on the equipment operation status feature time sequence significant propagation encoding vectors, and generate the early warning prompts based on the monitoring result.

[0039] Optionally, the device monitoring result generating unit is configured to:

[0040] Inputting the equipment operating state feature time series significant propagation coding vector into a classifier-based operating state monitoring module to obtain the monitoring result, which is used to indicate whether the operating state of the key equipment of the thermal power plant is abnormal;

[0041] If the monitoring result indicates that the working status of the key equipment of the thermal power plant is abnormal, the early warning prompt is generated.

[0042] Compared to existing technologies, this system first installs sensors on key thermal power plant equipment to collect data such as boiler temperature, pressure, steam flow, and equipment vibration amplitude. The collected data is then transmitted to a PLC controller via a communication network. The PLC controller then sends the data in batches to a central monitoring server for the thermal power plant equipment. The central monitoring server analyzes the collected data to obtain equipment monitoring results and generates early warning alerts based on the results. This allows for effective monitoring of equipment status and abnormality warnings, thereby ensuring the reliability and safety of thermal power plant equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings, and these exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0044] Figure 1 Flowchart of a method for monitoring and early warning the operating status of thermal power plant equipment based on PLC according to an embodiment of the present disclosure;

[0045] Figure 2 Flowchart of step S140 in the PLC-based thermal power plant equipment operating status monitoring and early warning method according to an embodiment of the present disclosure;

[0046] Figure 3 Flowchart of step S142 in the PLC-based thermal power plant equipment operating status monitoring and early warning method according to an embodiment of the present disclosure;

[0047] Figure 4 Flowchart of step S1423 in the PLC-based thermal power plant equipment operating status monitoring and early warning method according to an embodiment of the present disclosure;

[0048] Figure 5 4 is a system block diagram of a PLC-based thermal power plant equipment operation status monitoring and early warning system according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. However, it will be understood by those skilled in the art that in each embodiment of the present disclosure, many technical details are provided to enable readers to better understand the present disclosure. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present disclosure can be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present disclosure. The various embodiments can be combined and referenced with each other under the premise that there is no contradiction.

[0050] The stable operation of thermal power plant equipment is crucial to the safety and economy of the power grid. Patent CN117193087A proposes a method and device for monitoring the equipment status of thermal power plants. It collects data through sensors, uses Kalman filtering and weighted averaging technology to fuse the data to assess the health of the equipment, and displays it visually. When the correction value of the equipment operating status exceeds the preset threshold, the system will issue a fault warning. However, this method relies on preset parameters and is suitable for linear or approximately linear systems. It has limited adaptability to scenarios with nonlinear or large dynamic changes, and lacks the ability to mine deep features. In addition, reliance on preset thresholds may lead to false alarms or missed alarms, and the response to sudden failures is not fast enough.

[0051] Based on this, the present disclosure proposes a PLC-based thermal power plant equipment operating status monitoring and early warning method. Figure 1 FIG. 1 is a flow chart of a method for monitoring and warning the operation status of thermal power plant equipment based on PLC according to an embodiment of the present disclosure. Figure 1 As shown, the PLC-based thermal power plant equipment operation status monitoring and early warning method according to the embodiment of the present disclosure includes: S110, deploying a sensor component on the key equipment of the thermal power plant, the sensor component is used to collect equipment operation data, the equipment operation data including boiler temperature, boiler pressure, steam flow and equipment vibration amplitude value; S120, the sensor component uses a communication network to transmit the equipment operation data to the PLC controller; S130, the PLC controller transmits the equipment operation data in batches to the thermal power plant equipment central monitoring server; S140, on the thermal power plant equipment central monitoring server, analyzing the equipment operation data to obtain monitoring results, and generating early warning prompts based on the monitoring results.

[0052] In step S110, sensor assemblies are deployed on key equipment in a thermal power plant. These sensor assemblies collect equipment operating data, including boiler temperature, boiler pressure, steam flow rate, and equipment vibration amplitude. It should be understood that the boiler is a core piece of equipment in a thermal power plant, and its temperature is directly related to combustion efficiency, steam production, and safe equipment operation. Excessively high temperatures can damage equipment materials and increase scaling, shortening boiler lifespan; excessively low temperatures can affect steam quality and power generation efficiency. Boiler pressure is another key parameter, closely related to steam generation, transportation, and the operation of the entire thermal system. Excessively high pressure can cause serious safety incidents such as explosions, while excessively low pressure can result in an inability to meet power generation requirements. Steam flow rate reflects the boiler's steam production capacity and the distribution and utilization of steam throughout the system. Abnormal variations in steam flow rate indicate issues such as unstable boiler combustion, varying turbine loads, or pipeline blockages. Thermal power plant equipment vibrates during operation, and vibration amplitude is an important indicator of the equipment's mechanical condition. Excessive vibration amplitudes can indicate equipment imbalance, bearing wear, or loose foundations. These issues can severely impact equipment reliability and lifespan, and may even cause equipment failure. Therefore, collecting equipment vibration amplitude values can promptly identify mechanical anomalies. Overall, these data reflect the overall operating status of critical power plant equipment. Comprehensive analysis of this data allows for a more comprehensive and accurate assessment of any anomalies in the operating status of critical power plant equipment, enabling timely identification of potential issues and implementation of appropriate measures to ensure safe and stable equipment operation.

[0053] Deploying sensor components on key equipment in thermal power plants to collect accurate operational data is the foundation for effective monitoring and early warning of thermal power plant equipment. The following is a detailed description of its implementation:

[0054] The first step is to select the sensor and determine its installation location. For boiler temperature measurement, a sensor suitable for high-temperature environments and with high accuracy is required, such as a K-type thermocouple. This type of thermocouple has a wide temperature measurement range and is suitable for high-temperature environments like boilers.

[0055] When it comes to boiler pressure monitoring, high-precision, high-stability piezoresistive pressure sensors are the ideal choice. They accurately measure the high-pressure environment within the boiler and are typically installed at key pressure monitoring points, such as the boiler drum and main steam line.

[0056] Measuring steam flow relies on a suitable flowmeter, such as a differential pressure flowmeter or a vortex flowmeter. Differential pressure flowmeters have a proven measurement principle, while vortex flowmeters exhibit minimal pressure loss when used under steam conditions. After selecting an appropriate flowmeter based on the steam flow range and pipe size, install it on the steam pipe.

[0057] Accelerometers or displacement sensors are commonly used to monitor equipment vibration amplitude. Accelerometers are sensitive to high-frequency vibrations and excel at detecting small initial vibration changes in equipment, making them suitable for early fault warning. Displacement sensors, on the other hand, focus on measuring shaft vibration in large rotating equipment and accurately reflect the mechanical condition of the equipment. These sensors can be installed in areas such as bearing seats and housings.

[0058] Sensor wiring and protection are also critical components of the implementation process. Wiring planning requires careful design of routing paths, ensuring that sensor signal cables and power cables are laid separately to effectively prevent signal interference. Shielded signal cables are used to minimize the impact of external electromagnetic interference on measurement data, ensuring data accuracy. During the wiring process, cable protection measures must be fully considered, with targeted treatments tailored to specific environmental factors. For example, in high-temperature areas, cables must be equipped with heat-resistant casing to prevent damage from the heat. In corrosive environments, the cable sheath should be corrosion-resistant to prevent chemical attack.

[0059] Properly configuring the junction box is crucial for the transfer and centralized management of sensor signals. The junction box should be located in a suitable location and properly sealed to prevent the ingress of impurities such as dust and moisture, which can affect signal transmission quality and even endanger equipment safety. Wiring within the junction box must be performed strictly in accordance with specifications, ensuring secure and reliable connections and employing the correct wiring method to ensure accurate transmission of sensor signals to subsequent equipment.

[0060] Calibration and maintenance are essential to ensure the accuracy of sensor measurement data. Before installing a sensor, it should be rigorously calibrated using a standard signal source or by comparing it to a measuring device of known accuracy. Regular calibration is also required during use. For example, temperature sensors can be calibrated using a standard temperature source; pressure sensors can be calibrated using a high-precision pressure calibrator; and steam flow sensors require calibration using a flow standard device based on their type and accuracy requirements.

[0061] At the same time, establish a comprehensive routine sensor maintenance system. Regularly check the sensor's secure installation, inspect for external damage, and check for loose connections. For sensors susceptible to contamination, such as steam flow sensors, regularly clean the sensor probe to prevent dirt accumulation that can affect measurement accuracy. Additionally, check the integrity of the sensor's protective measures. If any damage is found, repair or replace it promptly to ensure the sensor is always in good working condition.

[0062] Finally, the data collection and transmission phase begins. The sensor data collection frequency should be appropriately set based on the equipment's operating characteristics and monitoring requirements. For relatively slowly changing parameters, such as temperature and pressure, the collection frequency can be appropriately reduced to meet monitoring requirements while reducing the burden of data transmission and processing. For more dynamic parameters, such as vibration, the collection frequency should be higher, as these parameters can quickly reflect changes in the equipment's operating status. This allows for the timely capture of subtle changes. The sensor data is transmitted to the PLC controller via a communications network. Wired communication methods, such as industrial Ethernet or RS485 bus, are preferred. This ensures stable and reliable data transmission, preventing data loss and interference during transmission. In some specialized environments or where mobility is required, wireless communication methods such as Wi-Fi or ZigBee may be considered as alternatives. However, wireless signal interference and coverage issues must be carefully considered, and appropriate measures must be taken to ensure data quality.

[0063] Through the above series of rigorous implementation steps, it is possible to ensure that the sensor components on the key equipment of the thermal power plant can accurately and reliably collect operating data, providing a solid data foundation for subsequent equipment operation status monitoring and early warning, thereby effectively ensuring the reliability and safety of thermal power plant equipment and improving the overall operating efficiency and stability of the thermal power plant.

[0064] In step S120, the sensor assembly transmits the equipment operation data to the PLC controller via the communication network. It should be understood that PLCs (programmable logic controllers) possess powerful data processing capabilities. The equipment operation data collected by the sensor assembly is diverse and generated in real time, such as boiler temperature, pressure, steam flow, and equipment vibration amplitude. Once this data is transmitted to the PLC controller, the PLC can perform preliminary data processing, such as data format conversion and filtering, to make the data more standardized and accurate, facilitating subsequent transmission and analysis. For example, the analog signals collected by the sensors can be converted to digital signals in the PLC to remove some interference noise and improve data quality. Furthermore, considering that the communication network may be subject to interference or failure during data transmission, the PLC controller can serve as a data cache and backup node. Specifically, when the communication network is temporarily interrupted, the PLC temporarily stores the sensor data and resumes transmission after the network is restored, ensuring that no data is lost. Furthermore, the PLC itself has high reliability and stability, enabling continuous and stable operation in complex industrial environments. This helps ensure the continuity of data transmission, thereby improving the reliability of the entire monitoring and early warning system.

[0065] In step S130, the PLC controller transmits the equipment operation data in batches to the thermal power plant equipment central monitoring server. It should be understood that the thermal power plant equipment central monitoring server possesses powerful computing and data storage capabilities. By transmitting the collected equipment operation data to the thermal power plant equipment central monitoring server, it is possible to centrally store and manage large amounts of equipment operation data. This data can also be effectively analyzed and mined to better predict potential equipment failures, thereby providing a basis for preventive maintenance. Accordingly, given that thermal power plant equipment operation data is generated in real time and in large volumes, transmitting all of the data at once could place a heavy load on the communication network, leading to network congestion, transmission delays, and even data loss. The PLC controller transmits the equipment operation data in batches, effectively balancing network traffic and ensuring stable and reliable data transmission. Furthermore, the transmitted data can be stored and managed in batches on the server, facilitating subsequent data query, tracing, and analysis. For example, when searching for changes in equipment operation data within a certain time period, relevant data can be quickly located based on batches, facilitating troubleshooting and operating status assessment, thereby improving the efficiency and accuracy of data management.

[0066] The PLC controller transmits equipment operation data in batches to the central monitoring server of the thermal power plant equipment. This implementation process requires comprehensive consideration of multiple aspects to ensure the efficiency, stability and reliability of data transmission. The following is a detailed explanation of its implementation process:

[0067] After receiving equipment operating data from sensor components, the PLC controller first establishes a data buffering mechanism. It maintains an internal buffer area of a certain capacity. The size of this buffer area is not arbitrarily set but is carefully calculated and considered. It requires a comprehensive balance of factors, including the rate at which equipment operating data is generated, the overall data volume, and network transmission capacity. For example, if critical equipment in a thermal power plant is operating at high load, generating data at a high rate while network bandwidth is relatively limited, a relatively large buffer area is necessary to prevent data overflow and loss during batch transmission due to insufficient buffer space. Simultaneously, while buffering data, the PLC controller simultaneously initiates data preprocessing. One of the key preprocessing tasks is uniform data format conversion. Sensor data often exists in analog form, while subsequent data transmission and processing typically require a digital signal format. Therefore, the PLC controller uses its built-in analog-to-digital conversion module to accurately convert analog signals into digital signals, ensuring that the data format meets the overall system communication and processing specifications. Data filtering is also essential. The complex environment of thermal power plants is subject to numerous interference factors, such as electromagnetic interference and equipment vibration. These factors can cause errors in sensor measurement data and introduce noise. PLC controllers use digital filtering algorithms, such as mean and median filtering, to process collected data, effectively removing noise interference and significantly improving data accuracy and reliability, making the data transmitted to the central monitoring server more valuable.

[0068] Batch division strategies play a key role in guiding the data transmission process. Dividing data batches based on time intervals is a common and effective method. For example, if every three minutes, the PLC controller transmits the data accumulated in the buffer during that time as a complete data batch. This strategy ensures data continuity over time, facilitating in-depth time series data analysis on the central monitoring server. In practice, the PLC controller maintains a precise time counter that starts counting from the moment data begins caching. When the timer reaches the pre-set three-minute interval, the data batch transmission is triggered. Another possible method for batch division is based on data volume. Depending on network load and server processing capabilities, when the data volume in the buffer reaches a certain size, such as 800 records, the PLC controller organizes the data into a single batch for transmission. This approach effectively avoids excessive network pressure from transmitting large amounts of data all at once and ensures stable network transmission. Furthermore, on the server side, dividing batches by data volume facilitates orderly data storage and management, facilitating efficient subsequent data query, retrieval, and analysis. The PLC controller monitors the changes in the data volume in the cache in real time, and once the data volume reaches the set threshold, it quickly starts the data batch transmission process.

[0069] The selection and configuration of transmission protocols are crucial to the quality and efficiency of data transmission. In the industrial environment of thermal power plants, industrial Ethernet protocols are widely used due to their superior performance. Industrial Ethernet protocols such as Profinet and Ethernet / IP offer significant advantages, such as high-speed transmission, strong stability, and high reliability, perfectly meeting the demanding requirements of massive data transmission in industrial sites. During implementation, technicians must carefully configure the Ethernet communication modules for PLC controllers and central monitoring servers. This requires accurate configuration of a series of network parameters, such as assigning unique IP addresses to devices to ensure clear identification within the network; setting appropriate subnet masks to divide network areas for effective network management; and configuring gateway addresses to ensure smooth data exchange between different network areas. Only by ensuring the correct configuration of these network parameters can the PLC controller and central monitoring server achieve stable and efficient communication within the same network environment. Furthermore, to further optimize data transmission, fine-tuning of relevant communication protocol parameters is necessary. For example, packet size adjustment requires a comprehensive consideration of network bandwidth and device performance. If the network bandwidth is high and the device performance is strong, the packet size can be appropriately increased and the number of packets can be reduced, thereby reducing the header overhead of data transmission and improving transmission efficiency. Conversely, if network conditions or device performance are limited, the packet size should be appropriately reduced to avoid delays or packet loss caused by excessively large packets during transmission. Properly setting the number of transmission retries and timeouts is also essential. If an error occurs during data transmission or a timeout occurs without a response, the PLC controller automatically retransmits according to the preset number of retries to ensure data integrity. For example, the number of retries can be set to three, with a certain interval between each retry. If the transmission is still unsuccessful after three retries, the transmission error information is recorded for reference by operations and maintenance personnel during subsequent troubleshooting.

[0070] During data transmission, the PLC controller must also conduct real-time, comprehensive monitoring of the transmission status. During data transmission, it closely monitors the progress of data transmission and accurately determines the ratio of the amount of data sent to the total data volume. It also monitors the response of the receiving party (central monitoring server), including whether the acknowledgment signal is returned promptly. Once data is successfully transmitted to the central monitoring server, the server sends a confirmation signal to the PLC controller as agreed upon, indicating that the data has been fully received. If the PLC controller does not receive the acknowledgment signal within the specified timeout period, it will proceed according to a pre-defined policy. This may include immediately initiating a data retransmission to ensure data is not lost. Furthermore, detailed transmission error information, including key information such as the batch number, transmission time, and error type, is recorded. This allows operations and maintenance personnel to quickly identify the root cause of the problem and implement effective remediation measures during subsequent troubleshooting. Furthermore, the PLC controller can record and compile data transmission status information, such as the number of batches transmitted, the number of transmission failures, and the average transmission delay. This information provides operations and maintenance personnel with a comprehensive understanding of the data transmission process, helping to promptly identify potential transmission issues and proactively implement optimization measures to ensure continuous and stable data transmission.

[0071] On the central monitoring server for thermal power plant equipment, a dedicated data reception service program is deployed to smoothly receive data batches transmitted by PLC controllers. This program is responsible for monitoring data transmission requests from PLC controllers and promptly receiving and processing incoming data. Received data is stored in an organized manner according to a carefully designed storage structure, typically using a mature database management system (such as SQL Server or Oracle) for efficient data storage and management. A reasonable data table structure is constructed based on key information such as data type (such as temperature, pressure, flow), acquisition time, and so on, ensuring that data is stored in a logically clear manner that is easy to query and retrieve. For example, for boiler temperature data, a data table containing fields such as temperature value, acquisition time, and equipment number is created. This facilitates rapid retrieval of required data based on various query requirements (such as querying by time range or by device). During data storage, the server also performs rigorous data integrity verification. Verification methods vary, and data length checking is a fundamental and effective method. By comparing the received data length with the expected data length, a preliminary determination of data integrity is made. Checksum calculation is also a common method. A checksum is calculated on the received data and compared with the checksum provided by the sender. A discrepancy between the two indicates a possible data transmission error. Upon discovering incomplete or erroneous data, the server immediately sends a request to the PLC controller to retransmit the corresponding data batch, ensuring the accuracy of the data stored on the server. Furthermore, the server can perform further data processing, such as data cleaning operations to remove duplicate records and outliers, further improving data quality and providing a solid foundation for subsequent complex data analysis and accurate equipment status monitoring.

[0072] In step S140, the equipment operation data is analyzed on the central monitoring server of the thermal power plant equipment to obtain monitoring results, and an early warning prompt is generated based on the monitoring results.

[0073] Accordingly, in the thermal power plant equipment central monitoring server, the equipment operating data is analyzed to obtain monitoring results, and based on the monitoring results, early warning prompts are generated. The technical concept of the present disclosure is to use AI-based data analysis and coding technology to arrange the equipment operating data in the time dimension, and then mine the implicit correlation features between the operating parameters in the equipment operating data at each time point after arrangement. In this way, based on the multiple gated time series propagation representations between the implicit correlation features between the operating parameters of each equipment, it is possible to intelligently judge whether there are any abnormalities in the working status of the key equipment of the power plant, and generate corresponding early warning prompts. In this way, the complex nonlinear time series dynamic changes between the operating parameters can be effectively captured, providing higher prediction accuracy. Moreover, instead of relying on fixed preset thresholds, the early warning standards are dynamically adjusted through intelligent algorithms, so that a rapid response to sudden failures can be made in a short period of time, thereby improving the effectiveness and reliability of the early warning.

[0074] Figure 2 FIG. 1 is a flow chart of step S140 in the PLC-based thermal power plant equipment operation status monitoring and early warning method according to an embodiment of the present disclosure. Figure 2 As shown, the step S140 includes: S141, performing time series arrangement and implicit correlation between parameters of the equipment operation data to obtain a time queue of implicit correlation feature vectors between equipment operation parameters; S142, performing multiple gated time series feature dynamic propagation on the time queue of implicit correlation feature vectors between equipment operation parameters to obtain a time series significant propagation coding vector of equipment operation status feature; S143, obtaining the monitoring result based on the time series significant propagation coding vector of equipment operation status feature, and generating the early warning prompt based on the monitoring result.

[0075] In step S141, the device operation data is time-sequentially arranged and implicitly associated with parameters to obtain a time queue of implicitly associated feature vectors between device operation parameters. Specifically, in the disclosed embodiment, step S141 includes: arranging the device operation data according to the time step dimension to obtain a time queue of device operation data; and inputting each device operation data in the time queue of device operation data into a device operation parameter association feature extractor based on a fully connected layer to obtain a time queue of implicitly associated feature vectors between device operation parameters.

[0076] It should be understood that the equipment operation data is constantly changing as time progresses, that is, the equipment operation data has characteristics in the time dimension. Therefore, in order to ensure that all collected data points are correctly arranged in the order of their occurrence, so that the time dependency of each time point can be captured and mined more conveniently and accurately in the future, in the technical solution of the present disclosure, the equipment operation data is arranged according to the time step dimension to obtain a time queue of the equipment operation data. In this way, time series analysis methods such as trend analysis, periodicity detection and outlier identification can be more conveniently applied, which helps to understand the long-term behavior and development trend of the equipment.

[0077] Next, considering that the equipment operation data of a thermal power plant usually contains multiple parameters (such as temperature, pressure, flow, etc.), there are implicit correlations and influence relationships between these parameters. Based on this, in order to capture those implicit correlation features that are not easy to be intuitively identified, so as to more comprehensively understand the working status of the equipment, in the technical solution disclosed in the present invention, each device operation data in the time queue of the equipment operation data is input into the equipment operation parameter correlation feature extractor based on the fully connected layer to obtain a time queue of implicit correlation feature vectors between the equipment operation parameters. It is worth mentioning that the fully connected layer is a type of layer in a neural network, in which each input node is directly connected to each output node of the next layer, and each connection has a weight parameter. The fully connected layer can process high-dimensional input data and learn the complex interactions between input features (various parameters of the equipment at each time point) to reflect the patterns and trends that change over time. For complex systems such as thermal power plants, the deep features between the various equipment parameters often contain important information about the health status of the equipment. By inputting the operating data into the feature extractor time by time, the implicit correlation features between the equipment operating parameters at each time point can be obtained. These features are arranged in chronological order to form a new time queue, providing a structured and information-rich time series data set for subsequent analysis.

[0078] In step S142, the time sequence of implicit correlation feature vectors between the device operating parameters is subjected to multi-gated dynamic propagation of temporal features to obtain a temporal significant propagation coding vector of the device operating state features. Specifically, the implicit correlation feature vectors between the various device operating parameters express the inherent implicit correlation between the various device operating parameters at a given time point, and the features at each time point have different weights and influences across the entire time domain. Therefore, in order to effectively capture the temporal dependencies between device operating parameters over long and short time spans, and to more accurately capture the interactions between different parameters and their impact on overall system performance, the technical solution disclosed herein uses multi-gated dynamic propagation of temporal features to process the time sequence of implicit correlation feature vectors between the device operating parameters to obtain a temporal significant propagation coding vector of the device operating state features. That is, by introducing the concept of multi-gating, the network can identify which time points or operating parameters have a particularly significant impact on the current device state. The generated significant propagation coding vector not only contains rich temporal information transmission features but also emphasizes the importance of key features, making the subsequent decision-making process more targeted.

[0079] Specifically, Figure 3 FIG. 1 is a flow chart of step S142 in the PLC-based thermal power plant equipment operation status monitoring and early warning method according to an embodiment of the present disclosure. Figure 3 As shown, the step S142 includes: S1421, performing Poincare space mapping on each implicit correlation feature vector between the device operating parameters in the time queue of the implicit correlation feature vector between the device operating parameters to obtain the time queue of the implicit correlation heterodomain modulation coding feature vector between the device operating parameters; S1422, performing feature sequence encoding on the time queue of the implicit correlation heterodomain modulation coding feature vector between the device operating parameters to obtain the implicit correlation feature heterodomain propagation representation vector between the device operating parameters; S1423, calculating the significant propagation representation between the implicit correlation feature heterodomain propagation representation vector between the device operating parameters and the time queue of the implicit correlation heterodomain modulation coding feature vector between the device operating parameters to obtain the device operating state feature time series significant propagation coding vector.

[0080] Specifically, first, each implicit correlation feature vector between device operating parameters in the time queue of implicit correlation feature vectors between device operating parameters is subjected to Poincare space mapping to obtain a time queue of implicit correlation heterogeneous modulation coding feature vectors between device operating parameters. The above process can be expressed as:

[0081] V={v1,v2,...,v i ,...,v n}

[0082] hi =W1v i W2

[0083] H={h1,h2,...,h i ,...,h n}

[0084] Where V is the time sequence of the implicit correlation feature vectors between the operating parameters of the equipment, v1, v2, v i and v n are the 1st, 2nd, i-th and b-th implicit correlation feature vectors between the equipment operating parameters in the time queue of the implicit correlation feature vectors between the equipment operating parameters, W1 and W2 are the first weight matrix and the second weight matrix, h1, h2, h i and h n are respectively the 1st, 2nd, i-th and n-th device operating parameters implicitly associated foreign-domain modulation and coding feature vectors in the time queue of the device operating parameters implicitly associated foreign-domain modulation and coding feature vectors, and H is the time queue of the device operating parameters implicitly associated foreign-domain modulation and coding feature vectors.

[0085] It should be understandable that by mapping the implicit correlation feature vectors between the operating parameters of each device in the time queue of the implicit correlation feature vectors between the operating parameters of the device to the Poincare space, the unique geometric properties of the Poincare hyperbolic space can be utilized to effectively capture the nonlinear relationship and hierarchical structure between the characteristics of the device operating parameters, that is, the semantic association between the implicit correlation features between the operating parameters of the device at different time nodes. This helps to understand the operating status of the device more comprehensively and deeply, and thus achieve more accurate fault prediction.

[0086] Specifically, in the embodiment of the present disclosure, step S1422 includes: performing forward LSTM sequence encoding on the time queue of the implicit correlation heterogeneous modulation coding feature vectors between the device operating parameters to obtain the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters. The above process can be expressed as:

[0087]

[0088] Where H is the time sequence of the implicitly associated heterogeneous modulation coding feature vectors between the device operating parameters, is the forward LSTM encoding, h n+1 It is the heterogeneous propagation representation vector of the implicit correlation characteristics between the operating parameters of the equipment.

[0089] It should be understood that by performing forward LSTM sequence encoding on the temporal queue of the implicitly correlated heterogeneous modulation coding feature vectors between the equipment operating parameters, the forward LSTM network's ability to capture long-term dependencies in the input sequence can be leveraged to better understand the temporal dependency characteristics between equipment operating parameters, thereby building a more effective thermal power plant equipment operating status monitoring and early warning system. When conditions such as abnormal boiler temperature fluctuations or abnormal equipment vibration amplitude occur, the encoded long-term dependency characteristics, which differ from those in the normal state, can be used to more quickly and accurately identify abnormal equipment conditions.

[0090] Specifically, Figure 4 FIG. 1 is a flow chart of step S1423 in the PLC-based thermal power plant equipment operation status monitoring and early warning method according to an embodiment of the present disclosure. Figure 4 As shown, the step S1423 includes: S14231, based on the heterogeneous propagation representation vector of the implicit correlation characteristics between the device operating parameters, calculating the modulation weight value of each heterogeneous modulation coding feature vector of the implicit correlation characteristics between the device operating parameters in the time queue of the heterogeneous modulation coding feature vector of the implicit correlation between the device operating parameters to obtain the time queue of the modulation weight of the implicit correlation transmission between the device operating parameters; S14232, based on the time queue of the modulation weight of the implicit correlation transmission between the device operating parameters, performing weighted compensation fusion on the time queue of the heterogeneous modulation coding feature vector of the implicit correlation between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation characteristics between the device operating parameters to obtain the device operating status feature time series significant propagation coding vector.

[0091] More specifically, in the embodiment of the present disclosure, step S14231 includes calculating the semantic correlation between each implicitly associated heterogeneous modulation and coding feature vector between device operating parameters in the time queue of the implicitly associated heterogeneous modulation and coding feature vector between device operating parameters and the heterogeneous propagation representation vector of the implicitly associated feature between device operating parameters to obtain a time queue of the semantic correlation between the implicitly associated device operating parameters. This process can be expressed as:

[0092]

[0093] α={α1,α2,...,α i ,...,α n}

[0094] Among them, h i is the i-th device operating parameter implicitly associated heterogeneous modulation coding feature vector in the time queue of the device operating parameter implicitly associated heterogeneous modulation coding feature vector, h n+1is the heterogeneous propagation representation vector of the implicit correlation characteristics between the operating parameters of the equipment, ‖·‖ is the length of the calculation vector, arccosh(·) is the inverse hyperbolic cosine function, α1, α2, α i and α n are the 1st, 2nd, i-th and n-th implicit association semantic correlations between equipment operation parameters in the time queue of implicit association semantic correlations between equipment operation parameters, and α is the time queue of implicit association semantic correlations between equipment operation parameters;

[0095] The semantic space jump value of each implicitly associated heterogeneous modulation and coding feature vector between device operating parameters in the time queue of the implicitly associated heterogeneous modulation and coding feature vector between the device operating parameters relative to the heterogeneous propagation representation vector of the implicitly associated feature between the device operating parameters is calculated to obtain the time queue of the semantic space jump value of the implicitly associated semantic space between the device operating parameters. The above process can be expressed as:

[0096] β i =‖h n+1 -h i ‖

[0097] β={β1,β2,...,β i ,...,β n}

[0098] Among them, h i is the i-th device operating parameter implicitly associated heterogeneous modulation coding feature vector in the time queue of the device operating parameter implicitly associated heterogeneous modulation coding feature vector, h n+1 is the heterogeneous propagation representation vector of the implicit correlation characteristics between the operating parameters of the equipment, ‖·‖ is the length of the calculation vector, β1, β2, β i and β n are the first, second, i-th and n-th jump values of the implicit association semantic space between device operating parameters in the time queue of the jump values of the implicit association semantic space between device operating parameters, and β is the time queue of the jump values of the implicit association semantic space between device operating parameters;

[0099] The time queue of the implicit semantic association degree between the device operating parameters and the time queue of the implicit semantic spatial jump value between the device operating parameters are subjected to multi-gated information transmission modulation to obtain the time queue of the implicit association transmission modulation weight between the device operating parameters. The above process can be expressed as:

[0100]

[0101] Among them, α is the time queue of the implicit semantic correlation between the equipment operation parameters, β is the time queue of the implicit semantic space jump value between the equipment operation parameters, W α and Wβ are the semantic association weight matrix and the semantic space weight matrix respectively, is matrix multiplication, b is the bias vector, sigmoid(·) is the sigmoid function, γ t It is a time queue that transfers modulation weights with implicit associations between equipment operating parameters.

[0102] It should be understood that by calculating the semantic correlation between each implicitly correlated heterogeneous modulation coding feature vector between device operating parameters in the time sequence of the implicitly correlated heterogeneous modulation coding feature vectors between device operating parameters and the heterogeneous propagation representation vector of the implicitly correlated heterogeneous modulation coding feature vectors between device operating parameters, the semantic intrinsic correlation between the temporal propagation characteristics of device operating parameter information and each implicitly correlated heterogeneous modulation coding feature vector between device operating parameters in the Poincare space can be expressed. This facilitates the subsequent introduction of an attention mechanism to enhance the model's focus on specific device operating parameter features, focusing on key parameter features that are more strongly correlated with potential equipment failures, while relatively weakening the attention to less important parameters, thereby more accurately capturing key information about device operation. Subsequently, the semantic space jump value of each implicitly correlated heterogeneous modulation coding feature vector between device operating parameters in the time sequence of the implicitly correlated heterogeneous modulation coding feature vectors between device operating parameters relative to the heterogeneous propagation representation vector of the implicitly correlated heterogeneous modulation coding feature vectors between device operating parameters is calculated. This not only reveals the dynamic characteristics of device operating parameter features but also helps discover periodic patterns and abnormal events in the raw device operation data. Then, the time queue of the implicit association semantic correlation between the device operating parameters and the time queue of the implicit association semantic space jump value between the device operating parameters are subjected to multiple gated information transmission modulation to obtain the time queue of the implicit association transmission modulation weight between the device operating parameters. By combining the implicit association semantic correlation between the device operating parameters and the implicit association semantic space jump value between the device operating parameters in this way, the direction and intensity of the device operating parameter timing information flow can be dynamically adjusted. Specifically, the model can adaptively and selectively strengthen or weaken the transmission of specific device operating parameter timing information according to the relationship between the device operating parameter timing characteristics, thereby improving the accuracy and robustness of the device operating parameter timing feature representation.

[0103] More specifically, in the embodiment of the present disclosure, step S14232 includes: using the time sequence of the modulation weights of the implicit correlation transmission between the device operating parameters as a weight sequence, calculating the position-weighted sum of the time sequence of the foreign-domain modulation coding feature vectors of the implicit correlation between the device operating parameters to obtain a significant propagation representation vector of the implicit correlation feature between the device operating parameters. The above process can be expressed as:

[0104]

[0105] Among them, h i is the i-th device operating parameter implicitly associated heterogeneous modulation coding feature vector in the time queue of the device operating parameter implicitly associated heterogeneous modulation coding feature vector, γ i It is H i The corresponding implicit correlation transfer modulation weight between the device operating parameters, n is the number of vectors in the time queue of the implicit correlation heterogeneous modulation coding feature vector between the device operating parameters, H c is a significant propagation representation vector of implicit correlation features between the operating parameters of the equipment;

[0106] Performing residual connection on the significant propagation representation vector of the implicit correlation feature between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters to obtain the significant propagation compensation representation vector of the heterogeneous propagation of the implicit correlation feature between the device operating parameters. The above process can be expressed as:

[0107] H n+1 ′ =H n+1 +H c

[0108] Among them, H c is the significant propagation representation vector of the implicit correlation characteristics between the equipment operating parameters, h n+1 is the heterogeneous propagation representation vector of the implicit correlation characteristics between the operating parameters of the equipment, h n+1 ′ is the compensation representation vector of the significant propagation of the heterogeneous features of the implicit correlation between the operating parameters of the equipment;

[0109] The implicit correlation between the equipment operating parameters and the significant propagation compensation representation vector of the heterogeneous features are mapped to the Euclidean space to obtain the significant propagation encoding vector of the equipment operating status feature time series. The above process can be expressed as:

[0110] V c =f euclidean (h n+1 ′ )=W3h n+1 ′ W4

[0111] Among them, h n+1 ′ is the implicit correlation between the operating parameters of the equipment and the significant propagation compensation representation vector of the alien features, W3 and W4 are the third and fourth weight matrices respectively, f euclidean (·) is the Euclidean space mapping operation, V c It is a significant propagation coding vector of the characteristic time series of the operating status of the device.

[0112] It should be understood that, by using the time sequence of the modulation weights for implicitly transmitting the correlation between the device operating parameters as a sequence of weights, the position-weighted sum of the time sequence of the heterogeneous modulation coding feature vectors for implicitly transmitting the correlation between the device operating parameters is calculated. This weighted summation method can comprehensively consider the importance and temporal changes of the device operating parameter characteristics, and obtain a significant propagation representation vector of the implicit correlation characteristics between the device operating parameters that reflects the evolution of the device operating parameter characteristics over time. This significant propagation representation vector can fully reflect the spatiotemporal characteristics of the original device operating data. Next, a residual connection is performed on the significant propagation representation vector of the implicit correlation characteristics between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation characteristics between the device operating parameters. This not only promotes the direct transmission of the temporal characteristics of the device operating parameters, but also enables the network to maintain good performance even in very deep situations, while also enhancing the richness and diversity of the representation of the temporal characteristics of the device operating parameters. Finally, the implicit correlation between the equipment operating parameters and the significant propagation compensation representation vector of the heterogeneous features are mapped to the Euclidean space to convert the complex processed equipment operating parameter features back to the traditional Euclidean space. This achieves an effective projection from the high-dimensional feature space to the low-dimensional space, which can facilitate the further processing of the subsequent feature data.

[0113] In step S143, the monitoring result is obtained based on the significant propagation coding vector of the equipment operating status characteristics, and the warning prompt is generated based on the monitoring result. Specifically, in the embodiment of the present disclosure, step S143 includes: inputting the significant propagation coding vector of the equipment operating status characteristics into a classifier-based operating status monitoring module to obtain the monitoring result, wherein the monitoring result indicates whether the operating status of the key equipment in the thermal power plant is abnormal; if the monitoring result indicates that the operating status of the key equipment in the thermal power plant is abnormal, the warning prompt is generated. It should be understood that the significant propagation coding vector of the equipment operating status characteristics contains complex characteristic information of the equipment operating status in the time series. However, this characteristic information requires further classification and judgment to determine whether the equipment operating status is abnormal. The classifier is the key component to achieve this classification and judgment function. Through extensive data training, it can classify the equipment status into different categories such as normal or abnormal based on the characteristic patterns in the coding vector. If the monitoring result indicates that the equipment operating status is abnormal, the system will generate a warning prompt based on this, prompting personnel to take appropriate measures in a timely manner, such as inspecting and repairing the equipment or adjusting operating parameters. This helps prevent the equipment failure from further deteriorating, thereby ensuring the safe and stable operation of the thermal power plant equipment. Specifically, the forms of early warning prompts include but are not limited to strong flashing lights and high-decibel sound alarms, SMS notifications containing information such as the equipment name, abnormal parameters and approximate fault type, and pop-up prompts on the monitoring software interface of the central monitoring server. In particular, in the embodiment of the present disclosure, one possible implementation method of inputting the equipment operating status feature time series significant propagation coding vector into the operating status monitoring module based on the classifier to obtain a monitoring result for indicating whether the working status of the key equipment of the thermal power plant is abnormal can be: using the fully connected layer of the classifier to fully connect the equipment operating status feature time series significant propagation coding vector to obtain the equipment operating status fully connected coding feature vector; inputting the equipment operating status fully connected coding feature vector into the Softmax classification function of the classifier to obtain the probability value of the equipment operating status feature time series significant propagation coding vector belonging to each classification label, the classification label including the one for indicating that the working status of the key equipment of the thermal power plant is abnormal and the one for indicating that the working status of the key equipment of the thermal power plant is not abnormal; and determining the classification label corresponding to the largest of the probability values as the monitoring result.

[0114] In summary, step S140 is clearly explained. It uses AI-based data analysis and coding technology to arrange the equipment operation data in the time dimension, and then mines the implicit correlation features between the operating parameters in the equipment operation data at each time point after the arrangement. In this way, based on the multiple gated time series propagation representations between the implicit correlation features between the operating parameters of each device, it can intelligently judge whether there is an abnormality in the working status of the key equipment of the power plant, and generate corresponding early warning prompts. In this way, the complex nonlinear time series dynamic changes between the operating parameters can be effectively captured, providing higher prediction accuracy. In addition, it no longer relies on fixed preset thresholds, but dynamically adjusts the early warning standards through intelligent algorithms, so that it can respond quickly to sudden failures in a short time, thereby improving the effectiveness and reliability of the early warning.

[0115] In summary, the PLC-based thermal power plant equipment operating status monitoring and early warning method based on the disclosed embodiment is explained. It first installs sensors on key thermal power plant equipment to collect data such as boiler temperature, pressure, steam flow, and equipment vibration amplitude. The collected data is then transmitted to the PLC controller via a communication network. The PLC controller then sends the data in batches to the thermal power plant equipment central monitoring server. The central monitoring server analyzes the collected data to obtain equipment monitoring results and generates early warning alerts based on the results. This allows for effective monitoring of equipment status and abnormality early warning, thereby ensuring the reliability and safety of thermal power plant equipment.

[0116] Figure 5 FIG. 1 is a system block diagram of a PLC-based thermal power plant equipment operation status monitoring and early warning system according to an embodiment of the present disclosure. Figure 5As shown, according to the embodiment of the present disclosure, the PLC-based thermal power plant equipment operation status monitoring and early warning system 100 includes: an equipment operation data acquisition module 110, which is used to deploy sensor components on key equipment of the thermal power plant, and the sensor components are used to collect equipment operation data, and the equipment operation data include boiler temperature, boiler pressure, steam flow and equipment vibration amplitude value; an equipment operation data transmission module 120, which is used for the sensor components to transmit the equipment operation data to the PLC controller using the communication network; a central monitoring server receiving data module 130, which is used for the PLC controller to transmit the equipment operation data in batches to the thermal power plant equipment central monitoring server; an equipment operation data analysis module 140, which is used to transmit the equipment operation data to the thermal power plant equipment central monitoring server in batches The device analyzes the device operation data to obtain monitoring results, and generates early warning prompts based on the monitoring results, wherein the device operation data analysis module includes: a device operation data encoding unit, which is used to perform time sequence arrangement and implicit correlation between parameters of the device operation data to obtain a time queue of implicit correlation feature vectors between device operation parameters; a device operation feature dynamic propagation unit, which is used to perform multiple gated time sequence feature dynamic propagation on the time queue of implicit correlation feature vectors between device operation parameters to obtain a device operation status feature time sequence significant propagation encoding vector; a device monitoring result generation unit, which is used to obtain the monitoring result based on the device operation status feature time sequence significant propagation encoding vector, and generate the early warning prompt based on the monitoring result.

[0117] Here, those skilled in the art will appreciate that the specific functions and operations of the various units and modules in the PLC-based thermal power plant equipment operation status monitoring and early warning system 100 have been described in detail above. Figures 1 to 4 The description of the PLC-based thermal power plant equipment operation status monitoring and early warning method has been introduced in detail, and therefore, its repeated description will be omitted.

[0118] In summary, the PLC-based thermal power plant equipment operating status monitoring and early warning system 100 based on the disclosed embodiment is described. It first installs sensors on key thermal power plant equipment to collect data such as boiler temperature, pressure, steam flow, and equipment vibration amplitude. The collected data is then transmitted to the PLC controller via a communication network. The PLC controller then sends the data in batches to the thermal power plant equipment central monitoring server. The central monitoring server analyzes the collected data to obtain equipment monitoring results and generates early warning alerts based on the results. This allows for effective monitoring of equipment status and abnormality early warning, thereby ensuring the reliability and safety of thermal power plant equipment.

[0119] Those skilled in the art will appreciate that the above-mentioned embodiments are specific embodiments for implementing the present disclosure, and that in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present disclosure.

Claims

1. A PLC-based thermal power plant equipment operating status monitoring and early warning method, characterized in that: include: Deploy sensor components on key equipment in thermal power plants, wherein the sensor components are used to collect equipment operating data, including boiler temperature, boiler pressure, steam flow, and equipment vibration amplitude; The sensor assembly transmits the device operation data to the PLC controller using a communication network; The PLC controller transmits the equipment operation data in batches to the central monitoring server of the thermal power plant equipment; On the thermal power plant equipment central monitoring server, the equipment operation data is analyzed to obtain monitoring results, and based on the monitoring results, an early warning prompt is generated, including: arranging the equipment operation data in time sequence and implicitly correlating parameters to obtain a time queue of implicit correlation feature vectors between equipment operation parameters; The time queue of the implicit correlation feature vectors between the equipment operating parameters is subjected to multiple gated timing feature dynamic propagation to obtain the equipment operating status feature timing significant propagation coding vector; based on the equipment operating status feature timing significant propagation coding vector, the monitoring result is obtained, and based on the monitoring result, the early warning prompt is generated.

2. The PLC-based thermal power plant equipment operation status monitoring and early warning method according to claim 1 is characterized in that: Performing time sequence arrangement on the device operation data and implicitly correlating parameters to obtain a time queue of implicit correlation feature vectors between the device operation parameters includes: Arranging the device operation data according to a time step dimension to obtain a time queue of the device operation data; Each device operation data in the time queue of the device operation data is input into a device operation parameter correlation feature extractor based on a fully connected layer to obtain a time queue of implicit correlation feature vectors between the device operation parameters.

3. The PLC-based thermal power plant equipment operation status monitoring and early warning method according to claim 2 is characterized in that: Performing multiple gated timing feature dynamic propagation on the time queue of the implicit correlation feature vectors between the device operating parameters to obtain a significant propagation coding vector of the device operating state feature timing, including: Performing Poincare space mapping on each implicit correlation feature vector between device operating parameters in the time queue of implicit correlation feature vectors between device operating parameters to obtain a time queue of implicit correlation heterogeneous modulation coding feature vectors between device operating parameters; Performing feature sequence encoding on the time queue of the implicitly associated heterogeneous modulation coding feature vectors between the device operating parameters to obtain a heterogeneous propagation representation vector of the implicitly associated feature between the device operating parameters; Calculate the significant propagation representation between the time queue of the implicit correlation feature heterogeneous propagation representation vector between the device operating parameters and the implicit correlation heterogeneous modulation coding feature vector between the device operating parameters to obtain the device operating status feature time series significant propagation coding vector.

4. The PLC-based thermal power plant equipment operation status monitoring and early warning method according to claim 3 is characterized in that: The method comprises the following steps: performing feature sequence encoding on the time queue of the implicitly associated heterogeneous modulation coding feature vectors between the device operating parameters to obtain the heterogeneous propagation representation vector of the implicitly associated feature between the device operating parameters, including: performing forward LSTM sequence encoding on the time queue of the implicitly associated heterogeneous modulation coding feature vectors between the device operating parameters to obtain the heterogeneous propagation representation vector of the implicitly associated feature between the device operating parameters.

5. The PLC-based thermal power plant equipment operation status monitoring and early warning method according to claim 4 is characterized in that: Calculating a significant propagation representation between a heterogeneous propagation representation vector of an implicit correlation feature between the device operating parameters and a time sequence of a heterogeneous modulation coding feature vector of an implicit correlation feature between the device operating parameters to obtain a significant propagation coding vector of a temporal sequence of the device operating state features, including: Calculating, based on the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters, the modulation weight value of each heterogeneous modulation and coding feature vector of the implicit correlation feature between the device operating parameters in the time queue of the heterogeneous modulation and coding feature vector of the implicit correlation feature between the device operating parameters to obtain the time queue of the modulation weight of the implicit correlation transmission between the device operating parameters; Based on the time queue of the modulation weights of the implicit correlation transmission between the device operating parameters, the time queue of the implicit correlation heterogeneous modulation coding feature vector between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation characteristics between the device operating parameters are weighted compensated and fused to obtain the device operating status feature time series significant propagation coding vector.

6. The PLC-based thermal power plant equipment operation status monitoring and early warning method according to claim 5 is characterized in that: The method further comprises calculating, based on the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters, a modulation weight value of each heterogeneous modulation coding feature vector of the implicit correlation feature between the device operating parameters in the time queue of the heterogeneous modulation coding feature vector of the implicit correlation feature between the device operating parameters to obtain a time queue of the modulation weight of the implicit correlation transmission between the device operating parameters, including: Calculating the semantic correlation between each implicitly associated heterogeneous modulation and coding feature vector between device operating parameters in the time queue of the implicitly associated heterogeneous modulation and coding feature vector between device operating parameters and the heterogeneous propagation representation vector of the implicitly associated feature between device operating parameters to obtain a time queue of the semantic correlation between the implicitly associated device operating parameters; Calculating the semantic space jump value of each implicitly associated heterogeneous modulation and coding feature vector between device operating parameters in the time queue of the implicitly associated heterogeneous modulation and coding feature vector between the device operating parameters relative to the heterogeneous propagation representation vector of the implicitly associated feature between the device operating parameters to obtain a time queue of the semantic space jump value of the implicitly associated semantic space between the device operating parameters; Multiple gated information transmission modulation is performed on the time queue of the implicit association semantic association degree between the device operation parameters and the time queue of the implicit association semantic space jump value between the device operation parameters to obtain the time queue of the implicit association transmission modulation weight between the device operation parameters.

7. The PLC-based thermal power plant equipment operation status monitoring and early warning method according to claim 6 is characterized in that: Based on the time queue of the modulation weights of the implicit correlation transmission between the device operating parameters, weighted compensation fusion is performed on the time queue of the implicit correlation heterogeneous modulation coding feature vector between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters to obtain the device operating state feature time series significant propagation coding vector, including: Using the time sequence of the modulation weights of the implicit correlation transmission between the device operating parameters as a weight sequence, calculating the position-weighted sum of the time sequence of the heterogeneous modulation coding feature vectors of the implicit correlation between the device operating parameters to obtain a significant propagation representation vector of the implicit correlation feature between the device operating parameters; Performing a residual connection on the significant propagation representation vector of the implicit correlation feature between the device operating parameters and the heterogeneous propagation representation vector of the implicit correlation feature between the device operating parameters to obtain a compensation representation vector of the significant propagation of the heterogeneous feature of the implicit correlation between the device operating parameters; The implicit correlation heterogeneous feature significant propagation compensation representation vector between the equipment operating parameters is mapped to the Euclidean space to obtain the equipment operating state feature time series significant propagation coding vector.

8. The PLC-based thermal power plant equipment operation status monitoring and early warning method according to claim 7 is characterized in that: Obtaining the monitoring result based on the significant propagation coding vector of the characteristic time series of the equipment operation status, and generating the early warning prompt based on the monitoring result, including: Inputting the equipment operating state feature time series significant propagation coding vector into a classifier-based operating state monitoring module to obtain the monitoring result, which is used to indicate whether the operating state of the key equipment of the thermal power plant is abnormal; If the monitoring result indicates that the working status of the key equipment of the thermal power plant is abnormal, the early warning prompt is generated.

9. A PLC-based thermal power plant equipment operation status monitoring and early warning system, characterized in that: include: An equipment operation data acquisition module is used to deploy sensor components on key equipment in a thermal power plant. The sensor components are used to collect equipment operation data, including boiler temperature, boiler pressure, steam flow, and equipment vibration amplitude values; A device operation data transmission module, used for the sensor component to transmit the device operation data to the PLC controller using a communication network; The central monitoring server receives data modules, which are used for the PLC controller to transmit the equipment operation data in batches to the central monitoring server of the thermal power plant equipment; The equipment operation data analysis module is used to analyze the equipment operation data in the central monitoring server of the thermal power plant equipment to obtain monitoring results, and generate early warning prompts based on the monitoring results. The equipment operation data analysis module includes: an equipment operation data encoding unit, which is used to perform time sequence arrangement and implicit correlation between parameters of the equipment operation data to obtain a time queue of implicit correlation feature vectors between equipment operation parameters; an equipment operation feature dynamic propagation unit, which is used to perform multiple gated time sequence feature dynamic propagation on the time queue of implicit correlation feature vectors between equipment operation parameters to obtain equipment operation status feature time sequence significant propagation encoding vectors; an equipment monitoring result generation unit, which is used to obtain the monitoring result based on the equipment operation status feature time sequence significant propagation encoding vectors, and generate the early warning prompts based on the monitoring result.

10. The PLC-based thermal power plant equipment operation status monitoring and early warning system according to claim 9 is characterized in that: The device monitoring result generating unit is used to: Inputting the equipment operating state feature time series significant propagation coding vector into a classifier-based operating state monitoring module to obtain the monitoring result, which is used to indicate whether the operating state of the key equipment of the thermal power plant is abnormal; If the monitoring result indicates that the working status of the key equipment of the thermal power plant is abnormal, the early warning prompt is generated.

Images