Gas-steam combined cycle unit operation control method and related device

Abstract

The invention belongs to a unit operation control method, and provides a gas-steam combined cycle unit operation control method and a related device for solving the technical problem that a traditional automatic start-stop control system is difficult to meet the strict requirements of a novel electric power system for safe, stable and economical operation of a gas-steam combined cycle unit. Time sequence operation data in the production process of the gas-steam combined cycle unit are input into the abnormity early warning model, warning information is obtained, the abnormity early warning model carries out early warning on the performance and faults of the gas-steam combined cycle unit, and the abnormity early warning model is a model constructed based on a modeling method combining a mechanism and mathematical theory and carries out early warning on the performance and faults of the gas-steam combined cycle unit. And inputting the alarm information into the abnormity diagnosis model to obtain a diagnosis report. And analyzing the diagnosis report to obtain a control instruction for adjusting the operation of the gas-steam combined cycle unit. Therefore, closed-loop autonomous operation under the abnormal condition of the unit can be realized.

Description

Technical Field

[0001] This application pertains to a unit operation control method, specifically relating to a gas-fired steam combined cycle unit operation control method and related devices. Background Technology

[0002] With the global energy structure transformation and the rapid development of renewable energy, gas-fired combined cycle (Gas-fired) power generation technology is increasingly becoming an important part of the power industry. To ensure grid stability and promote the consumption of clean energy, Gas-fired combined cycle units face frequent daily start-ups and shutdowns. However, traditional Automatic Power Plant Start-up and Shutdown (APS) systems rely entirely on the experience and manual real-time intervention of operators to quickly identify, locate, and handle abnormal operating conditions during unit start-up and shutdown processes and normal operation. This not only significantly increases the workload of operators but also makes it easier to cause human error risks under complex operating conditions, making it difficult to meet the stringent requirements of new power systems for the safe, stable, and economical operation of Gas-fired combined cycle units. Summary of the Invention

[0003] This application addresses the technical problem that traditional automatic start-stop control systems, which rely entirely on the experience and judgment of operators and manual real-time intervention, are unable to meet the stringent requirements of new power systems for the safe, stable, and economical operation of gas-fired combined cycle units. It provides a method and related devices for the operation control of gas-fired combined cycle units.

[0004] To achieve the above objectives, this application adopts the following technical solution: Firstly, this application proposes a method for operating and controlling a gas-fired steam combined cycle unit, comprising: The time-series operation data of the gas-fired steam combined cycle unit during the production process is acquired and input into the anomaly early warning model to obtain alarm information; the anomaly early warning model includes a performance early warning module and a fault early warning module for the gas-fired steam combined cycle unit; the anomaly early warning model is a model constructed based on a modeling method that combines mechanism and mathematics; The alarm information is input into the anomaly diagnosis model to obtain a diagnosis report; the anomaly diagnosis model includes a performance diagnosis module and a fault diagnosis module for the gas-fired steam combined cycle unit; Analyzing the diagnostic report yields control commands for adjusting the operation of the gas-fired steam combined cycle unit; Adjust the operating status of the gas-fired steam combined cycle unit according to control commands.

[0005] Furthermore, a modeling method based on the combination of mechanism and mathematics is constructed, including: Screening of measurement points; the measurement points include sensor measurement points that collect time-series operation data during the production process of the gas-fired steam combined cycle unit, as well as mechanism calculation measurement points; The selected measurement points were analyzed using a correlation analysis algorithm; Based on the correlation with abnormal operating conditions during the production process of the gas-fired combined cycle unit, simulated measurement points are selected to characterize the occurrence of abnormal operating conditions; the abnormal operating conditions include abnormal performance conditions and abnormal fault conditions of the gas-fired combined cycle unit. An anomaly early warning model is obtained by using a modeling method that combines mechanism and mathematics based on simulated measurement points.

[0006] Furthermore, the alarm information includes real-time data of abnormal measuring points and abnormal associated measuring points; The abnormal measurement points include sensor measurement points and mechanism calculation measurement points where the operating parameter thresholds of the gas-steam combined cycle unit exceed the limits or the trends are abnormal during the production process. The abnormal correlation measurement points include measurement points that meet preset requirements after correlation analysis with abnormal measurement points.

[0007] Furthermore, the method executed in the anomaly diagnosis model includes: Based on alarm information, a clustering algorithm is used to search for data corresponding to abnormal operating conditions in a pre-set database; By comparing the real-time data of abnormal measuring points and abnormal related measuring points in the alarm information with the data corresponding to the abnormal operating conditions, the location of the abnormal operating conditions, the cause analysis results, and the handling methods can be identified.

[0008] Furthermore, the method executed in the anomaly diagnosis model also includes: By using a large language model, alarm information is matched with existing abnormal operating condition data in a pre-set database, and then associated with semantic reasoning based on a knowledge base of expert experience to infer the possible causes and handling methods of abnormal operating conditions.

[0009] Furthermore, the method for analyzing the diagnostic report to obtain control commands for adjusting the operation of the gas-fired steam combined cycle unit includes: extracting information from the abnormal operating condition handling methods involved in the diagnostic report through a natural language processing model and converting it into control commands.

[0010] Furthermore, adjusting the operating status of the gas-fired steam combined cycle unit according to control commands includes: The control commands are gradually sent to the gas-fired steam combined cycle unit. Each step in the gradual sending process is adjusted and determined based on the timing operation data of the current gas-fired steam combined cycle unit's production process.

[0011] Secondly, this application proposes an operation control system for a gas-fired steam combined cycle unit, comprising: The data module is used to acquire time-series operating data during the production process of the gas-fired steam combined cycle unit, input it into the anomaly early warning model, and obtain alarm information. The anomaly early warning model includes a performance early warning module and a fault early warning module for the gas-fired steam combined cycle unit. The anomaly early warning model is a model constructed based on a modeling method that combines mechanism and mathematics. The diagnostic module is used to input the alarm information into the anomaly diagnostic model to obtain a diagnostic report; the anomaly diagnostic model includes a performance diagnostic module and a fault diagnostic module for the gas-fired steam combined cycle unit. The analysis module is used to analyze the diagnostic report and obtain control commands for adjusting the operation of the gas-fired steam combined cycle unit; The control module is used to adjust the operating status of the gas-fired steam combined cycle unit according to control commands.

[0012] Thirdly, this application proposes an electronic device, including: a memory and one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the steps of the above-described gas-fired steam combined cycle unit operation control method.

[0013] Fourthly, this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described gas-fired combined cycle unit operation control method.

[0014] Compared with the prior art, this application has the following beneficial effects: This application proposes a method for the operation control of a gas-fired combined cycle (GC-COB) unit. The method inputs the time-series operating data of the GC-COB production process into an anomaly early warning model to obtain alarm information. This model, constructed using a modeling method combining mechanistic and mathematical approaches, can provide early warnings of GC-COB performance and faults. The alarm information is then input into an anomaly diagnosis model to obtain a diagnostic report. Finally, the diagnostic report is analyzed to obtain control commands for adjusting the operation of the GC-COB, and the operating state of the GC-COB is adjusted according to these commands. This achieves a closed-loop process of unmanned autonomous operation of the GC-COB, from trend early warning and intelligent diagnosis to autonomous anomaly control. This application can provide alarms before anomalies occur in the GC-COB, reserving sufficient response time for intelligent decision-making and control adjustments under abnormal conditions. The anomaly diagnosis model allows for the setting of various analysis algorithms, enabling intelligent decision-making from different perspectives, effectively improving the accuracy and efficiency of decision-making results, and ensuring the reliability of anomaly control. Ultimately, by combining intelligent algorithms, an integrated "perception-decision-control" abnormal autonomous operation system was constructed. This system can quickly identify, locate, and handle abnormal operating conditions during the start-up and shutdown process and normal operation of the gas-fired steam combined cycle unit, achieving closed-loop autonomous operation under abnormal conditions. This significantly reduces the workload of operators, improves the safety and reliability of the unit, and reduces the risk of accidents and unplanned shutdowns.

[0015] This application also proposes a gas-fired steam combined cycle unit operation control system, an electronic device, and a computer-readable storage medium, which possess all the advantages of the aforementioned gas-fired steam combined cycle unit operation control methods. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of the operation control method for a gas-fired steam combined cycle unit according to this application. Figure 2 A schematic diagram of an unmanned autonomous operation system structure constructed for this application; Figure 3 This is a schematic diagram of a specific structure of an unmanned autonomous operating system in the embodiments of this application; Figure 4 This is a flowchart illustrating two integrated algorithms for anomaly diagnosis and an anomaly control module in the embodiments of this application. Figure 5 This is a schematic diagram of a specific example of the operation control method for the gas-fired steam combined cycle unit of this application; Figure 6 This is a schematic diagram of the operation control system of the gas-fired steam combined cycle unit of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0021] In the description of the embodiments of this application, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. In addition, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0023] In the description of the embodiments of this application, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0024] As the global energy structure transitions towards cleaner and lower-carbon energy, the installed capacity and power generation of renewable energy sources such as wind and solar power continue to grow rapidly, placing higher demands on the flexibility and stability of power systems. Combined Cycle Gas Turbine (CCGT) power generation technology, with its advantages of fast start-up, strong peak-shaving capacity, and low carbon emissions, has gradually become a key support for connecting baseload power and renewable energy generation. It is widely used in power industry scenarios such as peak shaving and power supply guarantee, and grid frequency regulation, playing an irreplaceable role, especially in addressing fluctuations in renewable energy output and ensuring grid supply and demand balance. In actual operation, to adapt to grid load changes and promote the consumption of clean energy, CCGT units often need to perform frequent daily start-up and shutdown operations. They start generating electricity during the day according to peak demand and shut down at night or during off-peak hours. This operating mode has become the mainstream operating mode for this type of unit under the current new power system. Meanwhile, to ensure the safe and efficient operation of the unit, an Automatic Power Plant Startup and Shutdown System (APS) is traditionally configured. The APS is mainly responsible for the automated execution of routine steps during the start-up and shutdown process of the unit, such as valve opening and closing, speed regulation, load increase and decrease, and other basic operations. It is the basic equipment for realizing the automated operation of gas-fired steam combined cycle units.

[0025] During the frequent start-ups, shutdowns, and normal operation of gas-fired combined cycle (GasCO) units, various abnormal operating conditions can occur, such as parameter anomalies caused by sensor malfunctions, operational fluctuations due to equipment performance degradation, and system imbalances caused by sudden changes in external loads. If these abnormal conditions are not identified and handled in a timely manner, they can affect the operating efficiency of the GasCO unit, and even lead to equipment damage, unplanned shutdowns, and in severe cases, threaten grid stability. However, traditional Automation System (APS) has significant limitations in dealing with these abnormal conditions. It can only perform preset routine automated operations and lacks the ability to proactively identify and handle abnormal conditions. Once an anomaly occurs, it relies entirely on operators' personal experience for real-time judgment and manual intervention. Therefore, this leads to the following significant shortcomings: First, frequent start-ups and shutdowns significantly increase the monitoring intensity and workload of operators, and anomaly judgment under complex conditions requires operators to quickly integrate multi-dimensional data; second, the response speed of manual intervention is limited by human reaction time, making it difficult to meet the need for faster response under abnormal conditions, and errors may occur during operation, further amplifying the risks.

[0026] To address the aforementioned issues, current improvements primarily focus on optimizing the functionality of traditional APS (Automatic Power Supply) systems. This includes adding parameter monitoring points and implementing more refined threshold alarm mechanisms to promptly issue alerts when parameters exceed normal ranges, prompting operators to take action. Some solutions also incorporate simple logic judgment functions, automatically triggering preset protection actions when a critical parameter exceeds a threshold, reducing reliance on manual judgment. While existing technologies have improved upon the shortcomings of traditional methods to some extent, many unresolved issues remain, hindering truly unmanned autonomous operation and failing to completely solve the problems faced by gas-fired combined cycle units in handling abnormal operating conditions.

[0027] Based on the above, this application proposes a method and related apparatus for the operation control of a gas-fired steam combined cycle unit. The following is a detailed description of this application in conjunction with the embodiments and accompanying drawings.

[0028] like Figure 1 The diagram shown is a flowchart illustrating one possible method for operating and controlling a gas-fired combined cycle unit according to this application, which may include: S101: Acquire the time-series operating data of the gas-fired combined cycle unit during production, input it into the anomaly early warning model, and obtain alarm information. The anomaly early warning model includes a performance early warning module and a fault early warning module for the gas-fired combined cycle unit. The anomaly early warning model is constructed based on a modeling method that combines mechanistic and mathematical approaches.

[0029] It should be noted that time-series operating data typically refers to the continuous operational data collected over time during the production process of a gas-fired combined cycle (GC) unit, such as temperature, pressure, and flow rate at different times. GC units generate a large amount of time-series operating data reflecting their own state during operation, and the patterns of these data changes are closely related to whether the GC unit is operating normally. The anomaly warning model analyzes this data to identify characteristics that deviate from normal patterns. Then, the performance warning module determines whether there is a performance degradation trend, and the fault warning module determines whether there is a risk of equipment failure, ultimately generating alarm information.

[0030] In practical applications, after acquiring time-series operational data, preprocessing can be performed, such as noise removal and missing value filling. As an example, the performance warning module in the anomaly warning model can predict and determine whether the current time-series operational data exceeds the normal operating performance parameter range of the gas-fired combined cycle unit. If it does, a performance anomaly warning is issued. The fault warning module can compare the characteristics of historical fault data with the current time-series operational data; if they match, a fault has occurred. Finally, the two types of warnings are integrated to form an alarm message.

[0031] The anomaly early warning model is based on a modeling method combining mechanistic and mathematical approaches. By integrating the physical laws governing unit operation with historical data statistical characteristics, it allows for in-depth analysis of time-series operational data through performance early warning and fault early warning modules. As an example, the anomaly early warning model balances mechanistic and mathematical aspects. On the mechanistic side, mathematical equations, such as thermodynamic cycle equations and fluid dynamics equations, can be constructed to ensure the anomaly early warning model conforms to the physical operating laws of gas-fired steam combined cycle units, avoiding misjudgments caused by data noise, and demonstrating stronger stability, especially under complex operating conditions such as unit start-up and shutdown, and variable loads. On the mathematical side, algorithms such as machine learning and statistical analysis can be used to train historical time-series operational data, capturing subtle data correlations that are difficult to cover in the mechanistic aspect, such as minor parameter drifts caused by equipment aging, thus improving the sensitivity of the early warning.

[0032] The above methods can capture abnormal signs in the operation of gas-fired steam combined cycle units in real time and accurately, issue early warnings, buy time for subsequent handling, and avoid the abnormality from escalating to more serious consequences. At the same time, warnings are issued separately for performance and faults, making the warnings more targeted.

[0033] S102, input the alarm information into the anomaly diagnosis model to obtain a diagnosis report. The anomaly diagnosis model includes a performance diagnosis module and a fault diagnosis module for the gas-fired steam combined cycle unit.

[0034] In practical applications, alarm information can only indicate that there is an abnormality in the gas-fired steam combined cycle unit, but it cannot specify the specific cause of the abnormality or the scope of its impact.

[0035] Therefore, this application can deeply analyze alarm information, identify key information about the anomaly, provide a basis for formulating precise control instructions, and avoid blind handling that could lead to waste of resources or expansion of the problem.

[0036] S103, analyze the diagnostic report and obtain control commands for adjusting the operation of the gas-fired steam combined cycle unit.

[0037] In practical applications, diagnostic reports can include key information such as the cause of the abnormality and the handling methods. Based on this information, combined with the parameter requirements and operating procedures for the normal operation of the gas-fired steam combined cycle unit, specific operating instructions can be derived to eliminate the abnormality and restore the gas-fired steam combined cycle unit to normal operation. These instructions serve as control commands, ensuring the effectiveness and relevance of the adjustment operations.

[0038] S104, adjust the operating status of the gas-fired steam combined cycle unit according to the control command.

[0039] In practical applications, the control system of a gas-fired steam combined cycle unit can receive and execute control commands. By adjusting the operating parameters of relevant equipment inside the gas-fired steam combined cycle unit, it can change the working status of the unit's fuel supply, power output, and other aspects, thereby adjusting the overall operating status of the unit in the direction of eliminating abnormalities.

[0040] This application addresses the reliance on manual judgment and intervention in the operation and control of traditional gas-fired combined cycle (GWC) units, constructing an automated and intelligent operation and control system. By combining anomaly early warning and anomaly diagnosis models, it not only achieves early and accurate identification of unit anomalies but also deeply analyzes the root causes, providing a scientific basis for subsequent handling and avoiding the escalation of anomalies caused by insufficient human experience or delayed judgment in traditional methods. Simultaneously, the entire scheme forms a closed-loop control system encompassing data acquisition, anomaly early warning, anomaly diagnosis, control command generation, and state adjustment, enabling rapid response and efficient handling of GWC unit operational anomalies, significantly improving the safety and stability of GWC unit operation. Furthermore, this application utilizes a modeling method combining mechanistic and mathematical principles, taking into account both the physical laws of unit operation and the statistical characteristics of historical data, resulting in higher accuracy in anomaly early warning and diagnosis, effectively reducing misjudgments and omissions, thereby lowering the probability of unplanned shutdowns of GWC units and improving unit operating efficiency and economy.

[0041] As one way to implement the above-mentioned operation control method for gas-fired combined cycle units, such as Figure 2The diagram shown is a schematic of an unmanned autonomous operation system constructed according to this application. It may include an autonomous sensing module, an intelligent decision-making module, and an anomaly control module, wherein the autonomous sensing module and the anomaly control module are respectively connected to a gas-fired steam combined cycle unit. In this embodiment, it specifically includes: The autonomous sensing module is used to acquire time-series operating data of the gas-fired steam combined cycle unit's production process, input it into the anomaly early warning model, generate alarm information, and send it to the intelligent decision-making module.

[0042] The intelligent decision-making module is used to obtain a diagnostic report based on alarm information through the anomaly diagnosis model and send it to the anomaly control module. The anomaly diagnosis model integrates the performance diagnosis and fault diagnosis functions of the gas-fired steam combined cycle unit through the performance diagnosis module and the fault diagnosis module.

[0043] The anomaly control module is used to obtain control commands based on the diagnostic report and send them to the gas-fired steam combined cycle unit to realize the closed-loop process of unmanned autonomous operation of the gas-fired steam combined cycle unit.

[0044] like Figure 3 The diagram shown illustrates a specific structure of the above embodiment. The autonomous sensing module includes a real-time monitoring unit and an anomaly early warning unit. The real-time monitoring unit acquires time-series operating data and historical data from the time-series data source of the gas-fired combined cycle unit during the production process. It performs preliminary processing and filtering of the time-series operating data and historical data before inputting them into the anomaly early warning unit to ensure data accuracy and completeness. The time-series data source may include DCS (Distributed Control System) data such as sensor measurement points and mechanism calculation measurement points obtained through the OPC (OLE for Process Control) protocol or other relevant data interfaces, which are transmitted to the time-series database of the real-time monitoring unit. The anomaly early warning unit has a preset anomaly early warning model. This model is constructed using a modeling method combining mechanistic and mathematical approaches. It performs online analysis and machine learning inference calculations on the input time-series operating data and existing historical data. When it calculates that the system or equipment operating parameters in the gas-fired combined cycle unit exceed threshold limits or exhibit abnormal trends, it generates alarm information in real time and forwards it to the intelligent decision-making module. As an example, alarm information may include sensor measurement points, mechanism calculation measurement points, and real-time data of relevant measurement points obtained by correlation analysis with abnormal measurement points for the operating parameters of the gas-fired steam combined cycle unit that exceed the threshold or show abnormal trends.

[0045] In some embodiments of this application, the specific calculation method in the anomaly warning model is as follows: A correlation analysis algorithm is used to analyze the measurement points involved in the anomaly warning model modeling. Based on the analysis results, a series of relevant analog measurement points that can quickly and accurately characterize the occurrence of abnormal operating conditions are selected. Modeling is then performed using a modeling method combining mechanistic and mathematical approaches, and these modeled measurement points exhibit strong correlations. Subsequently, the anomaly warning model is integrated into the anomaly warning unit and trained using advanced intelligent algorithms combined with professional knowledge and experience, thereby obtaining the anomaly warning model.

[0046] In other embodiments of this application, the autonomous sensing module can also monitor the power plant's on-site production systems and equipment in real time. Using the anomaly warning unit, in the early stages of a system or equipment failure or performance degradation in a gas-fired steam combined cycle unit, the anomaly warning model will generate alarms through specific measurement points strongly correlated with the fault, reminding operation and maintenance personnel to pay attention and use the intelligent decision-making module to process the alarm information.

[0047] The intelligent decision-making module includes an anomaly diagnosis unit, which uses an anomaly diagnosis comprehensive algorithm to accurately diagnose the input alarm information, obtain the precise location of the abnormal working condition, the result of the cause analysis and the handling method, and input the diagnosis result into the anomaly control module.

[0048] This embodiment illustrates the specific implementation methods of three comprehensive anomaly diagnosis algorithms, such as... Figure 4 The diagram shown illustrates two integrated anomaly diagnosis algorithms and the flowchart of the anomaly control module: (1) Fault detection strategy based on early warning monitoring model: Abnormal operating condition data in historical data is found by using clustering algorithm, and then machine learning is used to train the early warning monitoring model, so that the early warning monitoring model can identify possible problems or abnormal operating conditions by comparing the similarity between the current measurement point data of the equipment or system and the known abnormal operating condition data. The essence of this strategy is to perform diagnostic analysis from the perspective of measurement point data and to diagnose based on the similarity of data operation. This strategy does not rely on complex equipment mechanism knowledge and can uncover anomalies missed by traditional methods, which is especially suitable for scenarios with many measurement points and complex operating conditions, such as gas-fired steam combined cycle units.

[0049] (2) Fault Diagnosis Strategy Based on Knowledge Mining and Large Language Model Analysis and Reasoning: This strategy matches alarm information with existing data in the case library using a large language model, then correlates it with semantic reasoning based on expert experience to infer possible causes of anomalies and handling methods, providing diagnostic results. This strategy also possesses real-time learning capabilities, updating based on new fault data and user feedback to ensure data up-to-dateness and accuracy. The essence of this strategy is semantic analysis. Through large language model technology, combined with expert knowledge bases from mainstream manufacturers and typical event case libraries, it comprehensively analyzes fault phenomena from the perspective of phenomenon descriptions in the knowledge base and anomaly case library for diagnostic analysis. The large language model used here can better understand and handle the complex operating states and fault modes of gas-fired combined cycle units, thereby improving diagnostic accuracy.

[0050] It should be noted that the aforementioned knowledge base refers to a fault knowledge base built upon expert domain knowledge. Through mechanistic analysis and expert experience and insights into equipment operation and maintenance, equipment failure modes are extracted. Simultaneously, artificial intelligence technology is combined to construct a deep learning model of failure modes, forming a knowledge base based on expert experience. This achieves the integration of domain knowledge while addressing the problem of knowledge not being generalizable and shared across different units. Specifically, the knowledge base in this embodiment performs a structured transformation of expert knowledge. It is not simply a piling up of expert experience, but rather, through mechanistic analysis, combined with the experience accumulated by experts in equipment operation and maintenance, it transforms fragmented experience into definable and callable structured knowledge. Then, based on this structured knowledge, a deep learning model is introduced, allowing the deep learning model to transform expert knowledge into computable judgment logic. The deep learning model can also handle complex scenarios not covered by expert experience, supplementing and optimizing the knowledge system through data learning. This allows one knowledge base to serve multiple units, significantly reducing the cost of knowledge reconstruction. The case library, on the other hand, is constructed by operators reviewing existing historical cases, building a case library based on typical abnormal operating conditions that occurred historically. In practical applications, operators can collect typical abnormal operating condition cases that have occurred in the past of gas-fired steam combined cycle units, sort them out, and store them to form a case library.

[0051] Of the two strategies mentioned above, the fault-detection strategy based on the early warning monitoring model utilizes machine learning technology to perform quantitative feature learning and pattern recognition on the operating data and historical abnormal operating data of the gas-fired combined cycle unit, based on raw point data from sensor measurement points or data from mechanistic calculation measurement points. This is similar to Western medicine requiring patients to undergo various examinations before diagnosis and making a judgment based on the data on the test reports. The fault diagnosis strategy based on knowledge mining and large language model analysis and reasoning uses a large language model to correlate abnormal phenomena with semantic reasoning in a knowledge base and case base to infer possible causes of faults and solutions. This is similar to traditional Chinese medicine comparing symptoms with relevant diseases recorded in medical books to diagnose and prescribe medicine. Based on this "combination of traditional Chinese and Western medicine" diagnostic approach, the accuracy and efficiency of fault diagnosis results can be effectively improved, enabling knowledge sharing and generalization analysis capabilities across multiple gas-fired combined cycle units.

[0052] The third strategy combines the above two approaches, achieving intelligent and precise condition monitoring and early warning for gas-fired combined cycle (GWC) units, thus strengthening the defenses for their safe and efficient operation. This involves two dimensions: first, from the perspective of measurement point data, combining historical measurement points with the similarity of operational data behavior (i.e., the similarity of measurement point data curves) for diagnostic analysis; second, from the perspective of semantic analysis, using large language model technology combined with expert knowledge bases from mainstream manufacturers and typical event case libraries to comprehensively analyze fault phenomena from the perspective of phenomenon descriptions in the knowledge base and anomaly case library for diagnostic analysis. Both approaches can perform independent diagnostic analyses and can also combine their results to provide a recommended ranking of diagnostic suggestions. Furthermore, they possess learning and improvement capabilities, receiving user feedback, including both positive and negative feedback, to perform reinforcement learning and continuously optimize diagnostic effectiveness.

[0053] In this embodiment, based on the early fault judgment given by the autonomous perception module, the intelligent decision-making module uses methods such as equipment mechanism, language model, knowledge base, and case library to form anomaly diagnosis of specific alarm information, thereby achieving more accurate positioning, proposing solutions to the problem, and generating a diagnostic report for input into the anomaly control module.

[0054] In this embodiment, the anomaly control module includes an anomaly control unit. This unit analyzes the input diagnostic report using a natural language processing model, extracts key information from the anomaly handling schemes within the report, and transforms it into operable control commands. In practical applications, to enhance control precision, control commands can be automatically issued step-by-step based on real-time feedback from the gas-fired combined cycle unit, according to the anomaly handling scheme. Therefore, after intelligent decision-making, the anomaly control module can directly perform automatic control operations on the gas-fired combined cycle unit based on the diagnostic results, thereby achieving a closed-loop system for unmanned autonomous diagnostic operation and maintenance of the gas-fired combined cycle unit.

[0055] like Figure 5 The diagram shown is a specific example of the gas-fired steam combined cycle unit operation control method of this application. This example also verifies the effectiveness of the gas-fired steam combined cycle unit operation control method of this application.

[0056] A 400MW F-class single-shaft "one-to-one" gas-fired steam combined cycle unit is equipped with two feedwater pumps, designated as Feedwater Pump A and Feedwater Pump B, with one pump in operation and the other on standby. The lubricating oil required for each feedwater pump is supplied by its own dedicated lubrication station, which is equipped with two oil pumps, one in operation and one on standby. There are no monitoring devices for the lubricating oil quality; analysis must be performed manually.

[0057] When the method described in this application is put into operation, the gas-fired steam combined cycle unit is running at a load of 260MW. During the operation of feedwater pump A, the system autonomously detects an abnormal increase in the bearing temperature at both the drive and non-drive ends of pump A, issuing an alarm and pushing the alarm information to the intelligent decision-making module. The specific alarm information is as follows: Anomaly description: Abnormal increase in bearing temperature at the drive end of feedwater pump A; Anomaly measurement points: Bearing temperatures at both the drive and non-drive ends of feedwater pump A; Relevant measurement point data: Feedwater pump outlet pressure, bearing lubricating oil temperature, and lubricating oil cooling water temperature. Based on the alarm information, the intelligent decision-making module analyzes the data using a comprehensive diagnostic algorithm and generates the following diagnostic report: (1) Root cause analysis: Based on the case library, the lubricating oil filter screen was replaced during the maintenance period. The cause may be that the oil quality deteriorated, leading to an abnormal increase in bearing temperature.

[0058] (2) Solution: ① Start water pump B; ② Determine that pump B is operating normally: pump vibration, current, pump outlet pressure, etc. are within the normal range; ③ Stop water pump A.

[0059] The anomaly control module, in conjunction with the diagnostic report, generates the following control scheme for switching the operation of feedwater pump A to feedwater pump B: ① Command: Start water pump B; Feedback: Determine that pump B is operating normally: pump vibration, current, pump outlet pressure, etc. are normal.

[0060] ②Instruction: Stop water pump A; Feedback: Water pump A has stopped.

[0061] Then, the above control commands are gradually sent to the gas-fired steam combined cycle unit based on the real-time feedback from the unit, and finally, after the lubricating oil of feedwater pump A deteriorates, the operation is switched to feedwater pump B.

[0062] In summary, the specific process of "perception-decision-execution" for the above-mentioned abnormal operating conditions is as follows: the autonomous perception module detects the abnormal rise in the temperature of the feedwater pump bearing in advance. After comprehensive analysis by the intelligent decision-making module, it is found that the cause may be the deterioration of the feedwater pump lubricating oil. After providing a solution to switch to the operation of the standby feedwater pump, the abnormal control system forms a control plan and gradually sends it to the gas-steam combined cycle unit based on the real-time feedback from the unit. Finally, the unmanned autonomous operation operation of switching to the operation of feedwater pump B after the deterioration of the lubricating oil of feedwater pump A is completed.

[0063] This application addresses industry pain points by introducing automated control and artificial intelligence algorithms to construct an integrated control system encompassing "perception-decision-execution" for abnormal operating conditions. This system can quickly identify, locate, and handle abnormal operating conditions during start-up, shutdown, and normal operation, achieving autonomous decision-making and closed-loop control of the gas-fired combined cycle unit throughout the entire process from cold start-up to hot shutdown. This significantly reduces the reliance on operator experience and judgment for gas-fired combined cycle units, minimizing the possibility of human error during power generation. It also improves the operating efficiency and reliability of gas-fired combined cycle units under complex conditions such as frequent start-ups and shutdowns, achieving unattended autonomous operation during start-up, shutdown, and operation.

[0064] like Figure 6 The diagram shown is a schematic representation of the operation control system for a gas-fired combined cycle unit according to this application, which may include: The data module is used to acquire time-series operating data during the production process of the gas-fired steam combined cycle unit, input it into the anomaly early warning model, and obtain alarm information. The anomaly early warning model includes a performance early warning module and a fault early warning module for the gas-fired steam combined cycle unit. The anomaly early warning model is a model constructed based on a modeling method that combines mechanism and mathematics. The diagnostic module is used to input the alarm information into the anomaly diagnostic model to obtain a diagnostic report; the anomaly diagnostic model includes a performance diagnostic module and a fault diagnostic module for the gas-fired steam combined cycle unit. The analysis module is used to analyze the diagnostic report and obtain control commands for adjusting the operation of the gas-fired steam combined cycle unit; The control module is used to adjust the operating status of the gas-fired steam combined cycle unit according to control commands.

[0065] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of each block is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple blocks may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0066] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0067] This application also provides an electronic device, which may include one or more processors, memory and communication interfaces.

[0068] The memory, communication interface, and processor are coupled together. For example, the memory, communication interface, and processor can be coupled together via a bus.

[0069] The communication interface is used for data transmission with other devices. The memory stores computer program code. This computer program code includes computer instructions, which, when executed by the processor, cause the electronic equipment to perform the steps of the aforementioned combined cycle gas turbine unit operation control method.

[0070] The processor can be a processor or controller, such as a Central Processing Unit (CPU), a general-purpose processor, a Digital Signal Processor (DSP), an Application-Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with this disclosure. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The processor can be used to support an electronic device in performing the method steps provided in the above embodiments.

[0071] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. These buses can be categorized as address buses, data buses, control buses, etc.

[0072] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the above-described gas-steam combined cycle unit operation control method.

[0073] The computer-readable storage media involved in this application include random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage media known in the art.

[0074] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for operating and controlling a gas-fired steam combined cycle unit, characterized in that, include: The time-series operation data of the gas-fired steam combined cycle unit during the production process is acquired, input into the anomaly early warning model, and alarm information is obtained. The anomaly warning model includes a performance warning module and a fault warning module for gas-fired combined cycle units; The anomaly early warning model is a model constructed based on a modeling method that combines mechanistic and mathematical approaches. The alarm information is input into the anomaly diagnosis model to obtain a diagnosis report; the anomaly diagnosis model includes a performance diagnosis module and a fault diagnosis module for the gas-fired steam combined cycle unit; Analyzing the diagnostic report yields control commands for adjusting the operation of the gas-fired steam combined cycle unit; Adjust the operating status of the gas-fired steam combined cycle unit according to control commands.

2. The method for operating control of a gas-fired steam combined cycle unit according to claim 1, characterized in that, The modeling method, which combines mechanistic and mathematical approaches, is constructed and includes: Screening of measurement points; the measurement points include sensor measurement points that collect time-series operation data during the production process of the gas-fired steam combined cycle unit, as well as mechanism calculation measurement points; The selected measurement points were analyzed using a correlation analysis algorithm; Based on the correlation with abnormal operating conditions during the production process of the gas-fired combined cycle unit, simulated measurement points are selected to characterize the occurrence of abnormal operating conditions; the abnormal operating conditions include abnormal performance conditions and abnormal fault conditions of the gas-fired combined cycle unit. An anomaly early warning model is obtained by using a modeling method that combines mechanism and mathematics based on simulated measurement points.

3. The method for operating control of a gas-fired steam combined cycle unit according to claim 1, characterized in that, The alarm information includes real-time data of abnormal measuring points and abnormal related measuring points; The abnormal measurement points include sensor measurement points and mechanism calculation measurement points where the operating parameter thresholds of the gas-steam combined cycle unit exceed the limits or the trend is abnormal during the production process. The abnormal correlation measurement points include measurement points that meet preset requirements after correlation analysis with abnormal measurement points.

4. The method for operating control of a gas-fired steam combined cycle unit according to claim 3, characterized in that, The methods executed in the anomaly diagnosis model include: Based on alarm information, a clustering algorithm is used to search for data corresponding to abnormal operating conditions in a pre-set database; By comparing the real-time data of abnormal measuring points and abnormal related measuring points in the alarm information with the data corresponding to the abnormal operating conditions, the location of the abnormal operating conditions, the cause analysis results, and the handling methods can be identified.

5. The method for operating control of a gas-fired steam combined cycle unit according to claim 4, characterized in that, The method executed in the anomaly diagnosis model further includes: By using a large language model, alarm information is matched with existing abnormal operating condition data in a pre-set database, and then associated with semantic reasoning based on a knowledge base of expert experience to infer the possible causes and handling methods of abnormal operating conditions.

6. The method for operating control of a gas-fired steam combined cycle unit according to claim 4 or 5, characterized in that, The method for analyzing the diagnostic report to obtain control commands for adjusting the operation of the gas-fired steam combined cycle unit includes: extracting information from the abnormal operating condition handling methods involved in the diagnostic report through a natural language processing model and converting it into control commands.

7. The method for operating control of a gas-fired steam combined cycle unit according to claim 6, characterized in that, The adjustment of the operating status of the gas-fired steam combined cycle unit according to control commands includes: The control commands are gradually sent to the gas-fired steam combined cycle unit. Each step in the gradual sending process is adjusted and determined based on the timing operation data of the current gas-fired steam combined cycle unit's production process.

8. A gas-fired steam combined cycle unit operation control system, characterized in that, include: The data module is used to acquire time-series operating data during the production process of the gas-fired steam combined cycle unit, input it into the anomaly early warning model, and obtain alarm information; The anomaly warning model includes a performance warning module and a fault warning module for gas-fired combined cycle units; The anomaly early warning model is a model constructed based on a modeling method that combines mechanistic and mathematical approaches. The diagnostic module is used to input the alarm information into the anomaly diagnostic model to obtain a diagnostic report; the anomaly diagnostic model includes a performance diagnostic module and a fault diagnostic module for the gas-fired steam combined cycle unit. The analysis module is used to analyze the diagnostic report and obtain control commands for adjusting the operation of the gas-fired steam combined cycle unit; The control module is used to adjust the operating status of the gas-fired steam combined cycle unit according to control commands.

9. An electronic device, characterized in that, include: A memory, one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the steps of the gas-fired steam combined cycle unit operation control method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the gas-fired steam combined cycle unit operation control method as described in any one of claims 1-7.

Images