Gas turbine power plant accident early warning decision-making method based on hidden danger data integrated analysis

Abstract

The invention relates to the technical field of gas turbine power plant safety monitoring, and particularly discloses a gas turbine power plant accident early warning decision-making method based on hidden danger data integrated analysis, and the method comprises the steps: monitoring the wall surface temperature distribution and cooling efficiency of a combustion chamber in real time through a high-precision sensor network; and carrying out deep analysis on the acquired data by utilizing fast Fourier transform and Haar wavelet transform, and evaluating the damage of the wall surface material of the combustion chamber and the health state of the cooling system. The comprehensive feature vectors are processed in combination with a random forest model, the potential safety hazard level in the combustion chamber operation process is accurately judged, once the high-risk situation is detected, the system immediately starts an intelligent emergency plan, automatically adjusts the cooling water flow and the fan rotating speed or activates an additional cooling device and other measures, and potential hazards are prevented from further worsening.

Description

Technical Field

[0001] The present invention relates to the technical field of safety monitoring of gas turbine power plants, and in particular to a gas turbine power plant accident early warning decision method based on integrated analysis of hidden danger data. Background Art

[0002] As an important part of the modern power system, the operational safety and stability of gas turbine power plants have a vital impact on the reliability of the power grid and environmental protection. The combustion chamber is one of the core components of a gas turbine power plant. It works in a high temperature and high pressure environment for a long time. It is easy to cause thermal stress damage and equipment failure due to temperature fluctuations and reduced cooling efficiency, which in turn leads to serious safety accidents. Traditional monitoring methods mainly rely on regular manual inspections and simple sensor data collection, which makes it difficult to achieve real-time and comprehensive monitoring of the combustion chamber status. With the development of industrial Internet of Things technology and big data analysis methods, by integrating a variety of sensor devices and advanced data analysis algorithms, the damage of the combustion chamber wall material and the health of the cooling system can be more accurately evaluated, thereby effectively preventing potential safety hazards.

[0003] The prior art has the following deficiencies: The existing technology lacks the ability to monitor the temperature distribution and cooling efficiency of the combustion chamber wall in real time through an integrated high-precision sensor network, and uses advanced signal processing techniques (such as fast Fourier transform and Haar wavelet transform) to conduct in-depth analysis of the collected data to accurately evaluate the damage of the combustion chamber wall material and the health of the cooling system. The analysis of the comprehensive feature vector combined with the random forest model can accurately determine the level of safety hazards during the operation of the combustion chamber. Once a high-risk situation is detected, the system will immediately activate the intelligent emergency plan, automatically adjust the cooling water flow, fan speed or activate additional cooling devices, etc., to quickly reduce the temperature in the combustion chamber and prevent the hidden dangers from further deteriorating. Summary of the invention

[0004] The purpose of the present invention is to provide a gas turbine power plant accident early warning decision method based on hidden danger data integrated analysis to solve the above-mentioned background problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data comprises the following steps: S1: Real-time monitoring of the temperature distribution and cooling efficiency of the combustion chamber wall during the monitoring period of the gas turbine power plant; S2: Evaluate the damage of the combustion chamber wall material according to the degree of change in the temperature distribution of the combustion chamber wall; S3: judging whether the working state of the cooling system is stable according to the change range of the cooling efficiency of the combustion chamber; S4: Comprehensively analyze the temperature distribution of the combustion chamber wall and the cooling efficiency of the combustion chamber, and determine the safety hazard level of the combustion chamber during operation according to the analysis results, including high risk level, medium risk level and low risk level; S5: If it is determined that the safety hazard level of the combustion chamber during operation is at a high risk level, immediately activate the emergency plan, adjust the combustion mode and add additional cooling measures.

[0006] As a further solution of the present invention: the damage of the combustion chamber wall material is evaluated according to the degree of change of the temperature distribution of the combustion chamber wall, specifically including: During the monitoring period of the gas turbine power plant, the combustion chamber wall is divided into several areas, and the temperature data of the combustion chamber wall in each area is collected respectively. The fluctuation degree of the temperature data in each area of ​​the combustion chamber wall is analyzed respectively, and according to the abnormal distribution of the temperature data, the thermal stress distribution characteristic value of the combustion chamber wall is calculated to determine whether the thermal stress distribution characteristic value of the combustion chamber wall is greater than or equal to a preset threshold. If so, the combustion chamber wall material is damaged, and if not, the combustion chamber wall material is not damaged.

[0007] As a further solution of the present invention: the process of obtaining the characteristic value of the thermal stress distribution is: Get the location in each area and time The temperature measurement value under represents the position coordinates on the combustion chamber wall, Indicates time; Preprocess the acquired temperature data, including removing outliers and smoothing; Apply the fast Fourier transform algorithm to the temperature time series in each region to obtain the frequency domain representation; calculate the square of the modulus of the frequency domain expression obtained after conversion by the fast Fourier transform algorithm to obtain the energy of each frequency component; The thermal stress at the corresponding position is obtained by calculating the elastic modulus of the material multiplied by its thermal expansion coefficient and the energy product of all frequency components at the corresponding position; For each area divided by the combustion chamber wall, the thermal stress distribution characteristic value of the entire wall is calculated according to its average thermal stress value.

[0008] As a further solution of the present invention: judging whether the working state of the cooling system is stable according to the variation range of the cooling efficiency of the combustion chamber specifically includes: During the monitoring period of the gas turbine power plant, the cooling efficiency of the combustion chamber is monitored in real time. According to the variation range of the cooling efficiency of the combustion chamber, the characteristic value of the abnormal variation of the cooling efficiency is calculated to determine whether the characteristic value of the abnormal variation of the cooling efficiency is greater than or equal to a preset threshold. If so, it is determined that the working state of the cooling system is stable; if not, it is determined that the working state of the cooling system is unstable.

[0009] As a further solution of the present invention: the process of obtaining the abnormal change characteristic value of the cooling efficiency is: Get the time in each monitoring cycle The cooling efficiency measurement value; The cooling efficiency time series is decomposed into wavelet coefficients of different scales through Haar wavelet transform; The detail coefficient is taken as the main fluctuation component; The squares of all detail coefficients are added together to obtain the total energy of the detail coefficients at the corresponding scale. The detail coefficient with the highest frequency is obtained. The ratio of the energy of the detail coefficient with the highest frequency to the average energy of the detail coefficients at all scales is calculated to obtain the characteristic value of abnormal change in cooling efficiency.

[0010] As a further solution of the present invention: the comprehensive analysis of the temperature distribution of the wall of the combustion chamber and the cooling efficiency of the combustion chamber specifically includes: The thermal stress distribution characteristic values ​​and the abnormal change characteristic values ​​of the cooling efficiency of the combustion chamber during the monitoring period of the gas turbine power plant are obtained, and the thermal stress distribution characteristic values ​​and the abnormal change characteristic values ​​of the cooling efficiency are constructed into a comprehensive feature vector as the input of the machine learning model. The output of the model is the safety hazard score. According to the safety hazard score output by the model, the safety hazard level of the combustion chamber during operation is determined. The machine learning model is a random forest model.

[0011] As a further solution of the present invention: the training process of the machine learning model is: The historical sets of combustion chamber thermal stress distribution characteristic values, cooling efficiency abnormal change characteristic values ​​and safety hazard scores are obtained as training data sets to train the random forest model. During the training process, the random forest determines the final output safety hazard score by constructing multiple decision trees and voting on the results of each tree.

[0012] As a further solution of the present invention: the step of determining the potential safety hazard level of the combustion chamber during operation specifically includes: The safety hazard score of the combustion chamber of the gas turbine power plant during operation is compared with the preset first threshold. If the safety hazard score is greater than or equal to the preset first threshold, it is recorded as a high risk level. If the safety hazard score is less than the preset first threshold and greater than the preset second threshold, it is recorded as a medium risk level. If the safety hazard score is less than or equal to the safety hazard score, it is recorded as a low risk level.

[0013] As a further solution of the present invention: if it is determined that the safety hazard level of the combustion chamber during operation is a high-risk level, the emergency plan is immediately activated, specifically including: If the safety hazard level of the combustion chamber during operation is determined to be high-risk, the system will immediately activate the early warning mechanism, send notifications to operators and emergency response teams, adjust the cooling water flow, fan speed and activate additional cooling devices to quickly reduce the temperature in the combustion chamber.

[0014] Beneficial effects of the present invention: (1) The present invention realizes real-time, high-precision monitoring of the temperature distribution and cooling efficiency of the combustion chamber wall of a gas turbine power plant by integrating advanced sensing technology and signal processing methods. Fast Fourier transform is used to analyze the combustion chamber wall temperature data, and spectral features are extracted to evaluate the thermal stress distribution and its impact on material damage. At the same time, Haar wavelet transform is used to decompose the cooling efficiency time series, identify local abnormal changes, and ensure that the health status of the cooling system is fully monitored. Based on these in-depth analysis results, a comprehensive feature vector is constructed and input into a random forest model, which accurately determines the level of safety hazards during the operation of the combustion chamber through a voting mechanism of multiple decision trees. Once a high-risk situation is detected, the system immediately activates an intelligent emergency plan, automatically adjusts the cooling water flow, fan speed, or activates additional cooling measures to quickly reduce the temperature inside the combustion chamber and prevent the hazard from further deteriorating. This full-process closed-loop management from real-time data acquisition, advanced signal processing, intelligent data analysis to automated emergency response not only significantly improves the safety and reliability of gas turbine power plants, but also effectively prevents major safety accidents caused by equipment failures, ensures the safety of personnel and equipment, and provides an efficient, flexible and adaptive technical solution for industrial safety. In addition, the system ensures long-term stable operation, reduces maintenance costs, and extends equipment life by continuously optimizing control strategies. It has important practical application value and broad promotion prospects.

[0015] (2) The present invention can not only accurately identify potential safety hazards at an early stage, but also achieve efficient and flexible emergency response and optimization control through an integrated intelligent decision support module. When a high-risk situation is detected, the system can automatically adjust the cooling water flow, fan speed or activate additional cooling devices to quickly reduce the temperature in the combustion chamber and effectively avoid the further expansion of hidden dangers. The system has excellent adaptive capabilities and can dynamically adjust parameter settings according to real-time operating conditions to ensure the best cooling effect and system stability. Compared with the traditional manual monitoring and adjustment method, the automated solution provided by the present invention significantly shortens the response time and greatly improves the ability and efficiency to deal with emergencies. In addition, the system's built-in continuous monitoring and feedback mechanism can continuously optimize the cooling strategy to ensure the long-term and efficient operation of the cooling system, thereby reducing maintenance costs and extending the service life of the equipment. Through this full-process closed-loop management from data collection, intelligent analysis to dynamic adjustment, the present invention provides an efficient, reliable and highly adaptive technical means for industrial safety. Its advanced algorithm and intelligent design not only improve the safety and reliability of gas turbine power plants, but also have wide applicability and promotion value. It is suitable for a variety of complex industrial environments and provides solid technical support for ensuring the safe and stable operation of key equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 It is a flowchart of the specific steps of a gas turbine power plant accident early warning decision-making method based on integrated analysis of hidden danger data of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] See also Figure 1 As shown, the present invention is a method for early warning decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data, comprising the following steps: S1: Real-time monitoring of the temperature distribution and cooling efficiency of the combustion chamber wall during the monitoring period of the gas turbine power plant; S2: Evaluate the damage of the combustion chamber wall material according to the degree of change in the temperature distribution of the combustion chamber wall; S3: judging whether the working state of the cooling system is stable according to the change range of the cooling efficiency of the combustion chamber; S4: Comprehensively analyze the temperature distribution of the combustion chamber wall and the cooling efficiency of the combustion chamber, and determine the safety hazard level of the combustion chamber during operation according to the analysis results, including high risk level, medium risk level and low risk level; S5: If it is determined that the safety hazard level of the combustion chamber during operation is at a high risk level, immediately activate the emergency plan, adjust the combustion mode and add additional cooling measures.

[0020] In S1, during the monitoring period of the gas turbine power plant, the temperature distribution and cooling efficiency of the combustion chamber wall are monitored in real time, including: In S2, the damage of the combustion chamber wall material is evaluated according to the degree of change of the thermal stress distribution on the combustion chamber wall, including: During the monitoring cycle of a gas turbine power plant, the temperature distribution of the combustion chamber wall is first monitored in real time. This step is achieved through a network of high-precision temperature sensors deployed at key locations in the combustion chamber. These sensors can continuously collect temperature data at different points on the combustion chamber wall. The acquisition of temperature data includes not only the absolute temperature value, but also the temperature change trend over time and the temperature difference between different areas. In order to ensure the accuracy and reliability of the data, all temperature sensors need to be calibrated regularly and a redundant design is adopted to avoid single point failures. In addition, through an integrated data acquisition system, these scattered temperature information can be aggregated and transmitted to the central control system for subsequent analysis and processing.

[0021] High-precision temperature sensors are used to measure the inlet temperature of the cooling medium before it enters the combustion chamber cooling system and the outlet temperature after heat exchange. Flow meters are used to accurately record the flow of the cooling medium. These data are transmitted to the central monitoring system for processing. Based on these raw data, the actual heat removed by the cooling system is first calculated. The calculation expression is: ,in It represents the mass flow rate converted from the flow rate. represents the specific heat capacity of the cooling medium, represents the inlet temperature, represents the outlet temperature, Represents heat load. The cooling efficiency is calculated based on the ratio of actual heat load to theoretical maximum heat load, where the theoretical maximum heat load is estimated based on the design parameters of the combustion chamber. The obtained cooling efficiency data will also be integrated into the central control system to provide a basis for subsequent safety hazard analysis.

[0022] During the monitoring period of the gas turbine power plant, the combustion chamber wall is divided into several areas, and the temperature data of the combustion chamber wall in each area is collected respectively. The fluctuation degree of the temperature data in each area of ​​the combustion chamber wall is analyzed respectively, and according to the abnormal distribution of the temperature data, the thermal stress distribution characteristic value of the combustion chamber wall is calculated to determine whether the thermal stress distribution characteristic value of the combustion chamber wall is greater than or equal to a preset threshold. If so, the combustion chamber wall material is damaged, and if not, the combustion chamber wall material is not damaged.

[0023] The process of obtaining the characteristic value of thermal stress distribution is as follows: Get the location in each area and time Temperature measurement under ,in, represents the position coordinates on the combustion chamber wall, Indicates time; Preprocess the acquired temperature data, including removing outliers and smoothing; For each region, the temperature time series Apply the fast Fourier transform algorithm to obtain the frequency domain representation ,in represents frequency variable; Compute the square of the magnitude of the spectral density function: ,in, Represents the energy of each frequency component. According to the thermal stress theory, the thermal stress is calculated. The calculation expression is: ,in represents thermal stress, represents the elastic modulus, represents the coefficient of thermal expansion; The average thermal stress of each divided area is calculated. Based on the average thermal stress of each divided area, the average thermal stress of all areas is calculated to obtain the characteristic value of the thermal stress distribution of the entire combustion chamber wall.

[0024] It should be noted that the thermal stress distribution characteristic value reflects the thermal stress distribution characteristics of the combustion chamber wall, and the larger the thermal stress distribution characteristic value, the higher the degree of material damage to the corresponding combustion chamber wall.

[0025] In S3, judging whether the working state of the cooling system is stable according to the change range of the cooling efficiency of the combustion chamber, specifically includes: Get the time in each monitoring cycle The cooling efficiency measurement value under the condition of 100% cooling efficiency is obtained; the obtained cooling efficiency data is preprocessed, including removing outliers and smoothing, to ensure the data quality of subsequent analysis; The cooling efficiency time series is decomposed into wavelet coefficients of different scales through Haar wavelet transform, specifically including: for a given time series, Haar wavelet transform can decompose the original signal into a series of approximation coefficients and detail coefficients; the approximation coefficient represents the general trend of the signal, while the detail coefficient captures the local changes or abnormal fluctuations in the signal, and "scale" refers to the level of decomposition.

[0026] Calculate the energy of each detail coefficient, the calculation expression is: ,in, Indicates The energy sum of the layer detail coefficients, represents the number of layers of Haar wavelet transform, Indicates The detail coefficient, Indicates The first layer of Haar wavelet transform Detail coefficients; Indicates The first layer of Haar wavelet transform Detail coefficients; For the detail coefficient of the highest frequency, the sum of its energies is calculated, and the ratio of the sum of the energy to the average value of the detail coefficient energy at all scales is calculated to obtain the characteristic value of the abnormal change in cooling efficiency.

[0027] It should be noted that by using Haar wavelet transform to analyze the time series data of cooling efficiency and evaluating the health status of the cooling system based on the calculated abnormal change characteristic value of cooling efficiency, this method can not only effectively detect abnormal changes in cooling efficiency, but also provide a scientific basis for preventing equipment damage caused by cooling system failure.

[0028] In S4, a comprehensive analysis is performed on the temperature distribution of the combustion chamber wall and the cooling efficiency of the combustion chamber. Based on the analysis results, the safety hazard level of the combustion chamber during operation is determined, including high risk level, medium risk level and low risk level, specifically including: Obtaining the thermal stress distribution characteristic values ​​and the cooling efficiency abnormal change characteristic values ​​of the combustion chamber during the monitoring period of the gas turbine power plant, constructing the thermal stress distribution characteristic values ​​and the cooling efficiency abnormal change characteristic values ​​into a comprehensive characteristic vector as the input of the machine learning model, the output of the model is the safety hazard score, and determining the safety hazard level of the combustion chamber during operation according to the safety hazard score output by the model, wherein the machine learning model is a random forest model; The historical sets of combustion chamber thermal stress distribution characteristic values, cooling efficiency abnormal change characteristic values ​​and safety hazard scores are obtained as training data sets to train the random forest model. During the training process, the random forest determines the final output safety hazard score by constructing multiple decision trees and voting on the results of each tree. Each time a node is split, the algorithm selects the best split attribute based on information gain (such as Gini impurity or entropy), thereby ensuring that the constructed model has good generalization ability.

[0029] After the model training is completed, the random forest model needs to be rigorously evaluated to verify its performance on unseen data. This usually includes using cross-validation techniques to calculate the model's evaluation indicators of precision, recall, and F1 score. Once the model is confirmed to be accurate, it can be applied to real-time monitoring and output a safety hazard score based on the input comprehensive feature vector. Based on the preset first threshold and the preset second threshold, the safety hazard score can be mapped to different safety hazard levels, including: low risk level, medium risk level, and high risk level. This grading mechanism helps operators quickly identify potential hazards during the operation of the combustion chamber and take appropriate preventive measures to ensure the safe and stable operation of the equipment.

[0030] In S5, if the safety hazard level of the combustion chamber during operation is determined to be a high risk level, the emergency plan is immediately activated to adjust the combustion mode and add additional cooling measures, including: When the random forest model analysis shows that the level of safety hazards in the combustion chamber operation is at a high risk level, the system will immediately activate the early warning mechanism and send an emergency notification to the operator and the emergency response team to ensure that the relevant personnel can be informed and respond quickly. Then, the intelligent decision support module automatically intervenes and adjusts the cooling water flow, fan speed or activates additional cooling devices based on the real-time collected combustion chamber temperature data to quickly reduce the temperature in the combustion chamber and prevent the hidden dangers from further deteriorating. The entire process not only achieves a seamless connection from early warning to response, but also ensures the timeliness and effectiveness of cooling measures through intelligent means, thereby maximizing the safe and stable operation of the gas turbine power plant. This comprehensive response plan is significantly different from the traditional manual monitoring and adjustment method, and provides a more efficient and reliable automation solution.

[0031] The working principle of the present invention is as follows: during the monitoring period of a gas turbine power plant, a high-precision temperature sensor network deployed at a key position of the combustion chamber is used to monitor the temperature distribution of the combustion chamber wall in real time, and a flow meter and a temperature sensor are used to monitor the inlet temperature, outlet temperature and flow rate of the cooling medium, so as to calculate the actual heat load and cooling efficiency of the cooling system. These raw data are transmitted to a central control system for processing. For the combustion chamber wall, it is divided into several regions, and the fast Fourier transform algorithm is applied to the temperature data in each region to obtain the frequency domain representation and calculate the square of the modulus of the spectrum density function, and then the thermal stress distribution eigenvalue is estimated according to the thermal stress theory to evaluate the damage of the combustion chamber wall material. At the same time, the Haar wavelet transform is used to decompose the cooling efficiency time series, and the energy of the detail coefficient is calculated to identify the variation range of the cooling efficiency and determine whether the working state of the cooling system is stable. The thermal stress distribution eigenvalue of the combustion chamber and the abnormal change eigenvalue of the cooling efficiency are constructed into a comprehensive eigenvector as the input of the random forest model, and the model outputs a safety hazard score. The random forest model is trained by a historical data set, and the model performance is evaluated by a cross-validation technique to ensure that it has good generalization ability. According to the safety hazard score output by the model, combined with the preset threshold, the level of safety hazard during the operation of the combustion chamber can be determined. Once it is determined to be a high-risk level, the system immediately activates the early warning mechanism, notifies relevant personnel, and automatically adjusts the cooling water flow, fan speed, or activates additional cooling devices and other measures to quickly reduce the temperature in the combustion chamber and prevent the hidden danger from further deteriorating. Throughout the process, not only is the full-process automated management from data collection to safety hazard assessment to emergency response realized, but the timeliness and effectiveness of cooling measures are also ensured through the intelligent decision support module, significantly improving the ability to respond to emergencies. Compared with traditional manual monitoring and adjustment methods, the present invention provides an efficient and reliable automation solution, which greatly improves the safety and stability of gas turbine power plants, reduces unexpected downtime and economic losses caused by equipment failures, and provides a solid technical guarantee for industrial safety. In addition, the method can also flexibly adjust parameters according to actual conditions to adapt to different working conditions, and has strong practicality and promotion value.

[0032] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0033] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0034] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0035] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0036] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A gas turbine power plant accident early warning decision method based on hidden danger data integrated analysis, characterized in that: The following steps are involved: S1: Real-time monitoring of the temperature distribution and cooling efficiency of the combustion chamber wall during the monitoring period of the gas turbine power plant; S2: Evaluate the damage of the combustion chamber wall material according to the degree of change in the temperature distribution of the combustion chamber wall; S3: judging whether the working state of the cooling system is stable according to the change range of the cooling efficiency of the combustion chamber; S4: Comprehensively analyze the temperature distribution of the combustion chamber wall and the cooling efficiency of the combustion chamber, and determine the safety hazard level of the combustion chamber during operation according to the analysis results, including high risk level, medium risk level and low risk level; S5: If it is determined that the safety hazard level of the combustion chamber during operation is at a high risk level, immediately activate the emergency plan, adjust the combustion mode and add additional cooling measures.

2. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 1 is characterized in that: The method of evaluating the damage of the combustion chamber wall material according to the degree of change of the temperature distribution of the combustion chamber wall specifically includes: During the monitoring period of the gas turbine power plant, the combustion chamber wall is divided into several areas, and the temperature data of the combustion chamber wall in each area is collected respectively. The fluctuation degree of the temperature data in each area of ​​the combustion chamber wall is analyzed respectively, and according to the abnormal distribution of the temperature data, the thermal stress distribution characteristic value of the combustion chamber wall is calculated to determine whether the thermal stress distribution characteristic value of the combustion chamber wall is greater than or equal to a preset threshold. If so, the combustion chamber wall material is damaged, and if not, the combustion chamber wall material is not damaged.

3. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 2 is characterized in that: The process of obtaining the characteristic value of thermal stress distribution is as follows: Get the location in each area and time The temperature measurement value under represents the position coordinates on the combustion chamber wall, Indicates time; Preprocess the acquired temperature data, including removing outliers and smoothing; Apply the fast Fourier transform algorithm to the temperature time series in each region to obtain the frequency domain representation; calculate the square of the modulus of the frequency domain expression obtained after conversion by the fast Fourier transform algorithm to obtain the energy of each frequency component; The thermal stress at the corresponding position is obtained by calculating the elastic modulus of the material multiplied by its thermal expansion coefficient and the energy product of all frequency components at the corresponding position; For each area divided by the combustion chamber wall, the thermal stress distribution characteristic value of the entire wall is calculated according to its average thermal stress value.

4. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 1 is characterized in that: The step of judging whether the working state of the cooling system is stable according to the variation range of the cooling efficiency of the combustion chamber specifically includes: During the monitoring period of the gas turbine power plant, the cooling efficiency of the combustion chamber is monitored in real time. According to the variation range of the cooling efficiency of the combustion chamber, the characteristic value of the abnormal variation of the cooling efficiency is calculated to determine whether the characteristic value of the abnormal variation of the cooling efficiency is greater than or equal to a preset threshold. If so, it is determined that the working state of the cooling system is stable; if not, it is determined that the working state of the cooling system is unstable.

5. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 4 is characterized in that: The process of obtaining the abnormal change characteristic value of cooling efficiency is as follows: Get the time in each monitoring cycle The cooling efficiency measurement value; The cooling efficiency time series is decomposed into wavelet coefficients of different scales through Haar wavelet transform; The detail coefficient is taken as the main fluctuation component; The squares of all detail coefficients are added together to obtain the total energy of the detail coefficients at the corresponding scale. The detail coefficient with the highest frequency is obtained. The ratio of the energy of the detail coefficient with the highest frequency to the average energy of the detail coefficients at all scales is calculated to obtain the characteristic value of abnormal change in cooling efficiency.

6. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 1 is characterized in that: The comprehensive analysis of the temperature distribution of the combustion chamber wall and the cooling efficiency of the combustion chamber specifically includes: The thermal stress distribution characteristic values ​​and the abnormal change characteristic values ​​of the cooling efficiency of the combustion chamber during the monitoring period of the gas turbine power plant are obtained, and the thermal stress distribution characteristic values ​​and the abnormal change characteristic values ​​of the cooling efficiency are constructed into a comprehensive feature vector as the input of the machine learning model. The output of the model is the safety hazard score. According to the safety hazard score output by the model, the safety hazard level of the combustion chamber during operation is determined. The machine learning model is a random forest model.

7. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 6 is characterized in that: The training process of the machine learning model is: The historical sets of combustion chamber thermal stress distribution characteristic values, cooling efficiency abnormal change characteristic values ​​and safety hazard scores are obtained as training data sets to train the random forest model. During the training process, the random forest determines the final output safety hazard score by constructing multiple decision trees and voting on the results of each tree.

8. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 6 is characterized in that: Determining the potential safety hazard level of the combustion chamber during operation specifically includes: The safety hazard score of the combustion chamber of the gas turbine power plant during operation is compared with the preset first threshold. If the safety hazard score is greater than or equal to the preset first threshold, it is recorded as a high risk level. If the safety hazard score is less than the preset first threshold and greater than the preset second threshold, it is recorded as a medium risk level. If the safety hazard score is less than or equal to the safety hazard score, it is recorded as a low risk level.

9. The method for early warning and decision-making of a gas turbine power plant accident based on integrated analysis of hidden danger data according to claim 1, characterized in that: If it is determined that the safety hazard level of the combustion chamber during operation is at a high risk level, the emergency plan will be immediately activated, including: If the safety hazard level of the combustion chamber during operation is determined to be high-risk, the system will immediately activate the early warning mechanism, send notifications to operators and emergency response teams, adjust the cooling water flow, fan speed and activate additional cooling devices to quickly reduce the temperature in the combustion chamber.

Images