Fault diagnosis method and device for gas turbine system

Abstract

The present application relates to the field of gas turbine technology and discloses a fault diagnosis method and apparatus for a gas turbine system, which is applied to a turbine device in a gas turbine system, wherein the turbine blades in the turbine device have corresponding initial cleaning times. The method comprises: obtaining monitoring time series data of multiple specified parameters of the turbine blades; extracting single parameter features on a single time scale based on the monitoring time series data of the multiple specified parameters to obtain parameter-related features for each specified parameter; extracting multi-parameter features on multiple time scales to obtain multi-parameter correlation features; and generating maintenance measures suitable for the turbine blades based on the parameter-related features and multi-parameter correlation features for each specified parameter. The present application performs multi-parameter and multi-time scale analysis of the fouling condition of turbine blades in a gas turbine system and adjusts the initial cleaning time in real time based on the fouling condition of the turbine blades, effectively reducing the failure risk and maintenance cost of the gas turbine system.

Description

Technical Field

[0001] The present application relates to the technical field of gas turbines, and in particular to a method and device for diagnosing faults in a gas turbine system. Background Art

[0002] A gas turbine is a thermodynamic machine that converts high-temperature, high-pressure gas generated by combustion into usable mechanical energy by rotating a turbine. It boasts high thermal efficiency, lightweight, compact size, high power, and fast startup, and is widely used in a variety of fields, including aerospace, marine propulsion, and power generation. Gas turbine systems are complex, with their key components often operating under harsh conditions of high temperature and high pressure. Frequent starts and stops, as well as load fluctuations, make their actual operating conditions complex and variable, increasing the probability of failure and, in turn, impacting their normal operation.

[0003] Common fault types in gas turbine systems include compressor failure, combustion chamber failure, and turbine failure. Turbine failures include turbine blade fouling. To eliminate turbine blade fouling in gas turbine systems, ultrasonic inspection and cleaning methods are commonly used. However, these inspection and cleaning methods require improvement. Summary of the Invention

[0004] The present application provides a fault diagnosis method and device for a gas turbine system, which performs multi-parameter and multi-time-scale analysis on the fouling condition of turbine blades in the gas turbine system, and adjusts the cleaning time of the turbine blades in real time according to the fouling condition of the turbine blades, thereby generating maintenance measures that fit the actual fouling situation on the turbine blades, effectively reducing the failure risk and maintenance cost of the gas turbine system.

[0005] In order to achieve the above objectives, the main technical solutions adopted in this application include:

[0006] In a first aspect, an embodiment of the present application provides a fault diagnosis method for a gas turbine system, the method being applied to a turbine device of the gas turbine system, wherein turbine blades in the turbine device have an initial cleaning time. The method comprises:

[0007] Acquiring monitoring time series data of a plurality of designated parameters of the turbine blade; wherein the designated parameters are parameters affected or changed by the fouling condition on the turbine blade;

[0008] Based on the monitoring time series data of the multiple specified parameters, single parameter feature extraction is performed on a single time scale to obtain parameter-related features of each specified parameter; and multi-parameter feature extraction is performed on multiple time scales to obtain multi-parameter correlation features; wherein the parameter-related features are used to characterize the trend characteristics of the monitoring time series data of the corresponding specified parameters, and the multi-parameter correlation features are used to characterize the correlation between the specified parameters on multiple time scales;

[0009] A maintenance measure suitable for the turbine blade is generated according to the parameter-related characteristics of each designated parameter and the multi-parameter correlation characteristics; wherein the maintenance measure includes an adjustment suggestion for the initial cleaning time.

[0010] The gas turbine system fault diagnosis method proposed in the present embodiment extracts single-parameter features at a single time scale based on multiple monitoring time series data of turbine blades, and extracts multi-parameter features at multiple time scales based on the multiple monitoring time series data. This method obtains data features for each of the multiple specified parameters, as well as correlation features between the specified parameters at different time scales. Based on the parameter correlation features of the specified parameters and the multi-parameter correlation features of the multiple specified parameters, a multi-parameter and multi-time scale analysis of the turbine blade fouling condition is performed, thereby improving the accuracy of gas turbine system fault diagnosis, enabling appropriate maintenance measures to be taken for the gas turbine system and reducing the probability of gas turbine system failure. Compared with related art, the present invention analyzes the turbine blade fouling condition at multiple parameters and multiple time scales, and generates corresponding maintenance measures based on the turbine blade fouling condition, thereby adjusting the initial cleaning time of the turbine blades in real time. This effectively reduces the risk of gas turbine system failure caused by turbine blade fouling, reduces the probability of unnecessary turbine blade cleaning events, and reduces the cost of cleaning turbine blades, thereby reducing the maintenance cost of the gas turbine system to a certain extent.

[0011] Optionally, the plurality of specified parameters include flow, pressure and temperature at corresponding monitoring positions of the turbine equipment; the monitoring time series data includes flow time series data, pressure time series data and temperature time series data;

[0012] The monitoring time series data based on the multiple specified parameters is used to extract single parameter features on a single time scale to obtain parameter-related features of each specified parameter, including:

[0013] Extracting a single parameter feature on a single time scale based on the traffic time series data to obtain traffic-related features of the traffic time series data;

[0014] performing single parameter feature extraction on a single time scale according to the pressure time series data to obtain pressure-related features of the pressure time series data;

[0015] A single parameter feature is extracted on a single time scale based on the temperature time series data to obtain temperature-related features of the temperature time series data; wherein the parameter-related features include the flow-related features, the pressure-related features and the temperature-related features.

[0016] Optionally, the monitoring time series data based on the multiple specified parameters is subjected to multi-parameter feature extraction at multiple time scales to obtain multi-parameter correlation features, including:

[0017] constructing multi-parameter time series data using the flow time series data, the pressure time series data, and the temperature time series data;

[0018] Multi-parameter feature extraction is performed on multiple time scales based on the multi-parameter time series data to obtain the multi-parameter correlation feature.

[0019] Optionally, generating a maintenance measure suitable for the turbine blade according to the parameter-related feature of each designated parameter and the multi-parameter correlation feature includes:

[0020] classifying the fouling level of the turbine blade based on the parameter-related feature of each designated parameter and the multi-parameter correlation feature to obtain the fouling level of the turbine blade;

[0021] Maintenance measures for the turbine blades are determined according to the fouling level of the turbine blades.

[0022] Optionally, determining the maintenance measures for the turbine blade according to the fouling level of the turbine blade includes any one of the following situations:

[0023] If the fouling level is the first level, generating a maintenance measure to maintain the initial cleaning time unchanged; wherein the fouling situation represented by the first level has an impact on the performance of the turbine blade within an allowable range;

[0024] If the fouling level belongs to the second level, a recommended cleaning time corresponding to the second level is determined, and maintenance measures for shortening the initial cleaning time are generated based on the recommended cleaning time; wherein, the fouling condition represented by the second level has an impact on the performance of the turbine blade that exceeds the allowable range.

[0025] Optionally, generating a maintenance measure suitable for the turbine blade according to the parameter-related feature of each designated parameter and the multi-parameter correlation feature includes:

[0026] predicting a fouling condition of the turbine blade based on the parameter-related characteristics of each designated parameter and the multi-parameter correlation characteristics to obtain a predicted fouling condition of the turbine blade;

[0027] If the predicted scaling condition meets the infrared thermal imaging activation condition, obtaining the blade temperature distribution; wherein the blade temperature distribution is obtained by scanning the turbine blade with infrared thermal imaging equipment;

[0028] Determine appropriate maintenance measures for the turbine blades based on the blade temperature distribution.

[0029] Optionally, the monitoring time series data of the multiple specified parameters are constructed as multi-parameter time series data;

[0030] The monitoring time series data based on the multiple specified parameters, performing single parameter feature extraction on a single time scale to obtain parameter-related features of each specified parameter; and performing multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features, including:

[0031] Inputting the monitoring time series data of the multiple specified parameters into the corresponding single parameter feature extraction branch to perform single parameter feature extraction to obtain parameter-related features of each specified parameter;

[0032] The multi-parameter time series data is input into a multi-parameter feature extraction branch to perform multi-parameter feature extraction to obtain the multi-parameter correlation feature.

[0033] In a second aspect, an embodiment of the present application provides a fault diagnosis device for a gas turbine system, which is applied to a turbine device of the gas turbine system, wherein turbine blades in the turbine device have an initial cleaning time corresponding thereto; the device comprises:

[0034] a time series data acquisition module, configured to acquire monitoring time series data of a plurality of designated parameters of the turbine blade; wherein the designated parameters are parameters affected or changed by the fouling condition on the turbine blade;

[0035] A parameter feature extraction module is configured to perform single parameter feature extraction on a single time scale based on the monitoring time series data of the multiple specified parameters to obtain parameter-related features for each specified parameter; and to perform multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features; wherein the parameter-related features are used to characterize the trend features of the monitoring time series data of the corresponding specified parameters, and the multi-parameter correlation features are used to characterize the correlations between the specified parameters on multiple time scales;

[0036] A maintenance measure generation module is used to generate maintenance measures suitable for the turbine blades based on the parameter-related characteristics of each specified parameter and the multi-parameter correlation characteristics; wherein the maintenance measures include adjustment suggestions for the initial cleaning time.

[0037] In a third aspect, an embodiment of the present application provides a fault diagnosis device for a gas turbine system, comprising a monitoring unit and a data processing unit;

[0038] The monitoring unit is configured to monitor turbine blades of a turbine device of the gas turbine system to obtain monitoring time series data of a plurality of specified parameters of the turbine blades; wherein the specified parameters are parameters affected or changed by fouling on the turbine blades;

[0039] The data processing unit is configured to perform single parameter feature extraction on a single time scale based on the monitoring time series data of the multiple specified parameters to obtain parameter-related features of each specified parameter; and perform multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features;

[0040] A maintenance measure suitable for the turbine blade is generated according to the parameter-related characteristics of each designated parameter and the multi-parameter correlation characteristics; wherein the maintenance measure includes an adjustment suggestion for the initial cleaning time of the turbine blade.

[0041] Optionally, the fault diagnosis device further includes an infrared thermal imaging device;

[0042] The data processing unit is further configured to predict the scaling condition of the turbine blade based on the parameter-related characteristics of each designated parameter and the multi-parameter correlation characteristics, thereby obtaining the predicted scaling condition of the turbine blade;

[0043] The infrared thermal imaging device is used to scan the turbine blade to obtain the blade temperature distribution if the predicted fouling condition meets the infrared thermal imaging activation condition;

[0044] The data processing unit is further used to determine maintenance measures suitable for the turbine blades based on the temperature distribution of the blades. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the specific implementation methods or the description of the prior art. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0046] Figure 1A step diagram of a fault diagnosis method for a gas turbine system provided in an embodiment of the present application;

[0047] Figure 2 A diagram showing the steps for obtaining parameter-related features in an embodiment of the present application;

[0048] Figure 3 Schematic diagram of monitoring positions in a gas turbine according to an embodiment of the present application;

[0049] Figure 4 A diagram showing the steps for obtaining multi-parameter correlation features in an embodiment of the present application;

[0050] Figure 5 This is a diagram of the first step in determining maintenance measures in an embodiment of the present application;

[0051] Figure 6 A step diagram for generating maintenance measures regarding the initial cleaning time in an embodiment of the present application;

[0052] Figure 7 This is a diagram of the second step of determining maintenance measures in the embodiment of the present application;

[0053] Figure 8 A diagram showing the steps of performing feature extraction using a feature extraction model in an embodiment of the present application;

[0054] Figure 9 A module diagram of a fault diagnosis device for a gas turbine system provided in an embodiment of the present application;

[0055] Figure 10 A structural diagram of a fault diagnosis device for a gas turbine system provided in an embodiment of the present application;

[0056] Figure 11 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] To make the purpose, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0058] A gas turbine is a thermodynamic machine that converts high-temperature, high-pressure gas generated by combustion into usable mechanical energy by rotating a turbine. It boasts high thermal efficiency, lightweight, compact size, high power, and fast startup, and is widely used in a variety of fields, including aerospace, marine propulsion, and power generation. Gas turbine systems are complex, with their key components often operating under harsh conditions of high temperature and high pressure. Frequent starts and stops, as well as load fluctuations, make their actual operating conditions complex and variable, increasing the probability of failure and, in turn, impacting their normal operation.

[0059] Common gas turbine system failures include compressor failure, combustion chamber failure, and turbine equipment failure. Turbine equipment failures include turbine blade fouling, turbine blade corrosion, turbine blade wear, and mechanical damage to the turbine equipment. Failures such as turbine blade corrosion, turbine blade wear, and mechanical damage typically require repair or replacement of the turbine blades or turbine equipment to repair the defective portion. Regarding turbine blade fouling, regular cleaning of the turbine blades is commonly used in related technologies to remove the fouling.

[0060] However, in reality, due to the frequent start-up and shutdown of the gas turbine system and load fluctuations, the fouling of the turbine blades in the gas turbine system does not change in a fixed periodic manner. During the regular cleaning process, the turbine blades may only show slight fouling during the regular cleaning, and this slight fouling does not affect the actual operation for a period of time, resulting in a waste of maintenance costs; in other cases, the turbine blades may show severe fouling before the regular cleaning, which in turn affects the normal operation of the gas turbine system. Therefore, the method of relying solely on regular cleaning to remove turbine blade fouling in the related art not only fails to reduce the risk of failure of the gas turbine system, but also results in unnecessary waste of maintenance costs.

[0061] Based on the above problems, the present application provides a fault diagnosis method and device for a gas turbine system, which is applied to a turbine equipment of a gas turbine system, wherein the turbine blades in the turbine equipment have an initial cleaning time. The method includes: obtaining monitoring time series data of multiple specified parameters of the turbine blades; based on the monitoring time series data of the multiple specified parameters, performing single parameter feature extraction on a single time scale to obtain parameter-related features of each specified parameter; and performing multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features; generating maintenance measures suitable for the turbine blades based on the parameter-related features and multi-parameter correlation features of each specified parameter.

[0062] The gas turbine system fault diagnosis method proposed in this application extracts single-parameter features at a single time scale based on multiple monitoring time series data of turbine blades, and extracts multi-parameter features at multiple time scales based on multiple monitoring time series data. This method obtains data features for each of the multiple specified parameters, as well as correlation features between the specified parameters at different time scales. Based on the parameter correlation features of the specified parameters and the multi-parameter correlation features of the multiple specified parameters, a multi-parameter and multi-time scale analysis of the fouling of the turbine blades is performed, thereby improving the accuracy of gas turbine system fault diagnosis, enabling appropriate maintenance measures to be taken for the gas turbine system and reducing the probability of gas turbine system failure.

[0063] Compared with related technologies, the present application performs multi-parameter and multi-time-scale analysis on the fouling of turbine blades, and generates corresponding maintenance measures based on the fouling of turbine blades, so as to adjust the initial cleaning time of turbine blades in real time, thereby improving the cleaning effect of turbine blades, effectively reducing the failure risk of gas turbine system, and at the same time reducing the cost waste for cleaning turbine blades and reducing the maintenance cost of gas turbine system.

[0064] The gas turbine system fault diagnosis method provided in this specification can be applied to the turbine equipment of the gas turbine system to diagnose the fouling of turbine blades in the turbine equipment and generate corresponding maintenance measures. It is understood that, with adaptive modifications, this application can also be used to diagnose faults in other parts of the turbine equipment and other major components within the gas turbine system, such as the compressor or combustor within the gas turbine system.

[0065] According to an embodiment of the present application, an embodiment of a fault diagnosis method for a gas turbine system is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0066] Reference Figure 1 As shown, in this embodiment, a fault diagnosis method for a gas turbine system is provided, which is applied to a turbine device of the gas turbine system, wherein turbine blades in the turbine device have an initial cleaning time. The method includes:

[0067] S100. Acquire monitoring time series data of a plurality of designated parameters of a turbine blade; wherein the designated parameters are parameters that are affected by or cause changes in the scaling condition on the turbine blade.

[0068] S200. Based on the monitoring time series data of multiple specified parameters, single parameter feature extraction is performed on a single time scale to obtain parameter-related features of each specified parameter; and multi-parameter feature extraction is performed on multiple time scales to obtain multi-parameter correlation features; wherein, the parameter-related features are used to characterize the trend characteristics of the monitoring time series data of the corresponding specified parameters, and the multi-parameter correlation features are used to characterize the correlation between the specified parameters on multiple time scales.

[0069] S300. Generate maintenance measures suitable for turbine blades based on parameter-related features and multi-parameter correlation features of each designated parameter; wherein the maintenance measures include adjustment suggestions for initial cleaning time.

[0070] The initial cleaning time can be a pre-set time for periodically cleaning turbine blades in a gas turbine system. The initial cleaning time can be adjusted based on the generated maintenance measures. The time scale can be the time interval used to sample the monitoring time series data. The time scale can be determined based on the time length of the monitoring time series data or the changes in the monitoring time series data. Different time scales can be independent or interrelated.

[0071] Specifically, the present application is applied to turbine equipment in a gas turbine system, and analyzes the fouling of turbine blades by acquiring monitoring time series data of multiple specified parameters of turbine blades in the turbine equipment. The specified parameters can be determined based on the sensitivity of each parameter to the fouling of the turbine blades and the ability of each parameter to reflect the operating state of the turbine blades. It is understood that when fouling occurs on turbine blades, the values ​​of the specified parameters will be significantly affected by the fouling and will show significant changes, and the trend characteristics of the values ​​of the specified parameters affected by the fouling will follow specific patterns. By analyzing the trend characteristics of the specified parameters based on the known specific patterns, the fouling of the turbine blades can be diagnosed.

[0072] The types and reasons for selecting designated parameters are exemplified. The designated parameters may include at least one of the following parameters: flow rate, pressure, temperature, vibration, noise, and energy conversion efficiency of the turbine equipment. When scaling occurs on turbine blades, scaling alters the gas flow characteristics on the blade surface, increasing the flow resistance of the gas passing through the blades. This in turn affects the energy conversion efficiency of the blades, resulting in reduced gas flow and pressure in the turbine equipment and increased temperature within the turbine equipment. Furthermore, scaling on the turbine blades can affect their rotation, causing vibration and mechanical noise during operation, impacting the structural stability of the turbine equipment.

[0073] Furthermore, based on the monitoring time series data of multiple specified parameters, single parameter features are extracted on a single time scale for each of the monitoring time series data of the specified parameters to obtain parameter-related features for each specified parameter. The parameter-related features may be one or more of the maximum value, slope, rate of change, fluctuation frequency, and fluctuation amplitude of the monitoring time series data. It is understood that turbine blade fouling is typically uneven, and therefore turbine blade fouling will cause fluctuating changes in the specified parameters. Fluctuating changes in each specified parameter can be used to characterize the turbine blade fouling from the perspective of a single parameter.

[0074] Furthermore, based on the monitoring time series data of all specified parameters, multi-parameter feature extraction is performed at multiple time scales to obtain multi-parameter correlation features. This multi-parameter feature extraction method can be any of a number of methods, including convolution operations, autocorrelation and cross-correlation analysis, recurrent neural networks, or long short-term memory networks. The multi-parameter correlation features obtained through multi-parameter feature extraction characterize the correlation characteristics between the monitoring time series data of the specified parameters, thereby providing a more comprehensive data foundation for turbine blade fouling analysis and improving the accuracy of gas turbine system fault diagnosis.

[0075] It should be noted that multiple time scales can be determined based on the changes in the monitored time series data. For example, the periodicity of the monitored time series data can be analyzed and multiple time scales can be determined based on the periodicity analysis results. By extracting features at different time scales, short-term and long-term correlation features of the monitored time series data of multiple specified parameters can be obtained, thereby capturing the correlation of the monitored time series data at different time scales.

[0076] Furthermore, by combining the parameter-related characteristics of each specified parameter and the multi-parameter correlation characteristics on multiple time scales, the fouling of turbine blades is analyzed on a multi-parameter and multi-time scale basis based on the complex correlations between different specified parameters and time scales, thereby improving the analysis accuracy and efficiency of turbine blade fouling and thereby improving the accuracy of gas turbine system fault diagnosis.

[0077] Furthermore, after obtaining the parameter-correlation characteristics and multi-parameter correlation characteristics for each specified parameter, the scaling of the turbine blades is analyzed and predicted based on these characteristics to determine the current scaling status of the turbine blades. Based on this current scaling status, the initial cleaning time of the turbine blades is adjusted. For example, if the turbine blades are only slightly scaled, the initial cleaning time can be appropriately delayed. If the turbine blades are severely scaled, the initial cleaning time needs to be advanced to ensure timely cleaning of the turbine blades.

[0078] Compared with related technologies, the present application can accurately analyze and predict the fouling of turbine blades in a gas turbine system based on the parameter-related characteristics and multi-parameter correlation characteristics of each specified parameter, thereby adjusting the initial cleaning time of the turbine blades in real time according to the fouling condition of the turbine blades to adapt to the actual operating conditions of the gas turbine system, effectively improve the cleaning efficiency of the turbine blades, reduce the failure risk of the gas turbine system, and at the same time reduce the maintenance cost of the gas turbine system.

[0079] The gas turbine system fault diagnosis method provided in this embodiment extracts single-parameter features at a single time scale based on multiple monitoring time series data of turbine blades, and extracts multi-parameter features at multiple time scales based on the multiple monitoring time series data. This method obtains data features for each of the multiple specified parameters, as well as correlation features between the specified parameters at different time scales. Based on the parameter correlation features of the specified parameters and the multi-parameter correlation features of the multiple specified parameters, the fouling condition of the turbine blades is analyzed at multiple parameters and multiple time scales, thereby improving the accuracy of gas turbine system fault diagnosis, enabling appropriate maintenance measures to be taken for the gas turbine system and reducing the probability of gas turbine system failure.

[0080] Compared with related technologies, the present application performs multi-parameter and multi-time-scale analysis on the fouling of turbine blades, and generates corresponding maintenance measures based on the fouling of turbine blades, so as to adjust the initial cleaning time of turbine blades in real time, thereby improving the cleaning effect of turbine blades, effectively reducing the failure risk of gas turbine system, and at the same time reducing the cost waste for cleaning turbine blades and reducing the maintenance cost of gas turbine system.

[0081] Reference Figure 2 As shown, as an embodiment of the present application, the multiple specified parameters include flow, pressure, and temperature at corresponding monitoring positions of the turbine equipment; the monitoring time series data includes flow time series data, pressure time series data, and temperature time series data; based on the monitoring time series data of the multiple specified parameters, single parameter feature extraction is performed on a single time scale to obtain parameter-related features of each specified parameter, including:

[0082] S210. Perform single parameter feature extraction on a single time scale based on the traffic time series data to obtain traffic-related features of the traffic time series data.

[0083] S220. Perform single parameter feature extraction on a single time scale based on the pressure time series data to obtain pressure-related features of the pressure time series data.

[0084] S230. Perform single parameter feature extraction on a single time scale based on the temperature time series data to obtain temperature-related features of the temperature time series data; wherein the parameter-related features include flow-related features, pressure-related features, and temperature-related features.

[0085] Reference Figure 3 As shown in the figure, multiple sections of the gas turbine system are selected for parameter monitoring, including ① the gas turbine system inlet, ② the compressor inlet, ③ the compressor exhaust port, ④ the compressor outlet, ⑤ the combustor inlet, ⑥ the combustor outlet, ⑦ the turbine inlet, and ⑧ the turbine outlet. The pressure and temperature at the gas turbine system inlet are the same as those of the atmosphere, and the flow rate is related to the operating conditions of the gas turbine system. The corresponding monitoring locations of the turbine include the turbine inlet and turbine outlet. Under normal operation of the gas turbine system, the flow rates at the turbine inlet and turbine outlet are the same and are related to the flow rate of the system section preceding the turbine. Because the gas enters the turbine and exchanges energy with the turbine blades, the pressure at the turbine outlet is typically lower than the pressure at the turbine inlet, and the temperature at the turbine outlet is typically lower than the temperature at the turbine inlet. Similarly, the pressure and temperature at the turbine inlet and turbine outlet are also related to the pressure and temperature of the system section preceding the turbine. Therefore, the flow, pressure and temperature at the turbine equipment inlet and turbine equipment outlet not only characterize the fault condition of the turbine equipment itself, but also can characterize the fault condition of the entire gas turbine system to a certain extent.

[0086] Specifically, parameter-related features include flow-related features, pressure-related features, and temperature-related features. Parameter-related features can be one or more of the following features: the maximum value, slope, rate of change, fluctuation frequency, and fluctuation amplitude of the monitoring time series data. The slope of the monitoring time series data can be used to analyze the changing trend of the monitoring time series data to identify the failure risk of the gas turbine system. The rate of change of the monitoring time series data represents the instantaneous change speed of the monitoring time series data, so that sudden failures of the gas turbine system can be analyzed. By extracting single parameter features from the flow time series data, pressure time series data, and temperature time series data on a single time scale, the turbine equipment failures can be analyzed and predicted from the perspective of multiple parameters, thereby improving the accuracy and precision of fault diagnosis.

[0087] Reference Figure 4 As shown, as an embodiment of the present application, based on the monitoring time series data of multiple specified parameters, multi-parameter feature extraction is performed on multiple time scales to obtain multi-parameter correlation features, including:

[0088] S240. Construct multi-parameter time series data using flow time series data, pressure time series data, and temperature time series data.

[0089] S250. Perform multi-parameter feature extraction on multiple time scales based on the multi-parameter time series data to obtain multi-parameter correlation features.

[0090] Specifically, the process of constructing multi-parameter time series data can be as follows: based on the sampling frequencies of the time series data, the flow time series data, the pressure time series data, and the temperature time series data are represented as matrix sequences of corresponding lengths. The matrix sequences obtained from the flow time series data, the pressure time series data, and the temperature time series data are merged, and the missing data due to the different lengths of the time series data are filled with null values ​​to construct the multi-parameter time series data.

[0091] Furthermore, data sampling is performed on the multi-parameter time series data at multiple time scales to obtain multi-parameter time series data at different time scales. Multi-parameter feature extraction is performed on the multi-parameter time series data at different time scales to obtain multi-parameter correlation features. Among them, the method of multi-parameter feature extraction can be any one of the methods such as convolution operation, autocorrelation and cross-correlation analysis, recursive neural network or long short-term memory network. It should be noted that at different time scales, the flow time series data, pressure time series data and temperature time series data may have different correlations. By performing multi-parameter feature extraction on the multi-parameter time series data at multiple time scales, the correlation of the monitoring time series data at different time scales can be learned based on the multi-parameter correlation features, effectively capturing the correlation characteristics between the monitoring time series data, and improving the accuracy of the analysis and prediction of the fouling of the turbine blades.

[0092] Furthermore, by combining the parameter-related characteristics of flow, pressure and temperature and the multi-parameter correlation characteristics on multiple time scales, the fouling of turbine blades is analyzed on a multi-parameter and multi-time scale basis based on the complex correlation between multiple parameters and time scales, thereby improving the analysis accuracy and efficiency of turbine blade fouling and thereby improving the accuracy of gas turbine system fault diagnosis.

[0093] Reference Figure 5 As shown, as an embodiment of the present application, a maintenance measure suitable for a turbine blade is generated based on the parameter-related characteristics and multi-parameter correlation characteristics of each specified parameter, including:

[0094] S310. Classify the fouling level of the turbine blade based on the parameter-related characteristics and multi-parameter correlation characteristics of each designated parameter to obtain the fouling level of the turbine blade.

[0095] S320. Determine maintenance measures for turbine blades based on the fouling level of the turbine blades.

[0096] Specifically, a classification model can be pre-built based on the parameter-related features and multi-parameter correlation features of the specified parameters. The classification model can be any of a variety of models, including logistic regression, support vector machine, decision tree, and naive Bayesian classifiers. The parameter-related features and multi-parameter correlation features of the specified parameters are input into the classification model, and the classification model is used to classify the turbine blade fouling level based on the parameter-related features and multi-parameter correlation features. The turbine blade fouling level is obtained and serves as the basis for generating maintenance measures. The turbine blade fouling level can be pre-set and is typically related to the thickness, material type, and coverage of the fouling layer on the turbine blade.

[0097] To illustrate the fouling levels of turbine blades, the fouling levels can be divided into four categories based on the fouling characteristics: mild, moderate, severe, and extreme. Mild fouling is characterized by a thin fouling layer, which has little noticeable impact on turbine blade performance. Moderate fouling is characterized by a significantly increased fouling layer, which covers a large area of ​​the turbine blade, impacting performance. Severe fouling is characterized by a thicker fouling layer that covers a large area of ​​the turbine blade, significantly impacting performance. Extreme fouling is characterized by a thick and hard fouling layer that may damage the turbine blades, significantly impacting the operation of the turbine equipment and gas turbine system. Mild and moderate fouling can be treated with regular cleaning to ensure normal operation of the gas turbine system. Severe and extreme fouling require thorough cleaning or replacement of the turbine blades to remove the fouling.

[0098] Reference Figure 6 As shown, as an embodiment of the present application, determining the maintenance measures for the turbine blades according to the fouling level of the turbine blades includes any of the following situations:

[0099] S322. If the fouling level is level 1, generate maintenance measures to maintain the initial cleaning time unchanged; wherein the fouling situation represented by the first level has an impact on the performance of the turbine blades within an allowable range.

[0100] S324. If the fouling level is level 2, determine a recommended cleaning time corresponding to the level 2, and generate maintenance measures to shorten the initial cleaning time based on the recommended cleaning time; wherein the fouling situation characterized by level 2 has an impact on the performance of the turbine blades that exceeds an allowable range.

[0101] Specifically, the first level in this embodiment may include mild and moderate fouling. In these cases, the fouling layer has a minimal impact on turbine blade performance. Regular cleaning can remove the fouling from the turbine blades, ensuring normal operation of the gas turbine system. The second level in this embodiment may include severe and extreme fouling. In these cases, the fouling layer has a significant impact on turbine blade performance. Regular cleaning alone cannot effectively ensure normal operation of the gas turbine system. The turbine blades require immediate cleaning, and the cleaning intensity is greater than that of turbine blades in the first level. In some embodiments, if the impact of the fouling layer on turbine blade performance exceeds the allowable range, the impact of the fouling layer can be eliminated by replacing the turbine blades.

[0102] Reference Figure 7 As shown, as an embodiment of the present application, a maintenance measure suitable for a turbine blade is generated based on the parameter-related characteristics and multi-parameter correlation characteristics of each specified parameter, including:

[0103] S330. Predicting the scaling condition of the turbine blade based on the parameter-related characteristics and multi-parameter correlation characteristics of each designated parameter to obtain a predicted scaling condition of the turbine blade.

[0104] S340. If the predicted scaling condition meets the conditions for enabling infrared thermal imaging, obtain the blade temperature distribution; wherein the blade temperature distribution is obtained by scanning the turbine blade with infrared thermal imaging equipment.

[0105] S350. Determine appropriate maintenance measures for turbine blades based on blade temperature distribution.

[0106] Specifically, when scale forms on turbine blades, it can reduce heat exchange efficiency in areas covered by the scale layer, leading to localized temperature increases and the formation of hot spots on the blade surface. This localized heat accumulation in the turbine blades can increase thermal stress within the blades. Prolonged exposure to this condition can lead to deformation, cracks, or other structural damage, shortening the blade's service life and impacting the normal operation of the turbine equipment and gas turbine system.

[0107] Based on these issues, in this embodiment, the scaling condition of the turbine blades is predicted based on the parameter-related characteristics and multi-parameter correlation characteristics of each specified parameter, the predicted scaling condition of the turbine blades is obtained, and a judgment is made based on the activation conditions of infrared thermal imaging. If the scaling condition on the turbine blades is relatively mild, it will not affect the heat exchange efficiency of the turbine blades. At this time, the turbine blades will not have the problem of uneven temperature distribution, so there is no need to activate the infrared thermal imaging equipment. If the scaling condition on the turbine blades is relatively serious, the turbine blades may have local heat accumulation. Therefore, in order to accurately analyze the status of the turbine blades, it is necessary to activate the infrared thermal imaging equipment to scan the turbine blades to obtain the blade temperature distribution. According to the blade temperature distribution, the scaling location on the turbine blade can be determined, so that targeted cleaning and maintenance can be carried out to optimize the working condition of the turbine blades.

[0108] Reference Figure 8 As shown, as an embodiment of the present application, the monitoring time series data of multiple specified parameters are constructed as multi-parameter time series data; based on the monitoring time series data of multiple specified parameters, single parameter feature extraction is performed on a single time scale to obtain parameter-related features of each specified parameter; and multi-parameter feature extraction is performed on multiple time scales to obtain multi-parameter correlation features, including:

[0109] S260. Input the monitoring time series data of multiple specified parameters into the corresponding single parameter feature extraction branch to perform single parameter feature extraction to obtain parameter-related features of each specified parameter.

[0110] S270. Input the multi-parameter time series data into the multi-parameter feature extraction branch to perform multi-parameter feature extraction to obtain multi-parameter correlation features.

[0111] Specifically, a feature extraction model can be pre-constructed based on the parameter-related features and multi-parameter correlation features of multiple specified parameters. The feature extraction model includes a single-parameter feature extraction branch and a multi-parameter feature extraction branch. Multiple single-parameter feature extraction branches, one corresponding to each specified parameter, are used to extract single-parameter features from the monitoring time series data of the specified parameter, while the multi-parameter feature extraction branch is used to extract multi-parameter features from the multi-parameter time series data. The single-parameter feature extraction branch and the multi-parameter feature extraction branch are parallel structures, and the feature data output by both branches is used together to analyze and predict the fouling of turbine blades.

[0112] Furthermore, the monitoring time series data for each specified parameter is input into the corresponding single-parameter feature extraction branch to extract single-parameter features on a single time scale, thereby obtaining parameter-correlation features for each specified parameter. The multi-parameter time series data is input into the multi-parameter feature extraction branch to extract multi-parameter features on multiple time scales, thereby obtaining the multi-parameter correlation features. The parameter-correlation features and multi-parameter correlation features are used together to analyze and predict the scaling of the turbine blades, thereby determining the turbine blade scaling status and generating corresponding maintenance measures.

[0113] Accordingly, refer to Figure 9 As shown, an embodiment of the present application provides a fault diagnosis device for a gas turbine system, which is applied to a turbine device of the gas turbine system, wherein turbine blades in the turbine device have an initial cleaning time corresponding thereto; the device comprises:

[0114] The time series data acquisition module 910 is used to acquire monitoring time series data of multiple specified parameters of the turbine blades; wherein the specified parameters are parameters that are affected or changed by the fouling condition on the turbine blades.

[0115] The parameter feature extraction module 920 is used to perform single parameter feature extraction on a single time scale based on the monitoring time series data of multiple specified parameters to obtain parameter-related features of each specified parameter; and to perform multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features; wherein the parameter-related features are used to characterize the trend features of the monitoring time series data of the corresponding specified parameters, and the multi-parameter correlation features are used to characterize the correlation between the specified parameters on multiple time scales.

[0116] The maintenance measure generation module 930 is used to generate maintenance measures suitable for turbine blades based on the parameter-related characteristics and multi-parameter correlation characteristics of each specified parameter; wherein the maintenance measures include adjustment suggestions for the initial cleaning time.

[0117] In some optional embodiments, the plurality of specified parameters include flow, pressure, and temperature at corresponding monitoring locations of the turbine equipment; the monitoring time series data includes flow time series data, pressure time series data, and temperature time series data; and the parameter feature extraction module 920 includes:

[0118] The flow-related feature extraction unit is used to extract single parameter features on a single time scale based on the flow time series data to obtain flow-related features of the flow time series data.

[0119] The pressure-related feature extraction unit is used to extract a single parameter feature on a single time scale based on the pressure time series data to obtain the pressure-related features of the pressure time series data.

[0120] The temperature-related feature extraction unit is used to extract a single parameter feature on a single time scale based on the temperature time series data to obtain the temperature-related features of the temperature time series data; wherein the parameter-related features include flow-related features, pressure-related features and temperature-related features.

[0121] In some optional implementations, the parameter feature extraction module 920 further includes:

[0122] The multi-parameter time series data construction unit is used to construct multi-parameter time series data using flow time series data, pressure time series data and temperature time series data.

[0123] The multi-parameter correlation feature extraction unit is used to extract multi-parameter features on multiple time scales based on multi-parameter time series data to obtain multi-parameter correlation features.

[0124] In some optional implementations, the maintenance measure generation module 930 includes:

[0125] The turbine blade fouling grade classification unit is used to classify the turbine blade fouling grade based on the parameter-related features and multi-parameter correlation features of each specified parameter to obtain the turbine blade fouling grade.

[0126] The maintenance measure determination unit is used to determine the maintenance measures of the turbine blades according to the fouling level of the turbine blades.

[0127] In some optional implementations, the maintenance measure determination unit includes:

[0128] The first maintenance measure generating subunit is used to generate maintenance measures. If the fouling level belongs to the first level, a maintenance measure is generated to maintain the initial cleaning time unchanged; wherein the fouling situation represented by the first level has an impact on the performance of the turbine blade within an allowable range.

[0129] The second maintenance measure generation subunit is used to generate maintenance measures. If the fouling level belongs to the second level, the recommended cleaning time corresponding to the second level is determined, and a maintenance measure for shortening the initial cleaning time is generated based on the recommended cleaning time; wherein, the fouling condition represented by the second level has an impact on the performance of the turbine blades beyond the allowable range.

[0130] In some optional implementations, the maintenance measure generation module 930 further includes:

[0131] The turbine blade fouling condition prediction unit is used to predict the fouling condition of the turbine blade based on the parameter-related characteristics and multi-parameter correlation characteristics of each specified parameter, so as to obtain the predicted fouling condition of the turbine blade.

[0132] The temperature distribution acquisition unit is used to acquire the temperature distribution. If the predicted scaling condition meets the infrared thermal imaging activation condition, the blade temperature distribution is acquired. The blade temperature distribution is acquired by scanning the turbine blade with infrared thermal imaging equipment.

[0133] The maintenance measure determination unit is used to determine the maintenance measures suitable for the turbine blades according to the temperature distribution of the blades.

[0134] In some optional implementations, the monitoring time series data of multiple specified parameters are constructed as multi-parameter time series data; the parameter feature extraction module 920 further includes:

[0135] The single parameter feature extraction unit is used to input the monitoring time series data of multiple specified parameters into the corresponding single parameter feature extraction branch to perform single parameter feature extraction, and obtain parameter-related features of each specified parameter.

[0136] The multi-parameter feature extraction unit is used to input the multi-parameter time series data into the multi-parameter feature extraction branch to perform multi-parameter feature extraction to obtain multi-parameter correlation features.

[0137] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0138] The fault diagnosis device of the gas turbine system in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0139] Reference Figure 10 As shown, an embodiment of the present application further provides a fault diagnosis device for a gas turbine system, including a monitoring unit 1010 and a data processing unit 1020 .

[0140] The monitoring unit 1010 is used to monitor the turbine blades of the turbine equipment of the gas turbine system and obtain monitoring time series data of multiple specified parameters of the turbine blades; wherein the specified parameters are parameters that are affected or changed by the fouling condition on the turbine blades.

[0141] The data processing unit 1020 is used to perform single parameter feature extraction on a single time scale based on the monitoring time series data of multiple specified parameters to obtain parameter-related features of each specified parameter; and to perform multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features; generate maintenance measures suitable for turbine blades based on the parameter-related features and multi-parameter correlation features of each specified parameter; wherein the maintenance measures include adjustment suggestions for the initial cleaning time of the turbine blades.

[0142] In some embodiments, the fault diagnosis apparatus further includes an infrared thermal imaging device 1030 .

[0143] The data processing unit 1020 is further configured to predict the fouling condition of the turbine blade based on the parameter-related characteristics and multi-parameter correlation characteristics of each designated parameter, and obtain the predicted fouling condition of the turbine blade.

[0144] The infrared thermal imaging device 1030 is used to scan the turbine blades. If the predicted fouling situation meets the infrared thermal imaging activation condition, the turbine blades are scanned to obtain the blade temperature distribution.

[0145] The data processing unit 1020 is further used to determine appropriate maintenance measures for the turbine blades based on the temperature distribution of the blades.

[0146] See also Figure 11 , Figure 11 1 is a structural diagram of a computer device provided by an embodiment of the present application. As shown in the figure, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to an interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 11 A processor 10 is taken as an example.

[0147] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0148] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0149] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0150] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0151] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0152] The embodiments of the present application also provide a computer-readable storage medium. The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0153] An embodiment of the present application provides a computer program product, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform a method according to any embodiment of the present application.

[0154] Although the embodiments of the present application have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations shall fall within the scope defined by the appended claims.

[0155] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0156] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing this application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0157] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0158] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0159] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0160] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0161] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0162] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0163] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

[0164] Although the embodiments of the present application have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations shall fall within the scope defined by the appended claims.

Claims

1. A method for diagnosing a fault in a gas turbine system, characterized in that: The turbine device applied to the gas turbine system has an initial cleaning time corresponding to the turbine blades in the turbine device, and the initial cleaning time is a pre-set time for regularly cleaning the turbine blades of the gas turbine system; The method comprises: Selecting multiple system sections in the gas turbine system for parameter monitoring to obtain monitoring time series data of multiple specified parameters of the turbine blades; wherein the specified parameters are parameters that are affected or changed by the fouling condition on the turbine blades; Based on the monitoring time series data of the multiple specified parameters, single parameter feature extraction is performed on a single time scale to obtain parameter-related features of each specified parameter; and multi-parameter feature extraction is performed on multiple time scales to obtain multi-parameter correlation features; wherein, the multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features includes: data sampling of the multi-parameter time series data on multiple time scales to obtain multi-parameter time series data at different time scales; multi-parameter feature extraction is performed on the multi-parameter time series data at different time scales to obtain multi-parameter correlation features; the parameter correlation features are used to characterize the trend features of the monitoring time series data of the corresponding specified parameters, and the multi-parameter correlation features are used to characterize the correlation between the specified parameters on multiple time scales; Generating maintenance measures suitable for the turbine blades according to the parameter-related characteristics and the multi-parameter correlation characteristics of each specified parameter, including: inputting the parameter-related characteristics and the multi-parameter correlation characteristics into a pre-constructed classification model to analyze and predict the scaling condition of the turbine blades to obtain the predicted scaling condition of the turbine blades; if the predicted scaling condition meets the infrared thermal imaging activation condition, scanning the turbine blades with an infrared thermal imaging device to obtain the blade temperature distribution to more accurately obtain the scaling condition of the turbine blades; determining maintenance measures suitable for the turbine blades according to the blade temperature distribution; wherein the maintenance measures include adjustment suggestions for the initial cleaning time.

2. The method according to claim 1, characterized in that The plurality of specified parameters include flow, pressure and temperature at corresponding monitoring locations of the turbine equipment; the monitoring time series data includes flow time series data, pressure time series data and temperature time series data; The monitoring time series data based on the multiple specified parameters is used to extract single parameter features on a single time scale to obtain parameter-related features of each specified parameter, including: Extracting a single parameter feature on a single time scale based on the traffic time series data to obtain traffic-related features of the traffic time series data; performing single parameter feature extraction on a single time scale according to the pressure time series data to obtain pressure-related features of the pressure time series data; A single parameter feature is extracted on a single time scale based on the temperature time series data to obtain temperature-related features of the temperature time series data; wherein the parameter-related features include the flow-related features, the pressure-related features and the temperature-related features.

3. The method according to claim 2, characterized in that The monitoring time series data based on the multiple specified parameters is subjected to multi-parameter feature extraction at multiple time scales to obtain multi-parameter correlation features, including: constructing multi-parameter time series data using the flow time series data, the pressure time series data, and the temperature time series data; Multi-parameter feature extraction is performed on multiple time scales based on the multi-parameter time series data to obtain the multi-parameter correlation feature.

4. The method according to claim 1, wherein Generating maintenance measures suitable for the turbine blade according to the parameter-related characteristics of each designated parameter and the multi-parameter correlation characteristics includes: classifying the fouling level of the turbine blade based on the parameter-related feature of each designated parameter and the multi-parameter correlation feature to obtain the fouling level of the turbine blade; Maintenance measures for the turbine blades are determined according to the fouling level of the turbine blades.

5. The method according to claim 4, characterized in that Determining maintenance measures for the turbine blades according to the fouling level of the turbine blades includes: If the fouling level belongs to the first level, a maintenance measure is generated to maintain the initial cleaning time unchanged; wherein the fouling situation represented by the first level has an impact on the performance of the turbine blade that is within an allowable range.

6. The method according to claim 5, characterized in that The determining of maintenance measures for the turbine blades according to the fouling level of the turbine blades further includes: If the fouling level belongs to the second level, a recommended cleaning time corresponding to the second level is determined, and maintenance measures for shortening the initial cleaning time are generated based on the recommended cleaning time; wherein, the fouling condition represented by the second level has an impact on the performance of the turbine blade that exceeds the allowable range.

7. The method according to claim 1, characterized in that The monitoring time series data of the multiple specified parameters are constructed as multi-parameter time series data; The monitoring time series data based on the multiple specified parameters are used to extract single parameter features on a single time scale to obtain parameter-related features of each specified parameter; Furthermore, multi-parameter feature extraction is performed on multiple time scales to obtain multi-parameter correlation features, including: Inputting the monitoring time series data of the multiple specified parameters into the corresponding single parameter feature extraction branch to perform single parameter feature extraction to obtain parameter-related features of each specified parameter; The multi-parameter time series data is input into a multi-parameter feature extraction branch to perform multi-parameter feature extraction to obtain the multi-parameter correlation feature.

8. A fault diagnosis device for a gas turbine system, characterized in that: The turbine device applied to the gas turbine system has an initial cleaning time corresponding to the turbine blades in the turbine device, and the initial cleaning time is a pre-set time for regularly cleaning the turbine blades of the gas turbine system; The device comprises: a time series data acquisition module, configured to select a plurality of system sections in the gas turbine system for parameter monitoring, and acquire monitoring time series data of a plurality of specified parameters of the turbine blades; wherein the specified parameters are parameters affected or changed by the fouling condition on the turbine blades; A parameter feature extraction module is used to perform single parameter feature extraction on a single time scale based on the monitoring time series data of the multiple specified parameters to obtain parameter-related features of each specified parameter; and to perform multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features; wherein the multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features includes: sampling the multi-parameter time series data on multiple time scales to obtain multi-parameter time series data at different time scales; performing multi-parameter feature extraction on the multi-parameter time series data at different time scales to obtain multi-parameter correlation features; the parameter correlation features are used to characterize the trend features of the monitoring time series data of the corresponding specified parameters, and the multi-parameter correlation features are used to characterize the correlation between the specified parameters on multiple time scales; A maintenance measure generation module is used to generate maintenance measures suitable for the turbine blade based on the parameter-related characteristics of each specified parameter and the multi-parameter correlation characteristics, including: inputting the parameter-related characteristics and the multi-parameter correlation characteristics into a pre-constructed classification model based on the parameter-related characteristics and the multi-parameter correlation characteristics to analyze and predict the scaling condition of the turbine blade to obtain the predicted scaling condition of the turbine blade; if the predicted scaling condition meets the infrared thermal imaging activation condition, scanning the turbine blade with an infrared thermal imaging device to obtain the blade temperature distribution to more accurately obtain the scaling condition of the turbine blade; determining maintenance measures suitable for the turbine blade based on the blade temperature distribution; wherein the maintenance measures include adjustment suggestions for the initial cleaning time.

9. A fault diagnosis device for a gas turbine system, characterized in that: including a monitoring unit and a data processing unit; The monitoring unit is configured to select a plurality of system sections in the gas turbine system for parameter monitoring, monitor turbine blades of a turbine device of the gas turbine system, and obtain monitoring time series data of a plurality of specified parameters of the turbine blades; wherein the specified parameters are parameters affected or changed by fouling on the turbine blades; The data processing unit is configured to perform single parameter feature extraction on a single time scale based on the monitoring time series data of the multiple specified parameters to obtain parameter-related features of each specified parameter; and perform multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features; wherein the performing multi-parameter feature extraction on multiple time scales to obtain multi-parameter correlation features includes: performing data sampling on the multi-parameter time series data on multiple time scales to obtain multi-parameter time series data at different time scales; and performing multi-parameter feature extraction on the multi-parameter time series data at different time scales to obtain multi-parameter correlation features; The data processing unit is also used to generate maintenance measures suitable for the turbine blades based on the parameter-related characteristics of each specified parameter and the multi-parameter correlation characteristics, including: based on the parameter-related characteristics and the multi-parameter correlation characteristics, the parameter-related characteristics and the multi-parameter correlation characteristics are jointly input into a pre-constructed classification model to analyze and predict the scaling condition of the turbine blades to obtain the predicted scaling condition of the turbine blades; if the predicted scaling condition meets the infrared thermal imaging activation condition, the turbine blades are scanned by infrared thermal imaging equipment to obtain the blade temperature distribution to more accurately obtain the scaling condition of the turbine blades; and the maintenance measures suitable for the turbine blades are determined based on the blade temperature distribution; wherein the maintenance measures include adjustment suggestions for the initial cleaning time of the turbine blades.

10. The device according to claim 9, characterized in that The fault diagnosis device also includes an infrared thermal imaging device; The data processing unit is further configured to predict the scaling condition of the turbine blade based on the parameter-related characteristics of each designated parameter and the multi-parameter correlation characteristics, thereby obtaining the predicted scaling condition of the turbine blade; The infrared thermal imaging device is used to scan the turbine blades, and if the predicted fouling condition meets the infrared thermal imaging activation condition, the turbine blades are scanned to obtain the blade temperature distribution; The data processing unit is further used to determine maintenance measures suitable for the turbine blades based on the temperature distribution of the blades.

Images