Automatic optimization method and system for VMD decomposition parameters of thermal parameter sequence of gas turbine
By constructing a proxy model and optimizing the VMD decomposition parameters, the problem of low computational efficiency of gas turbine thermodynamic parameter sequences was solved, the automatic search for the global optimal solution was achieved, the decomposition accuracy and efficiency were improved, and gas turbine fault diagnosis was supported.
Patent Information
- Application Number
- CN202511459634.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-13
- Publication Date
- 2026-02-10
AI Technical Summary
In existing technologies, the VMD decomposition parameters of gas turbine thermodynamic parameter sequences are inefficient and it is difficult to obtain the global optimal solution.
An automatic optimization method is adopted, which optimizes the VMD decomposition parameters, including the search space of the modality number K and the penalty factor α, by constructing a surrogate model and acquisition function, and performs iterative optimization to find the global optimum.
It enables intelligent and efficient optimization of VMD decomposition parameters for gas turbine thermodynamic parameter sequences, improving decomposition accuracy and efficiency, and providing reliable technical support for gas turbine fault diagnosis.
Smart Images
Figure CN121503199A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent operation and maintenance technology for gas turbines, and in particular to an automatic optimization method and system for VMD decomposition parameters of gas turbine thermodynamic parameter sequences. Background Technology
[0002] Gas turbines generate massive sequences of high-precision (second-level or even millisecond-level) thermodynamic parameters during operation, such as temperature, pressure, flow rate, and power. These thermodynamic parameters contain crucial information characterizing unit performance degradation, component failures, and operational status. Effectively extracting fault characteristics and trend information from these complex, non-stationary, and nonlinear signals is a core challenge in achieving intelligent operation and maintenance and predictive health management of gas turbines. Variational Mode Decomposition (VMD), as an advanced adaptive signal processing method, is widely used in vibration signal analysis of rotating machinery due to its good noise robustness, and in recent years has been extended to the field of gas turbine thermodynamic parameter analysis. However, in the application of gas turbine thermodynamic parameter analysis, the selection of VMD decomposition parameters heavily relies on expert experience and manual trial and error, resulting in low computational efficiency and difficulty in obtaining globally optimal solutions. Summary of the Invention
[0003] This invention aims to at least solve the technical problems of low computational efficiency and difficulty in obtaining the global optimal solution of VMD decomposition parameters for gas turbine thermodynamic parameter sequences in the prior art. In particular, it innovatively proposes an automatic optimization method and system for VMD decomposition parameters of gas turbine thermodynamic parameter sequences.
[0004] On one hand, the present invention provides an automatic optimization method for VMD decomposition parameters of a gas turbine thermodynamic parameter sequence, the method comprising: S1. Collect the operating data of the gas turbine, preprocess the collected operating data, screen the key gas turbine thermodynamic parameters and normalize each thermodynamic parameter into an equal time interval sequence, and analyze each screened gas turbine thermodynamic parameter separately. S2. Using the number of modes K and the penalty factor α from VMD decomposition as optimization variables, and based on prior knowledge, set the search space for the optimization variables for each gas turbine thermodynamic parameter. ; S3, in the search space Internally, through spatial sampling, N initial parameter combinations of optimization variables are selected, VMD decomposition is performed on the gas turbine thermodynamic parameters, and the corresponding total envelope entropy is calculated. This forms the initial observation dataset D; S4. Construct a proxy model for the objective function. Perform probabilistic modeling to obtain the results at any untested point. The posterior probability distribution of the objective function value; S5. Define a data acquisition function A(·), which calculates the expected utility of untested points in the parameter space based on the posterior distribution provided by the surrogate model, and determines the optimal evaluation point. ; S6. Use the obtained optimal evaluation point Perform VMD decomposition on the gas turbine thermodynamic parameters to calculate their true objective function values. Add new observation data to the existing dataset D, and update the surrogate model with the expanded dataset D; S7. Repeat steps S4-S6 to form a closed-loop iterative process of "building a proxy model - optimizing the acquisition function - evaluating new points - updating the proxy model". When the number of iterations reaches the preset maximum number of evaluations, or the expected improvement value of multiple consecutive iterations is less than the set threshold, the iteration terminates. Select the optimal parameter combination from all evaluated parameter combinations. The smallest set is used as the optimal parameters. .
[0005] As an optional embodiment of the present invention, in step S1, the operating data may include gas turbine temperature, pressure, flow rate, and unit power.
[0006] As an optional embodiment of the present invention, the sampling accuracy of the running data may be no less than the second level.
[0007] As an optional embodiment of the present invention, the selection of key gas turbine thermodynamic parameters may include real monitoring data and data calculated based on thermodynamic mechanisms. The data calculated based on thermodynamic mechanisms includes compressor inlet / outlet air pressure and temperature, combustion chamber exhaust pressure, turbine exhaust temperature and turbine in-wheel temperature. The data calculated based on thermodynamic mechanisms includes compressor efficiency, turbine efficiency and turbine inlet temperature.
[0008] As an optional embodiment of the present invention, in step S3, the spatial sampling method is the Latin hypercube sampling method.
[0009] As an optional embodiment of the present invention, in step S5, a Gaussian process is used as a surrogate model.
[0010] As an optional embodiment of the present invention, the desired improvement in EI can be used as the acquisition function.
[0011] On the other hand, the present invention also provides an automatic optimization system for VMD decomposition parameters of gas turbine thermodynamic parameter sequences, used to execute the above-described automatic optimization method for VMD decomposition parameters of gas turbine thermodynamic parameter sequences. The system includes a data acquisition and preprocessing module, an automatic optimization module for VMD decomposition parameters, and a result display module. The data acquisition and preprocessing module is used to acquire gas turbine operating data and perform data preprocessing. The VMD decomposition parameter automatic optimization module is used to execute the VMD decomposition parameter automatic optimization method for a gas turbine thermodynamic parameter sequence. The results display module is used to display the VMD decomposition results of the gas turbine thermodynamic parameters, including each IMF component and the envelope entropy.
[0012] On the other hand, the present invention also provides a computer-readable storage medium, comprising: A memory on which computer programs are stored; A processor is configured to execute the program in the memory to implement the above-described method for automatic optimization of VMD decomposition parameters of gas turbine thermodynamic parameter sequences.
[0013] On the other hand, the present invention also provides an electronic device, comprising: One or more processors; A storage unit is used to store one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the VMD decomposition parameter automatic optimization method for the gas turbine thermodynamic parameter sequence described above.
[0014] The beneficial effects of this invention are as follows: This invention achieves automatic optimization of VMD decomposition parameters for gas turbine thermodynamic parameter sequences through steps such as: collecting and preprocessing historical operating data of gas turbines; setting the search space for optimization variables (modal number K and penalty factor α); selecting initial combinations of optimization variables for VMD decomposition and calculating the total envelope entropy to construct an initial observation dataset; constructing a surrogate model to probabilistically model the total envelope entropy; constructing a collection function to obtain the optimal evaluation point; using the optimal evaluation point and optimized variable combination for VMD decomposition to calculate the total envelope entropy and update the dataset and surrogate model; and performing iterative calculations of "constructing the surrogate model - optimizing the collection function - evaluating new points - updating the surrogate model" to minimize the total envelope entropy by selecting the modal number K and penalty factor α. By constructing a surrogate model and a collection function, this invention can intelligently and efficiently find the globally optimal VMD decomposition parameters for gas turbine thermodynamic parameter sequences, improving the accuracy and efficiency of parameter decomposition and providing more reliable technical support for gas turbine fault diagnosis.
[0015] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description
[0016] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1This is a flowchart of an automatic optimization method for VMD decomposition parameters of a gas turbine thermodynamic parameter sequence provided by the present invention; Figure 2 This is a structural block diagram of an automatic optimization system for VMD decomposition parameters of gas turbine thermodynamic parameter sequences provided by the present invention. Detailed Implementation
[0017] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.
[0018] Example 1 S1. Collect the operating data of the gas turbine, preprocess the collected operating data, screen the key gas turbine thermodynamic parameters and normalize each thermodynamic parameter into an equal time interval sequence, and analyze each screened gas turbine thermodynamic parameter separately. S2. Using the number of modes K and the penalty factor α from VMD decomposition as optimization variables, and based on prior knowledge, set the search space for the optimization variables for each gas turbine thermodynamic parameter. ; S3, in the search space Internally, through spatial sampling, N initial parameter combinations of optimization variables are selected, VMD decomposition is performed on the gas turbine thermodynamic parameters, and the corresponding total envelope entropy is calculated. This forms the initial observation dataset D; S4. Construct a proxy model for the objective function. Perform probabilistic modeling to obtain the results at any untested point. The posterior probability distribution of the objective function value; S5. Define a data acquisition function A(·), which calculates the expected utility of untested points in the parameter space based on the posterior distribution provided by the surrogate model, and determines the optimal evaluation point. ; S6. Use the obtained optimal evaluation point Perform VMD decomposition on the gas turbine thermodynamic parameters to calculate their true objective function values. Add new observation data to the existing dataset D, and update the surrogate model with the expanded dataset D; S7. Repeat steps S4-S6 to form a closed-loop iterative process of "building a proxy model - optimizing the acquisition function - evaluating new points - updating the proxy model". When the number of iterations reaches the preset maximum number of evaluations, or the expected improvement value of multiple consecutive iterations is less than the set threshold, the iteration terminates. Select the optimal parameter combination from all evaluated parameter combinations. The smallest set is used as the optimal parameters. .
[0019] As an optional embodiment of the present invention, in step S1, the operating data may include gas turbine temperature, pressure, flow rate, and unit power.
[0020] As an optional embodiment of the present invention, the sampling accuracy of the running data may be no less than the second level.
[0021] As an optional embodiment of the present invention, the selection of key gas turbine thermodynamic parameters may include real monitoring data and data calculated based on thermodynamic mechanisms. The data calculated based on thermodynamic mechanisms includes compressor inlet / outlet air pressure and temperature, combustion chamber exhaust pressure, turbine exhaust temperature and turbine in-wheel temperature. The data calculated based on thermodynamic mechanisms includes compressor efficiency, turbine efficiency and turbine inlet temperature.
[0022] As an optional embodiment of the present invention, in step S3, the spatial sampling method is the Latin hypercube sampling method.
[0023] As an optional embodiment of the present invention, in step S5, a Gaussian process is used as a surrogate model.
[0024] As an optional embodiment of the present invention, the desired improvement in EI can be used as the acquisition function.
[0025] Example 2 like Figure 2 As shown, an automatic optimization system for VMD decomposition parameters of a gas turbine thermodynamic parameter sequence is used to implement the automatic optimization method for VMD decomposition parameters of a gas turbine thermodynamic parameter sequence described in Example 1.
[0026] The system includes a data acquisition and preprocessing module, a VMD decomposition parameter automatic optimization module, and a result display module. The data acquisition and preprocessing module is used to acquire gas turbine operating data and perform data preprocessing. The VMD decomposition parameter automatic optimization module is used to execute the aforementioned method for automatic optimization of VMD decomposition parameters of a gas turbine thermodynamic parameter sequence. The result display module is used to display the VMD decomposition results of gas turbine thermodynamic parameters, including information such as each IMF component and envelope entropy.
[0027] In this embodiment, the principle of automatic optimization of VMD decomposition parameters for a gas turbine thermodynamic parameter sequence is as follows: The system first acquires the operating data of the gas turbine through a data acquisition and preprocessing module, and preprocesses these raw signals. The preprocessed signals are sent to the VMD decomposition parameter automatic optimization module, which uses the automatic optimization method of VMD decomposition parameters for a gas turbine thermodynamic parameter sequence, based on the principle of minimizing the total envelope entropy, to automatically calculate the optimal VMD decomposition parameters for each thermodynamic parameter of the gas turbine. Finally, the result display module presents the VMD decomposition results of each thermodynamic parameter of the gas turbine in an intuitive way, including information such as each IMF component and envelope entropy.
[0028] Example 3 A computer-readable storage medium comprising: A memory that stores computer programs.
[0029] A processor is configured to execute the program in the memory to implement the VMD decomposition parameter automatic optimization method for a gas turbine thermodynamic parameter sequence as described in Embodiment 1. The computer-readable medium may be included in the apparatus, device, or system of the present invention, or it may exist independently.
[0030] The computer-readable storage medium may be any tangible medium that contains or stores a program, and may be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device. More specific examples include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, optical fibers, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.
[0031] The computer-readable storage medium may also include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code, specific examples of which include, but are not limited to, electromagnetic signals, optical signals, or any suitable combination thereof.
[0032] This disclosure also provides an electronic device including one or more processors.
[0033] A storage unit is used to store one or more programs, which, when executed by one or more processors, enable the one or more processors to implement an automatic optimization method for VMD decomposition parameters based on a gas turbine thermodynamic parameter sequence as described above.
[0034] It should be noted here that the electronic device in the embodiments of this disclosure may also include an input device and an output device. The processor, memory, input device, and output device may be connected via a bus or other means, without specific limitations herein.
[0035] As a computer-readable storage medium, the memory can be used to store software programs, computer-executable programs, and various modules, such as the program or module corresponding to the VMD decomposition parameter automatic optimization method for a gas turbine thermodynamic parameter sequence according to an embodiment of this disclosure. The processor executes various functional applications and data processing of the electronic device by running the software programs or modules stored in the memory.
[0036] Input devices can be used to receive input digital numbers or signals. These signals can be key signals related to user settings and function control of the device / terminal / server. Output devices can include display devices such as screens.
[0037] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims
1. An automatic optimization method for VMD decomposition parameters of a gas turbine thermodynamic parameter sequence, characterized in that, The method includes: S1. Collect the operating data of the gas turbine, preprocess the collected operating data, screen the key gas turbine thermodynamic parameters and normalize each thermodynamic parameter into an equal time interval sequence, and analyze each screened gas turbine thermodynamic parameter separately. S2. Using the number of modes K and the penalty factor α from VMD decomposition as optimization variables, and based on prior knowledge, set the search space for the optimization variables for each gas turbine thermodynamic parameter. ; S3, in the search space Internally, through spatial sampling, N initial parameter combinations of optimization variables are selected, VMD decomposition is performed on the gas turbine thermodynamic parameters, and the corresponding total envelope entropy is calculated. This forms the initial observation dataset D; S4. Construct a proxy model for the objective function. Perform probabilistic modeling to obtain the results at any untested point. The posterior probability distribution of the objective function value; S5. Define a data acquisition function A(·), which calculates the expected utility of untested points in the parameter space based on the posterior distribution provided by the surrogate model, and determines the optimal evaluation point. ; S6. Use the obtained optimal evaluation point Perform VMD decomposition on the gas turbine thermodynamic parameters to calculate their true objective function values. Add new observation data to the existing dataset D, and update the surrogate model with the expanded dataset D; S7. Repeat steps S4-S6 to form a closed-loop iterative process of "building a proxy model - optimizing the acquisition function - evaluating new points - updating the proxy model". When the number of iterations reaches the preset maximum number of evaluations, or the expected improvement value of multiple consecutive iterations is less than the set threshold, the iteration terminates. Select the optimal parameter combination from all evaluated parameter combinations. The smallest set is used as the optimal parameters. .
2. The automatic optimization method for VMD decomposition parameters of gas turbine thermodynamic parameter sequences according to claim 1, characterized in that, In step S1, the operating data includes gas turbine temperature, pressure, flow rate, and unit power.
3. The automatic optimization method for VMD decomposition parameters of gas turbine thermodynamic parameter sequences according to claim 2, characterized in that, The sampling accuracy of the running data is no less than the second level.
4. The automatic optimization method for VMD decomposition parameters of gas turbine thermodynamic parameter sequences according to claim 1, characterized in that, The key gas turbine thermodynamic parameters selected include real monitoring data and data calculated based on thermodynamic mechanisms. The data calculated based on thermodynamic mechanisms include compressor inlet / outlet air pressure and temperature, combustion chamber exhaust pressure, turbine exhaust temperature, and turbine in-wheel temperature. The data calculated based on thermodynamic mechanisms also include compressor efficiency, turbine efficiency, and turbine inlet temperature.
5. The automatic optimization method for VMD decomposition parameters of gas turbine thermodynamic parameter sequences according to claim 1, characterized in that, In step S3, the spatial sampling method is the Latin hypercube sampling method.
6. The automatic optimization method for VMD decomposition parameters of gas turbine thermodynamic parameter sequences according to claim 1, characterized in that, In step S5, a Gaussian process is used as the surrogate model.
7. The automatic optimization method for VMD decomposition parameters of gas turbine thermodynamic parameter sequences according to claim 1, characterized in that, Use the expected increase in EI as the acquisition function.
8. An automatic parameter optimization system for VMD decomposition of gas turbine thermodynamic parameter sequences, used to execute the automatic parameter optimization method for VMD decomposition of gas turbine thermodynamic parameter sequences according to any one of claims 1-7, characterized in that, The system includes a data acquisition and preprocessing module, a VMD decomposition parameter automatic optimization module, and a result display module; The data acquisition and preprocessing module is used to acquire gas turbine operating data and perform data preprocessing. The VMD decomposition parameter automatic optimization module is used to execute the VMD decomposition parameter automatic optimization method for a gas turbine thermodynamic parameter sequence. The results display module is used to display the VMD decomposition results of the gas turbine thermodynamic parameters, including each IMF component and the envelope entropy.
9. A computer-readable storage medium, characterized in that, include: A memory on which computer programs are stored; A processor is configured to execute the program in the memory to implement the automatic optimization method for VMD decomposition parameters of the gas turbine thermodynamic parameter sequence according to any one of claims 1-7.
10. An electronic device, characterized in that, include: One or more processors; A storage unit is used to store one or more programs that, when executed by one or more processors, enable the one or more processors to implement the VMD decomposition parameter automatic optimization method for the gas turbine thermodynamic parameter sequence as described in any one of claims 1-7.