Flywheel energy storage system health degree assessment method and device

Abstract

The invention relates to the technical field of flywheel energy storage, and particularly provides a flywheel energy storage system health degree evaluation method and device, and the method comprises the steps: combining monitoring parameters and working condition parameters of a flywheel energy storage system according to a preset arrangement rule, and obtaining a pixel matrix corresponding to the flywheel energy storage system; taking the pixel matrix as the input of a pre-trained convolutional auto-encoder to obtain a reconstructed data matrix output by the decoding process of the pre-trained convolutional auto-encoder; the difference degree between the pixel matrix and the reconstructed data matrix is obtained, if the difference degree exceeds a threshold value, the flywheel energy storage system is in a healthy state, and otherwise, the flywheel energy storage system is in an unhealthy state. In the technical scheme provided by the invention, the pre-trained convolutional auto-encoder adopts unsupervised learning, does not need to label data, can learn low-dimensional representation of the data, is suitable for dimensionality reduction and feature extraction, is simple in structure, is easy to implement and expand, and can perform real-time online evaluation on the flywheel system.

Description

Technical Field

[0001] The present invention relates to the technical field of flywheel energy storage, and in particular to a method and device for evaluating the health of a flywheel energy storage system. Background Art

[0002] Flywheel energy storage systems have important applications in grid frequency regulation, backup power, and renewable energy integration. Because flywheel systems involve high-speed rotation, precise control, and complex mechanical structures, their health assessment is crucial.

[0003] In the existing technology, the health of the flywheel energy storage system is evaluated through a physical model-based method. Based on the dynamic characteristics of the flywheel rotor, parameters such as vibration and speed are analyzed. Finite element analysis is used to simulate the stress distribution and fatigue life of the flywheel rotor. By analyzing possible failure modes and their impacts, the system health status is evaluated.

[0004] However, flywheel systems involve the coupling of multiple physical fields (such as mechanical, electrical, and thermodynamic), making accurate modeling extremely complex. Parameters in actual systems (such as material properties and friction coefficients) may contain uncertainties, affecting model accuracy. High-precision models are computationally intensive, making real-time health assessment difficult.

[0005] Alternatively, the health of the flywheel energy storage system can be assessed using a signal processing-based method. By analyzing the vibration signal and swing signal of the flywheel system, such as the frequency domain and time domain characteristics, fault characteristics can be identified and compared with standard values ​​or health values ​​to perform a health assessment.

[0006] However, noise in real-world environments can affect the accuracy of signal analysis, making it difficult to extract certain fault characteristics from the signal. Analyzing the health of a specific component using only a single signal is incapable of comprehensively assessing the health of the flywheel system. Summary of the Invention

[0007] In order to overcome the above-mentioned defects, the present invention proposes a flywheel energy storage system health assessment method and device.

[0008] In a first aspect, a flywheel energy storage system health assessment method is provided, the flywheel energy storage system health assessment method comprising:

[0009] The monitoring parameters and operating parameters of the flywheel energy storage system are combined according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system;

[0010] Using the pixel matrix as input to a pre-trained convolutional autoencoder, and obtaining a reconstructed data matrix output by a decoding process of the pre-trained convolutional autoencoder;

[0011] A difference between the pixel matrix and the reconstructed data matrix is ​​obtained. If the difference exceeds a threshold, the flywheel energy storage system is in a healthy state; otherwise, the flywheel energy storage system is in an unhealthy state.

[0012] Preferably, the monitoring parameters include sensor parameters monitored at different positions of the flywheel energy storage system, and the operating condition parameters include at least one of the following: active power, reactive power, and rotational speed.

[0013] Furthermore, the sensor parameter is a vibration parameter.

[0014] Preferably, the training process of the pre-trained convolutional autoencoder includes:

[0015] Acquire monitoring parameters and operating parameters of a flywheel energy storage system in a healthy state, and combine the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system;

[0016] A convolutional autoencoder is used to perform unsupervised learning on the data.

[0017] Furthermore, in the process of using the convolutional autoencoder to perform unsupervised learning on the data, the training parameters of the encoder and decoder in the convolutional autoencoder are optimized by the backpropagation algorithm with the goal of minimizing the difference between the pixel matrix and the reconstructed data matrix.

[0018] Preferably, the difference between the pixel matrix and the reconstructed data matrix is ​​as follows:

[0019]

[0020] In the above formula, is the pixel matrix x and the reconstruction data matrix The difference between them, n is the total number of elements in the pixel matrix, x i is the value corresponding to the i-th element in the pixel matrix, is the value corresponding to the i-th element in the reconstructed data matrix.

[0021] In a second aspect, a flywheel energy storage system health assessment device is provided, the flywheel energy storage system health assessment device comprising:

[0022] A combination module is used to combine the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system;

[0023] An analysis module, configured to use the pixel matrix as input to a pre-trained convolutional autoencoder to obtain a reconstructed data matrix output by a decoding process of the pre-trained convolutional autoencoder;

[0024] The judgment module is used to obtain the difference between the pixel matrix and the reconstructed data matrix. If the difference exceeds a threshold, the flywheel energy storage system is in a healthy state; otherwise, the flywheel energy storage system is in an unhealthy state.

[0025] In a third aspect, a computer device is provided, comprising: one or more processors;

[0026] The processor is configured to store one or more programs;

[0027] When the one or more programs are executed by the one or more processors, the flywheel energy storage system health assessment method is implemented.

[0028] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed, the flywheel energy storage system health assessment method is implemented.

[0029] The above one or more technical solutions of the present invention have at least one or more of the following beneficial effects:

[0030] The present invention provides a method and device for evaluating the health of a flywheel energy storage system, comprising: combining the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system; using the pixel matrix as the input of a pre-trained convolutional autoencoder to obtain a reconstructed data matrix output by the decoding process of the pre-trained convolutional autoencoder; obtaining the difference between the pixel matrix and the reconstructed data matrix, and if the difference exceeds a threshold, the flywheel energy storage system is in a healthy state; otherwise, the flywheel energy storage system is in an unhealthy state. In the technical solution provided by the present invention, the pre-trained convolutional autoencoder adopts unsupervised learning, does not require labeled data, can learn a low-dimensional representation of data, is suitable for dimensionality reduction and feature extraction, has a simple structure, is easy to implement and expand, and can perform real-time online evaluation of the flywheel system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 1 is a flow chart of the main steps of the flywheel energy storage system health assessment method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0034] Example 1

[0035] See attached Figure 1 , Figure 1 FIG1 is a flow chart showing the main steps of a method for evaluating the health of a flywheel energy storage system according to an embodiment of the present invention. Figure 1 As shown, the flywheel energy storage system health assessment method in the embodiment of the present invention mainly includes the following steps:

[0036] Step S101: combining the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system;

[0037] Step S102: using the pixel matrix as input to a pre-trained convolutional autoencoder to obtain a reconstructed data matrix output by a decoding process of the pre-trained convolutional autoencoder;

[0038] Step S103: obtaining a difference between the pixel matrix and the reconstructed data matrix. If the difference exceeds a threshold, the flywheel energy storage system is in a healthy state; otherwise, the flywheel energy storage system is in an unhealthy state.

[0039] In this embodiment, the monitoring parameters include sensor parameters monitored at different positions of the flywheel energy storage system, and the operating condition parameters include at least one of the following: active power, reactive power, and rotational speed.

[0040] Wherein, the sensor parameter is a vibration parameter.

[0041] In a specific embodiment, the definition of the preset arrangement rule requires that the monitoring quantities associated with the working condition be distributed around the working condition parameters. When the convolutional neural network convolution kernel is 3*3 and the step size is 1, it can be ensured that the working condition parameters and the monitoring quantities associated therewith are covered by the convolution kernel every time the convolution kernel moves one step, for example, as shown in Table 1:

[0042] Table 1

[0043] Vibration 1 Vibration 4 Vibration 6 Vibration 1 Vibration 4 Vibration 6 Vibration 1 Vibration 4 Vibration 6 Vibration 2 Meritorious Vibration 7 Vibration 2 reactive power Vibration 7 Vibration 2 Speed Vibration 7 Vibration 3 Vibration 5 Vibration 8 Vibration 3 Vibration 5 Vibration 8 Vibration 3 Vibration 5 Vibration 8

[0044] Select m × n measurement points from the flywheel system condition monitoring and arrange them into an m × n matrix according to the rules in Table 1 (m and n can be equal and are both preset integers). This matrix is ​​equivalent to a single-channel grayscale image. Each element in the matrix is ​​a pixel, and each element value is a pixel value. Each image is equivalent to a [1, m, n] matrix. Alternatively, select m × n measurement points from the flywheel system condition monitoring and arrange them into a 3 × m × n matrix according to the rules in 1. This matrix is ​​equivalent to an RGB color image.

[0045] For another example, the operating condition parameters can be gathered together, and then the monitoring quantities associated with them can be distributed around the operating condition parameters, as shown in Table 2 below. The convolution kernel size is set to 4*4 and the step size is 1. This can also ensure that the operating condition parameters and the monitoring quantities associated with them are covered by the convolution kernel every time the convolution kernel moves one step.

[0046] Table 2

[0047] Vibration 1 Vibration 4 Vibration 6 Vibration 8 Vibration 1 Vibration 4 Vibration 6 Vibration 8 Vibration 2 Meritorious Speed Vibration 9 Vibration 2 Meritorious Speed Vibration 9 Vibration 3 Vibration 5 Vibration 7 Vibration 10 Vibration 3 Vibration 5 Vibration 7 Vibration 10 Vibration 1 Vibration 4 Vibration 6 Vibration 8 Vibration 1 Vibration 4 Vibration 6 Vibration 8 Vibration 2 Meritorious Speed Vibration 9 Vibration 2 Meritorious Speed Vibration 9 Vibration 3 Vibration 5 Vibration 7 Vibration 10 Vibration 3 Vibration 5 Vibration 7 Vibration 10

[0048] In this embodiment, the training process of the pre-trained convolutional autoencoder includes:

[0049] Acquire monitoring parameters and operating parameters of a flywheel energy storage system in a healthy state, and combine the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system;

[0050] A convolutional autoencoder is used to perform unsupervised learning on the data.

[0051] In one embodiment, in the process of performing unsupervised learning on the data using a convolutional autoencoder, the training parameters of the encoder and decoder in the convolutional autoencoder are optimized by a backpropagation algorithm with the goal of minimizing the difference between the pixel matrix and the reconstructed data matrix.

[0052] In this embodiment, the difference between the pixel matrix and the reconstructed data matrix is ​​as follows:

[0053]

[0054] In the above formula, is the pixel matrix x and the reconstruction data matrix The difference between them, n is the total number of elements in the pixel matrix, x i is the value corresponding to the i-th element in the pixel matrix, is the value corresponding to the i-th element in the reconstructed data matrix.

[0055] Example 2

[0056] Based on the same inventive concept, the present invention also provides a flywheel energy storage system health assessment device, the flywheel energy storage system health assessment device comprising:

[0057] A combination module is used to combine the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system;

[0058] An analysis module, configured to use the pixel matrix as input to a pre-trained convolutional autoencoder to obtain a reconstructed data matrix output by a decoding process of the pre-trained convolutional autoencoder;

[0059] The judgment module is used to obtain the difference between the pixel matrix and the reconstructed data matrix. If the difference exceeds a threshold, the flywheel energy storage system is in a healthy state; otherwise, the flywheel energy storage system is in an unhealthy state.

[0060] Preferably, the monitoring parameters include sensor parameters monitored at different positions of the flywheel energy storage system, and the operating condition parameters include at least one of the following: active power, reactive power, and rotational speed.

[0061] Furthermore, the sensor parameter is a vibration parameter.

[0062] Preferably, the training process of the pre-trained convolutional autoencoder includes:

[0063] Acquire monitoring parameters and operating parameters of a flywheel energy storage system in a healthy state, and combine the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system;

[0064] A convolutional autoencoder is used to perform unsupervised learning on the data.

[0065] Furthermore, in the process of using the convolutional autoencoder to perform unsupervised learning on the data, the training parameters of the encoder and decoder in the convolutional autoencoder are optimized by the backpropagation algorithm with the goal of minimizing the difference between the pixel matrix and the reconstructed data matrix.

[0066] Preferably, the difference between the pixel matrix and the reconstructed data matrix is ​​as follows:

[0067]

[0068] In the above formula, is the pixel matrix x and the reconstruction data matrix The difference between them, n is the total number of elements in the pixel matrix, x i is the value corresponding to the i-th element in the pixel matrix, is the value corresponding to the i-th element in the reconstructed data matrix.

[0069] Example 3

[0070] Based on the same inventive concept, the present invention also provides a computer device, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a flywheel energy storage system health assessment method in the above embodiment.

[0071] Example 4

[0072] Based on the same inventive concept, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It can be understood that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of a flywheel energy storage system health assessment method in the above embodiment.

[0073] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention 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.

[0074] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, 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 flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0075] 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.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating 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 The steps for the function specified in one or more boxes.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A flywheel energy storage system health assessment method, characterized in that: The method comprises: The monitoring parameters and operating parameters of the flywheel energy storage system are combined according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system; Using the pixel matrix as input to a pre-trained convolutional autoencoder, and obtaining a reconstructed data matrix output by a decoding process of the pre-trained convolutional autoencoder; A difference between the pixel matrix and the reconstructed data matrix is ​​obtained. If the difference exceeds a threshold, the flywheel energy storage system is in a healthy state; otherwise, the flywheel energy storage system is in an unhealthy state.

2. The method according to claim 1, wherein The monitoring parameters include sensor parameters monitored at different positions of the flywheel energy storage system, and the operating condition parameters include at least one of the following: active power, reactive power, and rotational speed.

3. The method according to claim 2, wherein The sensor parameter is a vibration parameter.

4. The method according to claim 1, wherein The training process of the pre-trained convolutional autoencoder includes: Acquire monitoring parameters and operating parameters of a flywheel energy storage system in a healthy state, and combine the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system; A convolutional autoencoder is used to perform unsupervised learning on the data.

5. The method according to claim 4, wherein In the process of using the convolutional autoencoder to perform unsupervised learning on the data, the training parameters of the encoder and decoder in the convolutional autoencoder are optimized by the back propagation algorithm with the goal of minimizing the difference between the pixel matrix and the reconstructed data matrix.

6. The method according to claim 1, wherein The difference between the pixel matrix and the reconstructed data matrix is ​​as follows: In the above formula, is the pixel matrix x and the reconstruction data matrix The difference between them, n is the total number of elements in the pixel matrix, x i is the value corresponding to the i-th element in the pixel matrix, is the value corresponding to the i-th element in the reconstructed data matrix.

7. A device based on the flywheel energy storage system health assessment method according to any one of claims 1 to 6, characterized in that: The device comprises: A combination module is used to combine the monitoring parameters and operating parameters of the flywheel energy storage system according to a preset arrangement rule to obtain a pixel matrix corresponding to the flywheel energy storage system; An analysis module, configured to use the pixel matrix as input to a pre-trained convolutional autoencoder to obtain a reconstructed data matrix output by a decoding process of the pre-trained convolutional autoencoder; The judgment module is used to obtain the difference between the pixel matrix and the reconstructed data matrix. If the difference exceeds a threshold, the flywheel energy storage system is in a healthy state; otherwise, the flywheel energy storage system is in an unhealthy state.

8. A computer device, characterized in that: include: one or more processors; The processor is configured to execute one or more programs; When the one or more programs are executed by the one or more processors, the flywheel energy storage system health assessment method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that A computer program is stored thereon, and when the computer program is executed, the flywheel energy storage system health assessment method according to any one of claims 1 to 6 is implemented.

Images