High-pressure valve-side sleeve modeling and prediction method based on a POD reduced-order model

Abstract

The application provides a high-pressure valve-side sleeve modeling and prediction method based on a POD reduced-order model, which comprises the following steps: S1. acquiring parameters of a target high-pressure valve-side sleeve, wherein the parameters comprise operating parameters and geometric parameters; S2. constructing a finite element simulation model of the high-pressure valve-side sleeve based on the parameters of the high-pressure valve-side sleeve, and determining a physical field distribution state of the high-pressure valve-side sleeve; S3. constructing a POD reduced-order model, and inputting the physical field distribution state into the POD reduced-order model for offline training; and S4. acquiring real-time high-pressure valve-side sleeve parameters, and inputting the real-time parameters into the POD reduced-order model to obtain a to-be-measured target of the current high-pressure valve-side sleeve; the method can quickly and accurately perform state prediction and evaluation on the high-pressure valve-side sleeve, and provides accurate and reliable data support for analysis and judgment of stability of a converter transformer and a power grid system and development of operation and maintenance measures.

Description

Technical Field

[0001] This invention relates to a method for predicting the condition of power equipment, and more particularly to a method for modeling and predicting the condition of high-voltage valve side bushings based on a reduced-order POD model. Background Technology

[0002] The high-voltage valve side bushing is a key insulating component on the converter transformer (commonly known as a converter transformer) that connects the converter valve inside the valve hall to the external high-voltage busbar (or line). Its condition seriously affects the safe operation of the converter transformer and the stability of the entire power grid system.

[0003] In the operation of high-pressure valve-side bushings, it is necessary to accurately assess their operating status. In existing technologies, the status is generally determined by obtaining the operating parameters of the high-pressure valve-side bushing. However, due to the complexity of the high-pressure valve-side bushing structure, such as its multi-scale and multi-medium coupling characteristics in its internal structure (from millimeter-level capacitor plates to meter-level overall structure, from metal conductors and epoxy resin insulation to SF6 gas medium, forming a complex three-dimensional composite system of "conductor-insulation-shielding"), it is impossible to accurately determine the current operating status of the high-pressure valve-side bushing based solely on a single operating parameter, thereby affecting the safe operation of the power grid system.

[0004] Therefore, in order to solve the above-mentioned technical problems, it is urgent to propose a new technical approach. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a modeling and prediction method for high-pressure valve-side bushings based on a reduced-order POD model. By obtaining the structural parameters of the high-pressure valve-side bushing, a corresponding finite element simulation model is established. The electromagnetic field, temperature field, and internal medium flow field states are determined by the operating parameters, and these three states form a coupled relationship. The POD reduced-order model is trained using the coupled three field states, thereby enabling rapid and accurate state prediction and evaluation of the high-pressure valve-side bushing. This provides accurate and reliable data support for the analysis and judgment of the stability of converter transformers and power grid systems, as well as the formulation of operation and maintenance measures.

[0006] This invention provides a method for modeling and predicting high-pressure valve side bushings based on a reduced-order POD model, comprising the following steps:

[0007] S1. Obtain the parameters of the target high-pressure valve side bushing, including operating parameters and geometric parameters;

[0008] S2. Construct a finite element simulation model of the high-pressure valve side bushing based on its parameters, and determine the physical field distribution state of the high-pressure valve side bushing.

[0009] S3. Construct a POD reduced-order model and input the physical field distribution state into the POD reduced-order model for offline training;

[0010] S4. Obtain the real-time parameters of the high-pressure valve side bushing and input the real-time parameters into the POD reduced-order model to obtain the target of the current high-pressure valve side bushing.

[0011] Furthermore, the geometric parameters include overall dimensional parameters, radial section parameters, insulation and capacitor structure parameters, and accessory parameters;

[0012] The operating parameters include operating current, temperature, and ambient temperature.

[0013] Furthermore, the physical field distribution includes electromagnetic field distribution, temperature distribution, and flow field distribution.

[0014] Furthermore, based on the operating current of the high-pressure bushing, the electromagnetic field distribution inside the bushing is calculated using the finite element method, and the electromagnetic loss of the high-pressure valve side bushing is determined from the electromagnetic field distribution.

[0015] Furthermore, using electromagnetic loss as a heat source, the temperature field distribution inside the high-pressure bushing was calculated using the finite element method.

[0016] Furthermore, the temperature field distribution is input into the CFD algorithm model to determine the flow field distribution of the high-pressure valve side sleeve.

[0017] Furthermore, when acquiring real-time high-pressure valve side bushing parameters, the Latin hypercube sampling method is used to perform high-density sampling in regions with strong nonlinearity, and the Latin hypercube sampling method is used to perform low-density sampling in regions with strong nonlinearity.

[0018] The beneficial effects of this invention are as follows: By obtaining the structural parameters of the high-pressure valve side bushing, a corresponding finite element simulation model is established. The electromagnetic field, temperature field, and internal medium flow field states are determined by the operating parameters, and these three states form a coupled relationship. The POD reduced-order model is trained using the coupled three field states, thereby enabling rapid and accurate state prediction and evaluation of the high-pressure valve side bushing. This provides accurate and reliable data support for the analysis and judgment of the stability of converter transformers and power grid systems, as well as the formulation of operation and maintenance measures. Attached Figure Description

[0019] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0020] Figure 1 This is a schematic diagram of the process of the present invention.

[0021] Figure 2 This is a schematic diagram of the sampling distribution of the present invention.

[0022] Figure 3 This is a schematic diagram of potential changes in a specific example of the present invention.

[0023] Figure 4The graph shows the potential and temperature changes in a specific example of the present invention. Detailed Implementation

[0024] The present invention will be further described in detail below:

[0025] This invention provides a method for modeling and predicting high-pressure valve side bushings based on a reduced-order POD model, comprising the following steps:

[0026] S1. Obtain the parameters of the target high-pressure valve side bushing, including operating parameters and geometric parameters;

[0027] S2. Construct a finite element simulation model of the high-pressure valve side bushing based on its parameters, and determine the physical field distribution state of the high-pressure valve side bushing.

[0028] S3. Construct a POD reduction model by inputting the physical field distribution state into the POD reduction model for offline training; the POD reduction model can be built using existing methods, which will not be elaborated here.

[0029] S4. Obtain real-time parameters of the high-pressure valve-side bushing and input these parameters into the POD reduced-order model to obtain the current test target of the high-pressure valve-side bushing. This invention establishes a corresponding finite element simulation model by obtaining the structural parameters of the high-pressure valve-side bushing. The electromagnetic field, temperature field, and internal medium flow field states are determined using operating parameters, and these three states form a coupled relationship. The POD reduced-order model is trained using these coupled three-field states, enabling rapid and accurate state prediction and evaluation of the high-pressure valve-side bushing. This provides accurate and reliable data support for the analysis and judgment of the stability of converter transformers and power grid systems, as well as the formulation of operation and maintenance measures.

[0030] In this embodiment, the geometric parameters include overall size parameters, radial section parameters, insulation and capacitor structure parameters, and accessory parameters;

[0031] The operating parameters include operating current, temperature, and ambient temperature.

[0032] Specifically, the physical field distribution includes electromagnetic field distribution, temperature distribution, and flow field distribution.

[0033] Based on the operating current of the high-pressure bushing, the electromagnetic field distribution inside the bushing is calculated using the finite element method, and the electromagnetic loss of the high-pressure valve side bushing is determined from the electromagnetic field distribution. The finite element method is an existing technology and will not be elaborated here.

[0034] Using electromagnetic loss as the heat source, the temperature field distribution inside the high-pressure bushing is calculated using the finite element method.

[0035] The temperature field distribution is input into the CFD algorithm model to determine the flow field distribution of the high-pressure valve side bushing. By coupling the three fields as described above, the influence of multiple factors can be considered during the prediction process, ensuring accuracy. Generally, the state prediction of the high-pressure bushing involves predicting both temperature and flow field conditions. The CFD algorithm used is an existing algorithm and will not be elaborated upon here.

[0036] In this embodiment, when acquiring real-time high-pressure valve side bushing parameters, the Latin hypercube sampling method is used for high-density sampling of regions with strong nonlinearity, and the Latin hypercube sampling method is used for low-density sampling of regions with strong nonlinearity. Figure 2 As shown: For data such as temperature and current, the nonlinearity of the data is relatively strong within the bolded range, so it is necessary to increase the number of sampling points. However, the nonlinearity of the data outside the range is weaker and tends to be linear, so only a few sampling points are needed. This method can reduce the amount of data and improve the computational efficiency.

[0037] The following specific examples further illustrate this point:

[0038] The potential physical quantities of the valve-side bushing are analyzed under normal operating conditions. The calculation results after order reduction by this invention are, for example... Figure 3 As shown, where Figure 3 (a) is the potential cloud diagram at the top of the center conductor of the valve side bushing, which is drawn in the data processing software after the full-order data of the finite element simulation is exported. Figure 3 (b) is a potential cloud map drawn using the data processing software after reducing the order of the full-order data. The two maps were drawn using the same data processing software. A comparison shows that the potential values ​​of the two maps are roughly the same, and the potential distribution and trend of change are basically consistent. The error of the circled part is 0.0178kV (0.006%), which is relatively small.

[0039] Figure 4 (a) and Figure 4 (b) Further, a comparison graph is provided showing the full-order simulation data of the potential and temperature at a sampling point on the valve-side bushing and the results of this invention. The solid red line represents the calculation results of the full-order data, and the dashed blue line represents the calculation results of this invention. Clearly, the two curves in the two graphs largely overlap, demonstrating the feasibility and accuracy of the method.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for modeling and predicting high-pressure valve-side bushings based on a reduced-order POD model, characterized in that: Includes the following steps: S1. Obtain the parameters of the target high-pressure valve side bushing, including operating parameters and geometric parameters; S2. Construct a finite element simulation model of the high-pressure valve side bushing based on its parameters, and determine the physical field distribution state of the high-pressure valve side bushing. S3. Construct a POD reduced-order model and input the physical field distribution state into the POD reduced-order model for offline training; S4. Obtain the real-time parameters of the high-pressure valve side bushing and input the real-time parameters into the POD reduced-order model to obtain the target of the current high-pressure valve side bushing.

2. The high-pressure valve-side bushing modeling and prediction method based on the POD reduced-order model according to claim 1, characterized in that: The geometric parameters include overall dimensional parameters, radial section parameters, insulation and capacitor structure parameters, and accessory parameters; The operating parameters include operating current, temperature, and ambient temperature.

3. The high-pressure valve-side bushing modeling and prediction method based on the POD reduced-order model according to claim 2, characterized in that: The physical field distribution includes electromagnetic field distribution, temperature distribution, and flow field distribution.

4. The high-pressure valve-side bushing modeling and prediction method based on the POD reduced-order model according to claim 3, characterized in that: Based on the operating current of the high-pressure bushing, the electromagnetic field distribution inside the bushing is calculated using the finite element method, and the electromagnetic loss of the high-pressure valve side bushing is determined from the electromagnetic field distribution.

5. The high-pressure valve-side bushing modeling and prediction method based on the POD reduced-order model according to claim 4, characterized in that: Using electromagnetic loss as the heat source, the temperature field distribution inside the high-pressure bushing is calculated using the finite element method.

6. The high-pressure valve-side bushing modeling and prediction method based on the POD reduced-order model according to claim 5, characterized in that: The temperature field distribution is input into the CFD algorithm model to determine the flow field distribution of the high-pressure valve side sleeve.

7. The high-pressure valve-side bushing modeling and prediction method based on the POD reduced-order model according to claim 1, characterized in that: When acquiring real-time high-pressure valve side bushing parameters, the Latin hypercube sampling method is used to perform high-density sampling in regions with strong nonlinearity, and the Latin hypercube sampling method is used to perform low-density sampling in regions with strong nonlinearity.

Images