High-altitude direct-current converter station valve hall hardware fitting structure parameter optimization method
By establishing a simulation model and training a BP neural network, the structure of high-altitude valve hall fittings was optimized, solving the problems of inaccurate parameters and low efficiency in existing technologies, and achieving stable operation of the power system.
Patent Information
- Application Number
- CN202411003333.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-08
AI Technical Summary
In existing technologies, the optimization of valve hall hardware structural parameters under high-altitude conditions lacks accuracy, leading to reduced power system stability and a complex and inefficient process.
A simulation model was established, and a BP neural network was used to predict the arrangement of fittings. The structural parameters of the fittings were optimized by considering the influence of multiple factors.
This improves the accuracy and optimization efficiency of hardware structural parameters, ensuring the safe and stable operation of the power system.
Smart Images

Figure CN118917144A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a method for optimizing structural parameters of hardware fittings, and in particular to a method for optimizing structural parameters of hardware fittings of a valve hall of a high-altitude DC converter station. Background Art
[0002] There are a large number of equipment in the valve hall, and the interconnections and arrangements are very complicated. The equipment is subjected to AC and DC superposition voltages at the same time. The electromagnetic field distribution in the valve hall is complex and difficult to explore. The structural parameters of the hardware will affect the stable operation of the power system, because at different altitudes, improper hardware structural parameters will cause corona on the surface of the hardware. The occurrence of corona will not only cause noise pollution, but also consume electric energy, cause equipment heating, and hinder wireless communications; even in locations with harsh operating conditions, long-term corona on the surface of the hardware will cause corrosion of the hardware surface, or discharge, which will weaken the external insulation safety of the equipment and thus affect the safe and stable operation of the power system.
[0003] In the prior art, the optimization of the hardware structure in the valve hall is mainly carried out through empirical measurement. This method cannot accurately determine the structural parameters of the hardware, especially under high altitude conditions, which leads to reduced stability of the power system. Moreover, the process in the prior art is complicated and inefficient.
[0004] Therefore, in order to solve the above technical problems, it is urgent to propose a new technical means. Summary of the invention
[0005] In view of this, the purpose of the present invention is to provide a method for optimizing the structural parameters of hardware fittings in the valve hall of a high-altitude DC converter station. A corresponding simulation model is established based on the structure of the valve hall and the structure of the hardware fittings, and then the surface electric field value of the hardware fittings is determined based on the simulation model. Then, the BP neural network is trained, and the hardware layout of the valve hall to be built is predicted based on the trained BP neural network. The influence of multiple factors is fully considered in the whole process, so as to effectively ensure the accuracy of the hardware structural parameters of the valve hall to be built, and it is more efficient and convenient to use, thereby ensuring the safe and stable operation of the power system.
[0006] The present invention provides a method for optimizing structural parameters of valve hall hardware of a high-altitude DC converter station, comprising the following steps:
[0007] S1. Establish a finite element three-dimensional simulation model for the valve hall of the DC converter station;
[0008] S2. Determine the valve hall simulation material and boundary voltage of the DC converter station;
[0009] S3. Apply voltage to the hardware in the finite element three-dimensional simulation calculation model, and obtain the electric field value and distribution state of the hardware surface;
[0010] S4. Repeat step S3 to obtain multiple sets of data, using the hardware parameters, hardware surface electric field values and distribution states of the DC converter station valve hall as sample data, and dividing the sample data into a training set and a test set;
[0011] S5. Construct a BP neural network, input the training set into the BP neural network to train the BP neural network, and then input the test set into the BP neural network for testing until the output error is less than the set value;
[0012] S6. Input the boundary voltage, altitude and hardware arrangement structure parameters of the valve hall of the DC converter station to be built into the BP neural network to obtain the hardware parameters of the valve hall of the DC converter station to be built, including hardware size and hardware arrangement structure.
[0013] Further, in step S2, the boundary voltage is the maximum operating voltage of the valve hall of the DC converter station;
[0014] The simulation materials of the DC converter station are determined based on the altitude of the valve hall, operating voltage, hardware structure, and hardware shape.
[0015] Further, in step S4, the hardware parameters include hardware shape, hardware size and hardware arrangement structure;
[0016] Among them, the hardware arrangement structure includes the minimum distance between the hardware and the valve hall wall, the hanging distance of the hardware and the distance of the hardware from the ground.
[0017] Furthermore, the electric field value on the surface of the hardware is the actual electric field value on the surface of the hardware, which is determined by:
[0018]
[0019] Where: E act is the actual electric field value on the surface of the hardware, E sim is the maximum electric field value on the fitting surface calculated by the finite element three-dimensional simulation model. i represents the current fitting, j represents the fitting in the valve hall whose distance from the current fitting i is less than the set value, S ij represents the distance between fittings i and j, S re is the set reference distance, k1, k2 and k3 represent the influence coefficients of temperature, humidity and surface contamination on the electric field on the surface of the hardware respectively.
[0020] Furthermore, before inputting into the BP neural network, it is necessary to determine E act Is it greater than or equal to the corona field strength E of the hardware? cor If so, remove the reorganized data.
[0021] The beneficial effects of the present invention are as follows: through the present invention, a corresponding simulation model is established based on the structure of the valve hall and the structure of the hardware, and then the surface electric field value of the hardware is determined based on the simulation model, and then the BP neural network is trained, and the layout of the hardware of the valve hall to be built is predicted based on the trained BP neural network. The influence of multiple factors is fully considered in the whole process, so as to effectively ensure the accuracy of the hardware structural parameters of the valve hall to be built, and it is more efficient and convenient to use, thereby ensuring the safe and stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:
[0023] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION
[0024] The present invention is further described in detail below:
[0025] The present invention provides a method for optimizing structural parameters of valve hall hardware of a high-altitude DC converter station, comprising the following steps:
[0026] S1. Establish a finite element three-dimensional simulation calculation model for the valve hall of the DC converter station; wherein, the three-dimensional simulation model is built by using existing software, such as ANSYS ELETRONICS software, CST and other existing software;
[0027] S2. Determine the valve hall simulation material and boundary voltage of the DC converter station; wherein the valve hall simulation material refers to the material of the valve hall and the material of the hardware, which need to be set in the three-dimensional simulation calculation model to determine the accurate surface electric field value;
[0028] S3. Apply voltage to the hardware in the finite element three-dimensional simulation calculation model. When applying voltage, the voltage application simulation is performed in a manner that is less than the boundary voltage, and the electric field value and distribution state of the hardware surface are obtained;
[0029] S4. Repeat step S3 to obtain multiple sets of data, using the hardware parameters, hardware surface electric field values and distribution states of the DC converter station valve hall as sample data, and dividing the sample data into a training set and a test set;
[0030] S5. Construct a BP neural network, input the training set into the BP neural network to train the BP neural network, and then input the test set into the BP neural network for testing until the output error is less than the set value;
[0031] S6. Input the boundary voltage, altitude and hardware arrangement structural parameters of the valve hall of the DC converter station to be built into the BP neural network to obtain the hardware parameters of the valve hall of the DC converter station to be built, including the hardware size and hardware arrangement structure. Through the above method, a corresponding simulation model is established based on the structure of the valve hall and the structure of the hardware, and then the surface electric field value of the hardware is determined based on the simulation model, and then the BP neural network is trained, and the hardware arrangement of the valve hall to be built is predicted based on the trained BP neural network. The influence of multiple factors is fully considered in the whole process, so that the accuracy of the hardware structure parameters of the valve hall to be built can be effectively ensured, and the efficiency is higher, the use of positive price is convenient, and the safe and stable operation of the power system can be ensured.
[0032] In this embodiment, in step S2, the boundary voltage is the maximum operating voltage of the valve hall of the DC converter station. Generally speaking, the maximum operating voltage will be higher than the rated voltage. For example, the rated voltage of the power grid is 800KV. However, in practice, there will be fluctuations, that is, in practice, there will be a situation of 820KV. When it reaches 830KV, it will affect the power grid system. Then the boundary voltage will be 830KV, and the transmission line will have a margin of 30KV.
[0033] The simulation materials of the DC converter station are determined based on the altitude of the valve hall, the operating voltage, the hardware structure, and the shape of the hardware. At different altitudes, different hardware structures and shapes, the materials of the hardware are also different to meet different operating voltages, and the determination process is determined through existing empirical methods.
[0034] In this embodiment, in step S4, the hardware parameters include hardware shape, hardware size and hardware arrangement structure;
[0035] Among them, the hardware arrangement structure includes the minimum distance between the hardware and the valve hall wall, the hanging distance of the hardware and the distance of the hardware from the ground.
[0036] In this embodiment, the electric field value on the surface of the hardware is the actual electric field value on the surface of the hardware, which is determined by:
[0037]
[0038] Where: E act is the actual electric field value on the surface of the hardware, E sim is the maximum electric field value on the fitting surface calculated by the finite element three-dimensional simulation model. i represents the current fitting, j represents the fitting in the valve hall whose distance from the current fitting i is less than the set value, S ij represents the distance between fittings i and j, S re is the set reference distance, i.e. the safe distance between two adjacent fittings in the power specification. k1, k2 and k3 represent the influence coefficients of temperature, humidity and surface contamination on the surface electric field of the fittings, respectively. εs It is the influence coefficient of the distance between the hardware. The above coefficients are tabulated in advance, and then determined by looking up the table according to the parameters corresponding to each coefficient in the actual environment. Through the above, since it is difficult to take the actual temperature, humidity and dirtiness into account during simulation, it is generally in an ideal state, resulting in low result accuracy. The above takes actual factors into consideration to ensure the accuracy of the final result.
[0039] In this embodiment, before inputting into the BP neural network, it is determined that E act Is it greater than or equal to the corona field strength E of the hardware? cor If so, the reorganized data is eliminated. Through this method, the accuracy of the final result can be ensured. The BP neural network determines the optimal structural parameters of the hardware through the genetic algorithm, including the size of the hardware, the suspension distance, the distance from the ground, and the distance from the side wall of the valve hall.
[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.
Claims
1. A method for optimizing structural parameters of valve hall hardware in a high-altitude DC converter station, characterized in that: The following steps are involved: S1. Establish a finite element three-dimensional simulation model for the valve hall of the DC converter station; S2. Determine the valve hall simulation material and boundary voltage of the DC converter station; S3. Apply voltage to the hardware in the finite element three-dimensional simulation calculation model, and obtain the electric field value and distribution state of the hardware surface; S4. Repeat step S3 to obtain multiple sets of data, using the hardware parameters, hardware surface electric field values and distribution states of the DC converter station valve hall as sample data, and dividing the sample data into a training set and a test set; S5. Construct a BP neural network, input the training set into the BP neural network to train the BP neural network, and then input the test set into the BP neural network for testing until the output error is less than the set value; S6. Input the boundary voltage, altitude and hardware arrangement structure parameters of the valve hall of the DC converter station to be built into the BP neural network to obtain the hardware parameters of the valve hall of the DC converter station to be built, including hardware size and hardware arrangement structure.
2. According to claim 1, the method for optimizing structural parameters of valve hall fittings in a high altitude DC converter station is characterized in that: In step S2, the boundary voltage is the maximum operating voltage of the valve hall of the DC converter station; The simulation materials of the DC converter station are determined based on the altitude of the valve hall, operating voltage, hardware structure, and hardware shape.
3. According to claim 1, the method for optimizing structural parameters of valve hall fittings in a high altitude DC converter station is characterized in that: In step S4, the hardware parameters include hardware shape, hardware size and hardware arrangement structure; Among them, the hardware arrangement structure includes the minimum distance between the hardware and the valve hall wall, the hanging distance of the hardware and the distance of the hardware from the ground.
4. According to claim 1, the method for optimizing structural parameters of valve hall hardware of a high altitude DC converter station is characterized in that: The electric field value on the surface of the fitting is the actual electric field value on the surface of the fitting, which is determined by: Where: E act is the actual electric field value on the surface of the hardware, E sim is the maximum electric field value on the fitting surface calculated by the finite element three-dimensional simulation model. i represents the current fitting, j represents the fitting in the valve hall whose distance from the current fitting i is less than the set value, S ij represents the distance between fittings i and j, S re is the set reference distance, k1, k2 and k3 represent the influence coefficients of temperature, humidity and surface contamination on the electric field on the surface of the hardware respectively.
5. According to claim 4, the method for optimizing structural parameters of valve hall hardware of high altitude DC converter station is characterized in that: Before entering the BP neural network, judge E act Is it greater than or equal to the corona field strength E of the hardware? cor If so, remove the reorganized data.
