国自电气-脉冲分压器专业研发厂商
Language: ∷  English

Technical information

The role of voltage balancing ring structure optimization in high-voltage DC voltage divider

作者: GOZ Electric 时间:2024-12-09 09:01:35 阅读:17


Abstract: The electric field distribution of the external insulation of the high-voltage DC divider is extremely uneven. The position and size parameters of the grading ring are the factors that determine the magnitude of the surface field strength of the grading ring. If the design is improper, the maximum surface field strength will exceed the corona starting field strength. In view of the problem that the optimal parameters of the grading ring are difficult to solve, and the traditional genetic algorithm has the phenomenon of local convergence, the small population real-coded genetic algorithm (SPRGA) and BF (back propagation) neural network are combined to optimize the design of the grading ring of the high-voltage DC divider. The calculation results show that the structural parameters of the grading ring optimized by combining SPRGA and BP neural network can reduce the surface field strength and surface field strength of the grading ring, reduce the probability of corona discharge, and compared with the traditional genetic algorithm, it can further improve the distribution of the potential and electric field of the high-voltage DC divider, which is conducive to improving the reliability and practicality of the design of the grading ring of the DC divider.


The high-voltage DC transmission system plays an important role in my country's power network. The reliability of the DC divider is crucial to the security and stability of the power network. In recent years, faults caused by flashover of DC voltage dividers have occurred frequently, causing huge losses to DC power transmission systems. Due to the high electric field strength at the end face of the DC voltage divider, corona discharge is very likely to occur during operation. Corona loss, local overheating, and increased leakage current may also occur, resulting in reduced voltage divider accuracy. Corona loss is an important economic and technical indicator of high-voltage DC transmission lines and DC resistor dividers. The distribution of voltage divider potential and electric field is closely related to the structure of the voltage grading ring. Therefore, optimizing the structural parameters of the voltage grading ring can reduce the maximum electric field strength on the surface and reduce the occurrence of corona discharge, which is conducive to improving the utilization efficiency of high-voltage DC voltage dividers and ensuring the safe and stable operation of the power grid.


With the advancement of science and technology, the use of finite element numerical calculation software for electromagnetic field research has become a common method in this field. By modeling and simulating the optimized design model, parameter evaluation can be performed more accurately and economically. The relevant literature shows the simulation image of the electric field intensity distribution of the high-voltage divider when considering the combination of multiple grading rings, and after comprehensive comparison, the optimal installation position of the grading ring is obtained; the relevant literature obtains the optimal solution of the grading ring structural parameters through neural network fitting, and this result is calculated with the insulator electric field distribution as the structural optimization target; the relevant literature shows the results of optimizing the electric field distribution using the "analysis-evaluation-correction" cycle process with the maximum field strength borne by the core rod and silicone rubber of the rod-shaped suspension composite insulator as the objective function.


Installing a grading ring is currently the best way to optimize the high-voltage divider, which can improve the potential and electric field distribution of the divider more economically and effectively. However, if the grading ring is not designed reasonably, a high field strength area will be generated on the surface of the equipment, causing safety risks to the equipment. When the electric field strength on the surface of the grading ring is greater than 2.2kV/mm, it will cause corona discharge of the grading ring itself. At present, there is no unified standard and method for the size and installation position of the grading ring in China, and the research on the optimization design of the parameters of the grading ring of ±500kV DC divider is still blank.


The current research targets the grading ring of the insulator, but there is a lack of research on the optimization design of the structure of the grading ring of the high-voltage DC divider. In addition, the optimization methods used in the research are exhaustive method and neural network optimization algorithm, which have the disadvantages of large workload, long calculation time, low optimization accuracy, etc., and it is easy to miss the optimal result. An optimization method combining genetic algorithm and neural network is proposed in the relevant literature, but the conventional genetic algorithm has the phenomenon of local convergence, which leads to the rapid reproduction of several individuals with high fitness in the population, and the convergence range is narrow, making it difficult to find the optimal solution.


To solve the above problems, this paper uses a small population real-coded genetic algorithm (SPRGA) to simulate and optimize the parameters of the grading ring of ±500kV high-voltage DC divider, and obtains a relatively optimal structure. The entire optimization process can be widely used in the design of optimized structural parameters of DC transmission projects.


联系我们

QQ: WhatsAPP&Twitter&Facebook:+1 7134804748

手机:+86 13349886706 Same as WeChat

电话: +86 027 81739173

邮箱: gozchina@163.com

地址:China Wuhan East Lake hi tech Development Zone (Optics Valley of China)