Search

The interior overvoltage in the extra-high voltage (EHV) power system determines the equipment insulation and safe operation of the system. At first, the paper expounds the problems brought from the application of traditional fixed high voltage shunt reactor to limit the overvoltage in our national EHV power system. Therefore this paper brings forward a proposition to solve these problems by adopting magnetically controlled shunt reactor. This paper mainly introduces a novel type of controlled shunt reactor: Magnetically controlled shunt reactor (MCR). Then its structure and principle of magnetically controlled shunt reactor is presented. The harmonic distribution characteristics, the voltage-current characteristics, control characteristics and response time of MCR are analyzed successively. Afterwards, this paper takes a 500 kV EHV transmission line from Sanyou to Jingzhou for an example, the line model and MCR model are built up. Based on Electromagnetic Transient Program (EMTP), analysis and numerical simulation has been performed to calculate frequency overvoltage and switching overvoltage in the EHV power system with magnetically controlled reactor. (7 pages)

Magnetically controlled reactor (MCR) applied in stead of traditional fixed shunt reactor in 500 kV Transmission Line from Sanyou to Jianglin is studied in this paper. MCR has the capabilities of smooth voltage and reactive power regulation, switching overvoltage suppression, secondary current limitation during single-phase fault reclosing. Structure and working principle of MCR are presented. The harmonic distribution characteristics, voltage-current characteristics, control characteristics and response time are analyzed. EMTP Simulation results show that in addition of smooth voltage and reactive power regulation MCR can also perform the same functions as those with conventional fixed reactor. (4 pages)

Presents a fuzzy and neural network integrated intelligence approach for fault diagnosis and monitoring. A generalized probability sum operator and a generalized probability product operator are proposed to implement the fuzzy neural network model for fuzzy inference. A nonlinear approximation model based on feedforward neural networks is constructed to achieve input fuzzification of fuzzy neural networks. Further, an application system is designed for condition monitoring of a marine diesel engine and system simulations for fault reasoning are discussed.