基于三相同时刻采样值算法的HVDC换相失败预测改进

Improvement of HVDC CFPREV based on a three phase simultaneity sampling values algorithm

换相失败问题是制约基于电网换相换流器的高压直流(LCC-HVDC)输电系统安全运行的重要约束。从事前预防的角度出发,目前实际工程中主要采用了换相失败预测控制(CFPREV)技术。根据现场录波,分析了现有预测控制存在故障检测速度不足的问题,认为提高故障检测算法灵敏度是预测控制提升的关键。在此基础上,分析对比了基于三相同时刻采样值的电压有效值算法与现有检测算法的优劣,并采用该算法对现有换相失败预测控制功能进行改进。仿真结果表明,在单相接地或两相短路接地故障中,基于三相同时刻采样值的电压有效值检测算法的故障检测速度比现有算法更快,采用改进型预测控制能够有效减少换相失败发生的几率。

Commutation failure is an important factor restricting the safe operation of a line-commutated converter-based high voltage direct current(LCC-HVDC)transmission system.From the perspective of prevention,up to now practical projects mainly adopt the commutation failure prevention control(CFPREV).According to the field fault records,the problem of insufficient fault detection speed in the existing predictive control is analyzed,and improving the sensitivity of the fault detection algorithm is the key to the improvement of predictive control.This paper compares the advantages and disadvantages of the voltage RMS algorithm based on the three phase simultaneity sampling values and the current fault detection algorithms,and improves the current CFPREV by the voltage RMS algorithm.The simulation results show that in the single phase grounding fault or two phase grounding fault,the fault detection speed of the voltage RMS detection algorithm is faster than the current algorithms,and the improved CFPREV can also effectively reduce the probability of commutation failure.