论文部分内容阅读
针对南方电网某500 kV变电站因某弹簧机构输出功不够,导致断路器出现偶尔合闸不到位的情况与BP算法存在局部极小点,收敛速度慢等缺点,在模式识别的基础上,以断路器机械特性数据为特征量,断路器合闸到位与合闸不到位为研究目标,采用改进BP神经网络算法,实现了断路器合闸到位平均速度的有效预测;算例计算结果的准确率高达99.68%,能有效预测断路器合闸到位的最低平均速度。
Aiming at a 500 kV substation of China Southern Power Grid due to a spring mechanism output is not enough, leading to occasional closing of the circuit breaker is not in place with the BP algorithm there is a local minimum point, slow convergence and other shortcomings, on the basis of pattern recognition, Mechanical characteristics of the characteristic data for the circuit breaker closing in place and the closing is not in place for the purpose of the study, the use of improved BP neural network algorithm to achieve an effective prediction of circuit breaker closing average speed; calculation results of the accuracy of up to 99.68%, can effectively predict the minimum average speed in circuit breaker closing in place.