复杂电力设备突发故障诊断方法研究与仿真

Research and Simulation on Complex Electric Power Equipment Sudden Fault Diagnosis

研究复杂电力设备突发故障的准确诊断问题。大型电力设备中,突发性的故障越来越多,故障的状态往往呈现较强的非线性关系,造成用于预判的相关电磁特征极其的不明显。传统的故障预测方法往往根据故障状态的预判性特征进行合理的预判断。大量的非线性信号特征给正常的电网故障状态带来了较大干扰,造成诊断精度不高。为了解决上述问题,提出一种改进人工鱼群算法的电力设备故障诊断方法,通过引入神经网络的优势,对人工鱼群算法进行优化,利用人工鱼群中的各种行为对神经网络进行有效的反馈,对干扰数据进行进一步的排除,实验结果表明,改进后的人工鱼群算法提高了复杂电力设备的故障诊断准确度。

Study the accurate diagnosis of complex electric equipment sudden fault. The paper put forward a kind of power equipment fault diagnosis method based on an improved artificial fish algorithm. Through introducing the ad- vantage of neural network, the artificial fish algorithm was optimized. All kinds of behaviors of the artificial fish were used to give the neural network effective feedback and to further eliminate the interference data. The experimental re- sults show that the improved artificial fish algorithm enhances the diagnosis accuracy of complex electric power equip- ment fault.