基于改进灰狼算法-BP神经网络的智能巡检机器人电磁兼容故障诊断

智能巡检机器人巡检电力线路时可能受到电磁干扰而影响工作甚至发生故障,为有效地完成智能巡检机器人电磁兼容故障的诊断,本文提出一种基于改进灰狼算法优化BP神经网络(TGWO-BP)的故障诊断模型。由于智能巡检机器人电磁兼容故障征兆与故障原因之间具有复杂的非线性关系,采用一般BP神经网络诊断模型存在着收敛速度较慢,易陷入局部最优,诊断准确率偏低的缺陷。针对以上问题,文中利用改进灰狼算法优化BP神经网络的权值与阈值,将优化后的BP神经网络应用于智能巡检机器人电磁兼容故障诊断。仿真结果表明,相比于GWO-BP神经网络和一般BP神经网络,TGWO-BP神经网络诊断模型收敛速度加快,网络泛化能力增强,故障诊断准确率提高。

Electromagnetic Compatibility Fault Diagnosis of Power Inspection Robot Based on TGWO-BP Neural Network

When Intelligent patrol robot inspects tour in the transmission lines, it may be affected by electromagnetic interference and fall in faults. In order to effectively complete the diagnosis of electromagnetic compatibility faults in Intelligent patrol robot, This paper proposes a fault diagnosis model based on transformed grey wolf optimizer algorithm to optimize BP neural network (TGWO-BP). Because of the complex nonlinear relationship between the symptoms of electromagnetic compatibility faults and the causes of faults, the normal BP neural network diagnosis model has a slow convergence rate, easy to fall into local extremum, and low diagnostic accuracy. In this paper, transformed grey wolf optimizer algorithm is used to optimize the weight and threshold of BP neural network, and the optimized BP neural network is applied to the electromagnetic compatibility fault diagnosis of inspection Intelligent patrol robot. The simulation results show that, compared to GWO-BP neural network and general BP neural network, the BP neural network diagnosis model optimized by transformed grey wolf optimizer algorithm has a faster convergence speed, enhanced network generalization ability and improved fault diagnosis accuracy.