改进的BP网络及其在电路故障诊断中的应用

Application of an Improved BP Neural Network in Circuit Fault Diagnosis

BP神经网络在模拟电路故障诊断中的应用是一个十分热门的研究课题,但BP网络自身的学习收敛速度慢、容易陷于局部极小等缺陷则限制了其应用。提出了一种改进型的BP网络诊断模拟电路故障的方法,详细介绍了自适应学习率动量法,对模拟电路故障进行诊断和分析研究。最后,给出了仿真实例,验证了自适应学习率动量法加快了样本学习收敛速度,具有更好的实时性和诊断效率。

Study on BP neural network applied in diagnosing the analog eircuit＇s fault is one of important problems. However, the problems about slow convergence rate and easily to fall into local minimums are considered in its application. An improved BP neural network is presented in this paper and applied in analog circuit fault diagnosis. It is the joints of additional momentum algorithm and self- adaptive learning rate algorithm. Finally, an illustrative example is given to validate that the convergence tempo is quickened in the improved algorithm. At the same time, the improved one is more feasible and efficient in contrast with the traditional one.