论文部分内容阅读
针对BP神经网络在提升机制动系统故障诊断中的局限性,如收敛速度慢和可靠性差等缺点,根据提升机制动系统的故障机理和特点,提出了一种基于遗传算法的模糊神经网络故障诊断方法。结合了神经网络和模糊逻辑的优点,在利用神经网络对提升机制动系统进行故障诊断的基础上,引入模糊逻辑的概念,采用模糊隶属函数来描述这些故障的程度,并利用遗传算法对网络的权值和阈值进行修正,加快了网络收敛的速度,克服了易陷入局部极小的问题。
Aimed at the limitations of the BP neural network in the fault diagnosis of the hoist braking system, such as the slow convergence and poor reliability, a fuzzy neural network fault diagnosis based on the genetic algorithm is proposed according to the fault mechanism and characteristics of the hoist braking system. method. Combining the advantages of neural network and fuzzy logic, this paper introduces the concept of fuzzy logic based on fault diagnosis of elevator braking system by using neural network, and uses fuzzy membership function to describe the degree of these faults. By using genetic algorithm, Weights and thresholds are amended to speed up the convergence of the network speed to overcome the easy to fall into the problem of local minimum.