基于模糊小波神经网络的提升机恒减速制动系统的研究
Research of Constant Deceleration Braking System of Hoist Based on Fuzzy Wavelet Neural Network
-
摘要: 针对提升机恒减速制动系统采用常规PID控制方式、模糊控制方式存在控制效果差的问题,提出了一种基于模糊小波神经网络的提升机恒减速制动系统的设计方案。该系统采用小波基函数作为模糊隶属函数,利用神经网络的自学习能力和小波基良好的局部特性来增强模糊控制的自适应能力,并采用遗传算法对小波基函数的平移、伸缩因子以及控制器的连接权值进行训练,使网络参数达到全局最优。Matlab仿真结果表明,该系统具有良好的动态特性和较高的控制精度。Abstract: In view of the problem of bad control effect of constant deceleration braking system of hoist with common PID control mode and fuzzy control mode, the paper proposed a design scheme of constant deceleration braking system of hoist based on fuzzy wavelet neural network. The system uses wavelet basis function as fuzzy membership function, uses self-learning ability of neural network and good local characteristics of wavelet basis to enhance self-adaptive ability of fuzzy control, and uses genetic algorithm to train displacement factor and dilation factor of wavelet basis function as well as connection weights of controller, thus achievs global optimization of the network parameters. The simulation result with Matlab showed that the system has good dynamic performance and higher control precision.