New fuzzy wavelet neural networks for system identification and control

摘要: By utilizing some of the important properties wavelets like denoising, compression, multiresolution along with concepts fuzzy logic and neural network, new two wavelet networks (FWNNs) are proposed for approximating any arbitrary non-linear function, hence identifying a system. The output discrete transform (DWT) block, which receives given inputs, is fuzzified in methods: one using compression property other property. We present type neuron model, each synapse characterized by set implication rules singleton weights their consequents. It shown that noise disturbance reference signal reduced also variation somatic gain, parameter controls slope activation function leads to more accurate output. Identification results found be speed convergence fast. Next, we simulate control system maintaining at desired level identified models. Self-learning FNN controller has been designed this simulation. Simulation show adaptive robust.