A comparative study of the wavenet PID controllers for applications in non-linear systems

In this paper, a comparative study of the wavenet PID controllers for applications in non-linear systems is presented. The wavenet PID controllers combine the neural network learning advantage with the wavelet representation for an efficient identification of non-linear dynamic systems, and these have application when we want to control plants of unknown mathematical model with highly non-linear characteristics. There exist different types of PID wavenet controllers i.e. wavenet PID, fuzzy-wavenet PID and multiresolution wavenet PID, these controllers tune online the proportional, integral and derivative gains of a classical discrete PID controller, through the identification of the plant using a radial basis neural network with different daughters wavelet activation functions. For this reason, the performance of the wavenet PID and the fuzzy-wavenet PID controlling a sub-actuated system are compared with the classical PID controller. The simulation results show that the fuzzy-wavenet PID controller has a good performance to carry out the control nonlinear system, for example the inverted pendulum.