Nonlinear System Parameterization and Control using Reduced Adaptive Kernel Algorithm

To develop a system for specific purpose, it needs to estimate its parameters (parameterization). It can be used in different fields like engineering, medicine, industry etc. In this work, authors used adaptive algorithm to model a system that is applicable in industry for control. The model is non-linear and works on kernel-based estimation with Least-mean square (LMS) algorithm. The kernels are verified with Polynomial and Gaussian. As the system is nonlinear, polynomial kernel-based algorithm fails to prove its efficacy, though it is of low complexity approach. Gaussian kernel-based application for nonlinear system control performs better as compared to polynomial kernel. Further, the complexity is reduced and used with Gaussian kernel in LMS algorithm for better performance. The result proves its performance in form of MSE, MAE, RMSE for identification and control that is very useful in industrial application. With the use of reduced Gaussian kernel application, the MSE, MAE and RMSE are found to be -54.622 dB, 0.0362,0.235 respectively, in 0.01062 sec that shows the time consumption is very less compared to other approaches.