State space parameters estimation using online genetic algorithms

Abstract EN

Accurate on-line estimates of critical system states and parameters are needed in a variety of engineering applications, such as condition monitoring, fault diagnosis, and process control.

In these and many other applications it is required to estimate a system variable which is not easily accessible for measurement, using only measured system inputs and outputs.

The classical identification methods, such as least-square method, are calculus-based search method.

They have many drawbacks such as requiring a good initial guess of the parameter and gradient or higher-order derivatives of the objective function are generally required also there is always a possibility to fall into a local minimum.

In this paper we develop on-line, robust, efficient, and global optimization identification for parameters estimation based on genetic algorithms.

The simulation results show that the proposed algorithm is very fast to find and adapt the estimated parameters.