Optimization of static and dynamic anti-windup compensator using new improved particle swarm optimization algorithm

In this paper, the optimal static and the dynamic anti-windup compensators (AWC) are designed using a new improved particle swarm optimization (PSO) algorithm. The existing optimization techniques for AWC make use of LMI (Linear Matrix Inequality) and various concepts of nonlinear control concepts that require complex mathematics. For this optimization, we have developed and utilized an improved version of PSO. In this new improved PSO algorithm, first the problem-space is explored with constant non-zero inertia factor until the fitness function saturates to a local or global minima. The solution obtained thus far is then further refined by rerunning the algorithm but with inertia factor set to zero. This slight improvement enables a better search for a solution within the already explored problem-space. The optimized static and dynamic AWC is compared with the existing LMI-based AWC for the performance index, and it is observed that the improved PSO-based AWC gave better results and optimized value of the performance index.