Rule-based PI controller autotuning for drive systems

Tuning a PI controller can be quite cumbersome for the non-expert as the closed-loop control system has to meet various requirements while the influence and interaction of the two degrees of freedom are not always clear. This paper addresses the design of an iterative PI controller autotuning for drive systems with the idea of imitating a human expert. In contrast to existing concepts, a new approach with multiple tuning strategies is applied which gives a compact rule base that is easy to modify. The performance of the proposed algorithm is illustrated through motion control testbench trials.