System identification using evolutionary computation and its application to internal adaptive model control

The requirement for the high quality control of complex and/or structure-unknown plant is growing in the real-world industrial machine. Indirect Adaptive Control (IAC), which identifies model and updates the compensators automatically, is expected as one of the promising ways to meet this requirement. The conventional IAC, however, required the information of the structure of the plant, i.e. the order of its transfer function in addvance. This paper presents a new IAC scheme which utilizes Genetic Algorithm (GA) in its Identification part and embeds it into a control system, In the proposed framework, the information on the order of the plant is not required, since GA can find both parameters of the plant and a structure of the plant dynamics autonomously. A two-degree-of-freedom Internal Model Control (IMC) is adopted as the basic controller architecture, because the indirect adaptation mechanism can be achieved seamlessly. The effectiveness of the proposed framework is verified through some numerical simulations and experiments applied to a velocity control of a multi-mass system.