Design of a Multi-variable Adaptive Stimuli for an Emotional Learning Based Controller for a MIMO Process

摘要

Most large scale systems manifest significant cross-coupling effects and operate over a wide region. Besides, mathematical models of such large scale systems often exhibit substantial deviation from the real-time model, which renders the static stimuli based emotional learning-based controller unreliable for practical purposes. In this correspondence, an adaptation based multi-variable stimuli design is pro-pounded which manifests the model uncertainties and the coupling effects associated with large scale systems. The proposed strategy is validated on a Quadruple Tank System in the non-minimum phase operating mode. The effectiveness of the proposed approach is established by comparing its performance with that of a static stimulus-based design for various levels of plant uncertainty.