Addressing real-time control problems in complex environments using dynamic multi-objective evolutionary approaches

摘要

The demand for increased automation of industrial processes generates control problemsthat are dynamic, multi-objective and noisy at the same time. The primaryhypothesis underlying this research is that dynamic evolutionary methods could beused to address dynamic control problems where conicting control criteria are necessary.The aim of this research is to develop a framework for on-line optimisationof dynamic problems that is capable of a) representing problems in a quantitativeway, b) identifying optimal solutions using multi-objective evolutionary algorithms,and c) automatically selecting an optimal solution among alternatives.A literature review identi es key problems in the area of dynamic multi-objectiveoptimisation, discusses the on-line decision making aspect, analyses existing Multi-Objective Evolutionary Algorithms (MOEA) applications and identi es researchgap. Dynamic evolutionary multi-objective search and on-line a posteriori decisionmaker are integrated into an evolutionary multi-objective controller that uses aninternal process model to evaluate the tness of solutions.Using a benchmark multi-objective optimisation problem, the MOEA abilityto track the moving optima is examined with di erent parameter values, namely,length of pre-execution, frequency of change, length of prediction interval and staticmutation rate. A dynamic MOEA with restricted elitism is suggested for noisyenvironments.To address the on-line decision making aspect of the dynamic multi-objectiveoptimisation, a novel method for constructing game trees for real-valued multiobjectiveproblems is presented. A novel decision making algorithm based on gametrees is proposed along with a baseline random decision maker.The proposed evolutionary multi-objective controller is systematically analysedusing an inverted pendulum problem and its performance is compared to Proportional{Integral{Derivative (PID) and nonlinear Model Predictive Control (MPC) approaches.Finally, the proposed control approach is integrated into a multi-agent frameworkfor coordinated control of multiple entities and validated using a case study of atra c scheduling problem.