A Hybrid Quantum Clone Evolutionary Algorithm-based scheduling optimization in a networked learning control system

Control and network performance rely on the design of system architecture, control algorithm and scheduling of network information. In this paper, a two-layer networked learning control system (NLCS) architecture is introduced, achieving better control performance, better interference rejection and increasing the adaptability to varying environment. We establish a multi-objective optimization (MOO) based on Hybrid Quantum Clone Evolutionary Algorithm (HQCEA) with rules of expert knowledge describing for the control performance and bandwidth requirements in the two-layer NLCS to dynamically allocate bandwidth of each control loop, aiming at realizing maximization of control performance and minimization of bandwidth consumption. Such theoretical results are confirmed by the simulations of the algorithm.