Quantum Computation With Applications To Intelligent Optimization And Control

Quantum computation is a novel inter-discipline that includes quantum mechanics and information science. As the representative of quantum computation, the quantum algorithm has great implication of superiority and strong vitality, and become the cutting-edge research field of many international scholars because of a high degree of parallelism, the index level of storage capacity, and the role of speeding up the classic heuristic algorithm. The integration of quantum computing and intelligent computing may improve optimization ability and convergence rate of the traditional algorithms by embedding the quantum computing into the traditional intelligent computing. Hence, the research of quantum algorithm has important theoretical and practical significance. This paper mainly studied the improvement strategy of Grover algorithm, the quantum-inspired optimization algorithm, the quantum neural networks model and algorithm, and application of quantum genetic algorithm in fuzzy controller parameter optimization. The specific content of the paper can be summarized as follows.Firstly, the existing problem in basic Grover algorithm is analyzed. By changing rotation phase of qubits and weighting the marked states, five improved Grover algorithms are proposed. These methods have improved the success probability of initial Grover's algorithm to varying degrees.Secondly, the encoding way of double chains chromosomes is constructed by directly regarding two probability amplitudes of qubit as genes in chromosome. On the basis, such algorithms are proposed as the quantum genetic algorithm with double chains, chaos quantum immune algorithm, quantum ant colony optimization algorithm, and quantum particle swarm optimization algorithm. By directly regarding the Bloch coordinates of qubit as genes in chromosome, a quantum-inspired evolution algorithm is proposed. With the application of double or triple chains encoding way, these algorithms evidently improved the optimization efficiency.Thirdly, through integration of the quantum computing and the neural networks theory, four quantum neural networks models and algorithms are presented. In these models, the processes of training are achieved by adjusting the related parameters of the quantum rotation gates or the controlled-non gates. The simulation results show that their approximation ability and convergence speed are evidently superior to the common neural networks.Fourthly, aiming at parameter optimization of fuzzy controller with the analytical fuzzy control rules, an optimization method is proposed based on quantum genetic algorithm. To solve the control of inverted pendulum, a normalized fuzzy neural networks controller is constructed, and its parameter is optimized by quantum genetic algorithm, which may achieve the automatic design of controller. This controller is successfully applied to inverted pendulum control system, and the simulation results show the optimization ability of quantum genetic algorithm is evidently superior to common genetic algorithm.Finally, the research works of this paper are summarized and the further exploration and research are discussed in quantum intelligent computing.