Study On The End Breakage Rate Optimization For Warp Based On Allele Quantum Evolutionary Algorithm

Quantum computation is a novel interdisciplinary field which combines with information science and quantum mechanics. Quantum computation is represented by quantum algorithm has high parallelism, storage capacity of exponential order and the acceleration to classic heuristic algorithm, therefore, quantum algorithm has superiority and vitality. The fusion between quantum computation and intelligent computation changed the traditional optimization method of intelligent computation, improved the search ability and convergence rate by introducing quantum computation mechanism to intelligent computation. In recent years, the study on quantum evolutionary algorithm has been one of the hotspot issues in the field of intelligent optimization.This paper analyzed the disadvantages and limitations of traditional quantum evolutionary algorithm by carding and reviewing the research on domestic and overseas intelligent optimization and intelligent modeling. Then, a real-coded quantum evolutionary algorithm based on allele is presented in which improved the coding method and updating strategy of quantum evolutionary algorithm is introduced. And the proposed quantum evolutionary algorithm is applied to the optimization of rules and consequent parameter initial values of interval type-2 fuzzy neural network. Finally, it is applied to optimize end breakage rate of warp. The main contents are presented as follows:(1) The basic principle of quantum computation and the coding method, updating strategy,the flow of quantum evolutionary algorithm are briefly introduced. The optimization results in Knapsack problem and continuous optimization problem respectively is compared with the results of genetic algorithm and ant colony algorithm, then the advantages and limitations of quantum evolutionary algorithm is analyzed.(2) For improved low precision, slow convergence speed and the limit of coding length of quantum evolutionary algorithm in continuous optimization problem, a real-coded quantum evolutionary algorithm based on allele is presented. In the proposed algorithm, the variables as alleles is coding to probability superposition, and the hybrid updating strategy based on the relative superior is introduced. The validity is verified by comparing the proposed algorithm with quantum evolutionary algorithm and double-chains quantum genetic algorithm.(3) For improving the problem for redundancy rule and the difficulty to determine the initial values of consequent parameter, an interval type-2 fuzzy neural network based on hybrid-coded quantum evolutionary algorithm is proposed. The proposed fuzzy neural network combined the hybrid-coded quantum evolutionary algorithm with the self-organizing interval type-2 fuzzy neural network which is based on Mamdani fuzzy model, eliminate the redundancy rules by combinatorial optimization, and gain superior initial values of consequent parameter by the real optimization part, Simultaneously ensure the model precision and avoid the redundancy rule problem by hybrid performance index. the effectiveness of interval type-2fuzzy neural network based on hybrid-coded quantum evolutionary algorithm is verified by comparing the proposed method with interval type-2 fuzzy neural network and adaptive neuro-fuzzy inference system.(4) Since there are strongly nonlinear and uncertainty in slashing process, the end breakage rate is affected directly which is the key index of warp weavability, then it will affect the high grade product rate and economic benefit. In this paper, a hybrid-coded quantum evolutionary algorithm applies to the prediction and optimization of end breakage rate in which employ the real-coded quantum evolutionary algorithm based on allele and the interval type-2 fuzzy neural network.The predictive model of end breakage rate is built by the interval type-2 fuzzy neural network based on hybrid-coded quantum evolutionary algorithm which size add-on and moisture regain are input variables and the end breakage rate of warp is output variable, the optimization problem is solved by the real-coded quantum evolutionary algorithm based on allele which its fitness function optimizes end breakage rate, so that the improvement of product quality.