一种改进的量子遗传模拟退火算法及其在神经网络智能故障诊断中的应用

摘要

The advantages and disadvantages of the simulated annealing algorithm, genetic algorithm and ordinary quantum genetic algorithm were analyzed. Aiming at the population diversity and convergence rapidity of the real-coded double-chain quantum genetic algorithm, it was combined with the simulated annealing algorithm, and real-coded double-chain quantum genetic simulated annealing algorithm was put forward on the basis of the simulation of cosmic evolution of celestial bodies. The initial weights and thresholds of BP neural network were improved with this new algorithm, and the improved BP neural network was used in intelligent fault diagnosis. The simulation results show that this algorithm has goad effective.%分析了模拟退火算法、遗传算法与普通量子遗传算法的优缺点,针对实数编码双链量子遗传算法的种群多样性和收敛快速性,将其与模拟退火算法相结合,在模拟天体宇宙演变的基础之上,提出实数编码双链量子遗传模拟退火算法,并用之改进BP神经网络的初始权值与阈值,并将改进后的BP神经网络运用于智能故障诊断中.仿真结果表明,该算法效果良好.