论文部分内容阅读
针对量子粒子群算法(quantum particle swarm optimization,QPSO)的收敛速度和寻优精度问题,提出了一种改进的QPSO算法。采用混沌序列初始化量子的初始角位置;在算法中加入变异处理,有效地增加了种群的多样性,避免早熟收敛。函数优化测试结果表明:该文提出的算法具有良好的优化效果。同时利用该文提出的算法对经典的具有无限冲激响应(infinite impulse response,IIR)的自适应递归滤波器模型进行了辨识,辨识结果证明了这种算法的有效性。利用此算法,在结合某分散控制系统的基础上,编制出了一种通用的热工对象模型辨识算法模块,并应用于某循环流化床电厂的辨识,取得了令人满意的辨识结果。
Aiming at the convergence speed and accuracy of QPSO (Quantum Particle Swarm Optimization) algorithm, an improved QPSO algorithm is proposed. The chaotic sequence is used to initialize the initial angular position of the quantum. Adding mutation to the algorithm effectively increases the diversity of the population and avoids premature convergence. The function optimization test results show that the proposed algorithm has a good optimization effect. At the same time, an adaptive recursive filter model with infinite impulse response (IIR) is identified by using the proposed algorithm. The recognition results prove the effectiveness of this algorithm. Based on this algorithm, a general model of thermal object model identification algorithm is developed based on a decentralized control system and applied to the identification of a circulating fluidized bed power plant, and a satisfactory identification result is obtained.