Abstract: The conventional parameter optimization of PID controller is easy to produce surge and big overshoot，and therefore heuristics such as Genetic Algorithm（GA），Particle Swarm Optimization（PSO） are employed to enhance the capability of traditional techniques.But the convergence speed of SGA is slowly and PSO may be trapped in the local optima of the objective and leads to poor performance.In this paper，a weight coefficient is introduced into Quantum-behaved Particle Swarm Optimization（QPSO） and an improved QPSO（WQPSO） for the parameter optimization of PID controller is proposed.The comparison of WQPSO，PSO and QPSO based on benchmark function is given.Finally，three examples are given to illustrate the design procedure and exhibit the effectiveness of the proposed method via a comparison study with an existing Z-N，GA and PSO approaches.

Key words: Quantum-behaved Particle Swarm Optimization（QPSO）, weight coefficient, PID controller, parameter optimization

摘要： 传统的PID控制器参数优化方法容易产生振荡和较大的超调量，因此智能算法如遗传算法（SGA）和粒子群算法（PSO）被用于参数优化，弥补传统算法的不足，但是遗传算法在进化过程中收敛速度慢，粒子群算法存在易于早熟的缺点。在分析量子粒子群算法（QPSO）的基础上，在算法中引入了权重系数，提出使用改进的量子粒子群算法（WQPSO）优化PID控制器参数。将改进量子粒子群算法与量子粒子群算法、粒子群算法通过benchmark测试函数进行了比较。最后，通过三个传递函数实例，分别使用Z-N、GA、PSO方法和改进的量子粒子群算法进行了PID控制器参数优化设计，并对结果进行了分析。

关键词: 量子粒子群算法, 权重系数, PID控制器, 参数优化