CAJ | 학술논문

为解决粒子群优化算法寻优过程中易出现种群趋同化而导致早熟收敛的问题，引入粒子群的进化速度和种群多样性适应度方差两个因素，构建了自适应的动态的惯性因子取值机制，并讨论了惯性因子的收敛性及参数的独立性，从而改进了传统粒子群优化算法的惯性因子线性取值机制。将改进的粒子群优化算法应用于某水库的优化调度中，验证了该算法能以较快的速度收敛得到全局极值，克服了易陷入局部最优的缺点，为水库优化调度问题提供了一条新途径。
위해결입자군우화산법심우과정중역출현충군추동화이도치조숙수렴적문제，인입입자군적진화속도화충군다양성괄응도방차량개인소，구건료자괄응적동태적관성인자취치궤제，병토론료관성인자적수렴성급삼수적독립성，종이개진료전통입자군우화산법적관성인자선성취치궤제。장개진적입자군우화산법응용우모수고적우화조도중，험증료해산법능이교쾌적속도수렴득도전국겁치，극복료역함입국부최우적결점，위수고우화조도문제제공료일조신도경。
In order to solve the problem of the homogenization of particle swarm prone to premature convergence resulting in the process of particle swarm optimization,this paper proposed a new mechanism to assign the value to inertia factor adaptively and dynamically with the evolution speed factor and swarm diversity fitness variance factor to improve the traditional linear method. The improved particle swarm optimization algorithm was applied to the actual reservoir optimal scheduling to verify that the algorithm had faster convergence speed to get the global extreme and had over-come the shortcomings of easily to fall into local optimum,which had provided a new way for the issue of reservoir optimal scheduling problem.