【作者】 汪荣贵； 李守毅； 孙见青；

【Author】 WANG Rong-gui LI Shou-yi SUN Jian-qing School of Computer and Information, Heifei University of Technology, Hefei Anhui 230009

【机构】 合肥工业大学计算机与信息学院；

【摘要】 针对标准粒子群优化算法(SPSO)中惯性权线性变化不能适应复杂的非线性优化搜索问题,提出一种新的自适应标准粒子群优化算法(ASPSO),该算法根据粒子距最优位置的距离测度值和粒子的进化速度动态地改变惯性权系数,并使用反馈策略维持进化过程中种群多样性,避免搜索陷入局部最优。对几个典型的函数的测试结果表明,ASPSO明显优于SPSO算法。将APSO算法应用于人脸检测,实验结果表明APSO在该应用上是有效的。更多还原

【Abstract】 A new adaptive standard particle swarm optimization algorithm (ASPSO) is presented to solve the problem that the linearly decreasing weight of the standard particle swarm algorithm, which can not adapt to the complex and nonlinear optimization process. In this algorithm, the inertia weights are changed dynamically according to the evolving speed of particle and the distance between the particle and the best position. Several functions are used in the experiment and the result proves that ASPSO outperforms SPSO. Then ASPSO is introduced into BPNN to improve BP learning algorithm. The new learning algorithm is adopted to train a neural network for face detection, and the experiment results testify its efficiency in this.更多还原