高级检索

    陈如清, 俞金寿. 基于带扰动项粒子群算法的软测量建模[J]. 华东理工大学学报(自然科学版), 2007, (3): 414-418.
    引用本文: 陈如清, 俞金寿. 基于带扰动项粒子群算法的软测量建模[J]. 华东理工大学学报(自然科学版), 2007, (3): 414-418.
    CHEN Ru-qing, YU Jin-shou. Soft Sensor Modeling Based on Particle Swarm Algorithm with Disturbance[J]. Journal of East China University of Science and Technology, 2007, (3): 414-418.
    Citation: CHEN Ru-qing, YU Jin-shou. Soft Sensor Modeling Based on Particle Swarm Algorithm with Disturbance[J]. Journal of East China University of Science and Technology, 2007, (3): 414-418.

    基于带扰动项粒子群算法的软测量建模

    Soft Sensor Modeling Based on Particle Swarm Algorithm with Disturbance

    • 摘要: 针对粒子群算法用于高维数、多局部极值点的复杂函数寻优时易陷入局部最优解现象,提出一种改进的带扰动项粒子群算法并进行收敛性分析。算法中引入进化速度因子,当粒子进化速度低于一定值时在粒子速度更新方程中添加扰动项使粒子逃离局部最优区而继续搜索。对几个复杂函数的寻优测试表明:改进算法的收敛速度、收敛精度和全局搜索性能均有显著提高。将本方法用于建立丙烯腈收率神经网络软测量建模,研究结果表明模型精度较高、泛化性能好,满足现场测量要求。

       

      Abstract: Traditional particle swarm optimization(PSO) algorithms often trap into local minima easily when used for the optimization of high-dimensional complex functions with a lot of local minima.To solve this problem an improved PSO algorithm with disturbance is presented and its convergence capability is studied then.The evolution speed factor of the swarm is introduced in this new algorithm.A disturbance will be added to the velocity update equation when the value of the factor is less than a certain value to he...

       

    /

    返回文章
    返回