摘要
针对目前应用在PID参数自整定中的算法实现复杂、控制效果差、计算复杂度高等问题,提出了一种基于Solis&Wets算法的PID参数自整定方法。但考虑到初值点的选取会影响算法寻优的效果,将改进粒子群优化融合进Solis&Wets算法中。标准粒子群算法加入以下策略进行改进:引入平均极值将种群分类,针对不同种群采用异步进化策略,增强种群间的协作;针对进化过程中同一粒子的不同维度所出现的维差异问题,通过引入距离因子,实现粒子按维动态改变惯性权重的策略。将改进算法用于PID参数自整定中,并与其他几种算法作比较,结果表明,所提算法不仅可获得更好的控制效果,计算复杂度也明显降低。
To solve current algorithms＇ problems when applying to the self-tuning of PID parameters, such as complex processes, bad controlling effects, high computational complexity, this paper proposed a method of self-tuning of PID parameters based on Solis＆Wets algorithm. However, considering the selection of the initial point would affect the effect of optimization, a modified particle swarm optimization was mixed with Soils＆Wets algorithm. This modified algorithm involved following strategies： introducing average extreme value to classify the population, adopting asynchronous evolutionary strategies for different populations while enhancing collaboration between populations, introducing the concept of distance factor to dynamically change the inertia weight by dimensions considering that the different dimensions of the same particle had differences during searching the best location. It applied the modified Solis＆Wets algorithm to the self-tuning of PID parameters while comparing with other algorithms. The result shows that the proposed algorithm not only has a better performance for control, but has a lower computational complexity.
出处
《计算机应用研究》
CSCD
北大核心
2015年第11期3349-3351,3355,共4页
Application Research of Computers
基金
上海市自然科学基金资助项目(12ZR1420700)
