基于知识的污水生化处理过程智能优化方法  

Knowledge-based Intelligent Optimal Control for Wastewater Biochemical Treatment Process

作　　者：乔俊飞[1,2] 韩改堂 周红标[1,2] 

机构地区：[1]北京工业大学信息学部,北京100124 [2]计算智能与智能系统北京市重点实验室,北京100124

出　　处：《自动化学报》2017年第6期1038-1046,共9页Acta Automatica Sinica

基　　金：国家自然科学基金(61533002);国家杰出青年科学基金项目(61225016)资助~~

摘　　要：针对污水处理过程控制能耗过大和水质超标严重等问题,本文提出一种基于知识的污水生化处理过程智能优化控制方法.该方法通过记忆多目标智能优化算法的动态处理信息,建立环境变量参数与最优解之间的知识模型.优化算法利用知识库中非支配解的引导,结合定向局部区域寻优以及随机全局寻优策略,提高了算法的收敛性,获取了更高质量的解.最后基于国际通用平台BSM1进行实验验证.结果表明,与其他优化算法相比,该方法能够在保证出水水质的前提下产生更少的能量消耗.In order to solve the problems of excessive energy consumption and serious water quality in wastewater treatment process, a wastewater treatment process intelligent optimization control method based on knowledge is proposed.Knowledge model of environment variable parameters and optimal solutions are built by memorizing the dynamic processing information of the multi-objective intelligent optimization algorithm. The optimization algorithm is guided by the non-dominated solution in the knowledge base, and combines the oriented local area search and the stochastic global search strategy to improve the convergence of the algorithm and obtains a higher quality solution. Finally, experiment verification is performed on the international common simulation platform BSM1. Results show that the proposed method can reduce energy consumption under the premise of ensuring the quality of the effluent.

关 键 词：污水处理过程 能耗 多目标优化 知识引导 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] X703[自动化与计算机技术—控制科学与工程]