基于案例推理的曝气过程智能控制方法  

Intelligent Control Method of Aeration Process Based on Case-based Reasoning

作　　者：吴桐 于广平[2] 袁德成 刘坚[2] 李健 孙宏存 WU Tong;YU Guangping;YUAN Decheng;LIU Jian;LI Jian;SUN Hongcun(School of Information Engineering,Shenyang University of Chemical Technology,Shenyang 110142,China;Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110142,China)

机构地区：[1]沈阳化工大学信息工程学院,辽宁沈阳110142 [2]中国科学院沈阳自动化研究所,辽宁沈阳110016

出　　处：《控制工程》2023年第11期2041-2047,共7页Control Engineering of China

基　　金：国家自然基金重大项目(61890935);国家自然基金重点资助项目(61533002);广东省科技项目(2020A0505100024)。

摘　　要：为保证出水水质,降低运行成本,污水处理过程的优化需要动态更新污水处理过程操作变量的最优设定值。因此,提出使用进化算法对溶解氧的设定值进行优化,并结合案例推理(case-based reasoning,CBR),提出一种污水处理的曝气过程智能控制方法。首先,建立入水数据与出水指标的神经网络模型,针对不同工况,使用优化算法获取操作变量的优化设定值,建立动态案例库,使用最近相邻法于案例匹配过程中,将案例重用后取得操作变量的优化设定值应用于基准仿真模型1号(benchmark simulation model No.1,BSM1)中,并得到性能评价指标。根据性能评价指标,更新操作变量的优化设定值和神经网络模型。使用BSM1对优化系统进行仿真,优化系统较原系统曝气能耗减少了18.5%,同时出水水质(effluent quality,EQ)指标得到了改善。In order to ensure the quality of effluent water and reduce operating costs,the optimization of the sewage treatment process requires dynamically updating the optimal set values of the operating variables of the sewage treatment process.Therefore,it is proposed to use an evolutionary algorithm to optimize the set value of dissolved oxygen,and combined with case-based reasoning(CBR),an intelligent control method for the aeration process of sewage treatment is proposed.Firstly,a neural network model of water inlet data and water outlet indicators is established.For different working conditions,an optimization algorithm is used to obtain the optimized setting values of operating variables,a dynamic case library is established,and the nearest neighbor method is used in the case matching process,and the cases are reused.Obtain the optimized setting value of the operating variable,apply it to the BSM1(benchmark simulation model No.1)model and obtain the performance evaluation index,and update the optimized setting value of the operating variable and the neural network model according to the performance evaluation index.BSM1 is used to simulate the optimized system.The aeration energy consumption of the optimized system is reduced by 18.5%compared with the original system,and the effluent quality index EQ(Effluent Quality)is improved.

关 键 词：曝气过程 污水处理 溶解氧 案例推理 神经网络 粒子群 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]