文献类型：期刊文献

中文题名：污水处理溶解氧的模糊神经网络控制器研究

英文题名：Fuzzy Neural Network Controller of Dissolved Oxygen in the Wastewater Treatment Process

作者：李雄军[1];罗健旭[1];黄志清[1];王文佳[1]

机构：[1]华东理工大学自动化系,上海200237

年份：2010

卷号：17

期号：S2

起止页码：64

中文期刊名：控制工程

外文期刊名：Control Engineering of China

收录：CSTPCD;;北大核心:【北大核心2008】；CSCD:【CSCD2011_2012】；

基金：国家自然科学基金资助项目(60974066)

语种：中文

中文关键词：模糊神经网络;粒子群算法;BP算法;溶解氧;BSM1

外文关键词：fuzzy neural network;;particle swarm optimization;;back-propagation algorithm;;dissolved oxygen;;benchmark simulation Model-1

摘要：针对基于活性污泥法处理工艺的污水生化处理系统对溶解氧控制的要求,提出了一种改进性模糊神经网络控制算法,对污水生化处理系统的溶解氧质量浓度进行控制。在该改进性模糊神经网络控制方法中,为了克服单独应用BP算法或粒子群(PSO)算法训练模糊神经网络控制器参数时存在的缺陷,提出了一种将BP算法和粒子群(PSO)算法二者相结合的改进算法(PSO+BP算法),以充分利用PSO算法的全局寻优能力和BP算法的局部搜索能力,从而更有效地提高训练模糊神经网络控制器参数的效率,以及提高控制器的控制效果。最后基于活性污泥法处理工艺的仿真基准模型BSM1仿真平台,对污水生化处理系统的溶解氧控制进行仿真研究,与模糊控制的效果进行对比分析,验证了该模糊神经网络控制器的有效性和可行性。
In order to meet the requirements of dissolved oxygen control in biological wastewater treatment system,an improved fuzzy neural network(FNN) control method is proposed to control dissolved oxygen.In order to overcome the shortcomings of the algorithm when applying the particle swarm optimization algorithm(PSO) or the back-propagation algorithm(BP)to train the fuzzy neural network controller parameters,a combined algorithm of both PSO and BP is proposed.The combined algorithm makes full use of the global optimization ability of PSO and local search ability of BP,and it improves the efficiency of training.The dissolved oxygen control using the adaptive FNN is simulated in biological wastewater treatment platform based on Benchmark Simulation Model-1(BSM1 ),and comparison with fuzzy control results is taken.Results show that the fuzzy neural network controller is effective and feasible.