基于区间二型模糊神经网络的出水氨氮软测量

发布时间：2019-07-05 13:09

【摘要】：针对污水处理过程出水氨氮(ammonia nitrogen,NH4-N)难以实时检测的问题,提出了一种基于区间二型模糊神经网络(interval type-2 fuzzy neural networks,IT2FNN)的软测量方法,建立了出水NH4-N的软测量模型,实现了出水NH4-N的实时检测。首先,采集和预处理相关过程变量的实际运行数据,通过主元分析法筛选出与出水NH4-N相关性较强的过程变量。其次,利用IT2FNN建立所选变量与出水NH4-N的软测量模型,通过梯度下降算法对模型相关参数进行修正。最后,将基于IT2FNN的出水NH4-N软测量模型应用于实际污水处理过程。实验结果表明,提出的出水NH4-N软测量方法不仅能够实现污水处理过程出水NH4-N的实时检测,而且具有较高的检测精度。
[Abstract]:In order to solve the problem that it is difficult to detect effluent ammonia nitrogen (ammonia nitrogen,NH4-N) in wastewater treatment process, a soft sensing method based on interval two fuzzy neural network (interval type-2 fuzzy neural networks,IT2FNN) is proposed. The soft sensing model of effluent NH4-N is established, and the real time detection of effluent NH4-N is realized. Firstly, the actual operation data of related process variables are collected and pretreated, and the process variables with strong correlation with effluent NH4-N are screened out by principal component analysis (PCA). Secondly, the soft sensing model of selected variables and effluent NH4-N is established by IT2FNN, and the related parameters of the model are modified by gradient drop algorithm. Finally, the effluent NH4-N soft sensing model based on IT2FNN is applied to the actual sewage treatment process. The experimental results show that the proposed soft sensing method of effluent NH4-N can not only realize the real-time detection of effluent NH4-N in sewage treatment process, but also has high detection accuracy.
【作者单位】：北京工业大学信息学部;计算智能与智能系统北京市重点实验室;
【基金】：国家自然科学基金项目(61622301,61533002) 北京市教育委员会科研计划项目(KZ201410005002,km201410005001)~~
【分类号】：X832;TP183

【相似文献】