基于知识模糊迁徙的城市污水处理膜污染决策  

Decision-making for Membrane Fouling Based on Knowledge Fuzzy Transfer in Municipal Wastewater Treatment

作　　者：何政[1,2] 赵楠 李杰[1] 陈行行 阜崴 顾剑 韩红桂 刘峥[3,4] HE Zheng;ZHAO Nan;LI Jie;CHEN Hanghang;FU Wei;GU Jian;HAN Honggui;LIU Zheng(Beijing City Drainage Refco Group Ltd,Beijing 100044,China;Beijing City Drainage Environment Development Ltd,Beijing 100044,China;Faculty of Information Technology,Beijing University of Technology,Beijing 100124,China;Beijing Key Laboratory of Computational Intelligence and Intelligent System,Beijing University of Technology,Beijing 100124,China)

机构地区：[1]北京城市排水集团有限责任公司,北京100044 [2]北京北排水环境发展有限公司,北京100044 [3]北京工业大学信息学部,北京100124 [4]北京工业大学计算智能与智能系统北京市重点实验室,北京100124

出　　处：《北京工业大学学报》2024年第3期299-306,共8页Journal of Beijing University of Technology

基　　金：国家自然科学基金资助项目(62125301);中国博士后科学基金资助项目(2022M720319)。

摘　　要：针对城市污水处理膜污染难以精准决策的问题,提出一种基于知识模糊迁徙的膜污染决策方法。首先,结合城市污水处理运行过程数据和运行经验,利用模糊规则的形式实现膜污染决策知识的表达;其次,提出一种知识重构机制(knowledge reconstruction mechanism,KRM),动态平衡源域与目标域之间的准确性和多样性,并采用知识迁徙的方法完成决策知识重构;最后,建立一种基于数据和知识驱动的区间二型模糊神经网络(data-knowledge-driven interval type-2 fuzzy neural network,DK-IT2FNN)的决策模型,利用模糊规则设计模型参数,采用迁徙梯度下降算法动态调整网络权值,提高决策精度。实验结果表明,该模型能够实现膜污染的精准决策。A decision-making method,based on a knowledge fuzzy transfer,was proposed to solve the problem of membrane fouling in wastewater treatment process.First,based on the data and experience collected from a real wastewater treatment process,the knowledge of membrane fouling decision-making was expressed in the form of fuzzy rules.Second,a knowledge reconstruction mechanism(KRM)was proposed to complete the knowledge reconstruction by balancing the matching accuracy and diversity between the source domain and the target domain with knowledge transfer method.Finally,a decision-making model based on data-knowledge-driven interval type-2 fuzzy neural network(DK-IT2FNN)was developed,the model parameters were designed by using fuzzy rules,and the transfer gradient descent algorithm was proposed to adjust the model weight.The decision-making accuracy has been improved.Results show that the proposed method can realize the decision of membrane fouling with high accuracy.

关 键 词：城市污水处理 膜污染 知识重构机制(knowledge reconstruction mechanism KRM) 模糊神经网络 模糊迁徙 梯度下降算法 

分 类 号：U461[机械工程—车辆工程] TP308[交通运输工程—载运工具运用工程]