【作者】 孙剑； 蒙西； 乔俊飞；

【Author】 Sun Jian;MENG Xi;QIAO Jun-fei;Faculty of Information Technology, Beijing University of Technology;Beijing Key Laboratory of Computational Intelligence and Intelligent System;

【机构】 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；

【摘要】 在城市固体废弃物焚烧(Municipal solid waste incineration,MSWI)过程中,主蒸汽流量对锅炉机组运行状况、性能监测、过程控制起着至关重要的作用。针对主蒸汽流量难以实时检测的问题,文中提出了一种基于平均影响值(Mean impact value,MIV)和径向基函数(Radial basis function, RBF)神经网络的主蒸汽流量软测量方法。首先,根据实际生产过程中锅炉的工艺特点,基于机理及经验知识选取与主蒸汽流量相关的变量;然后,采用MIV算法进行特征选择确定软测量模型的输入变量;最后,通过RBF神经网络建立主蒸汽流量软测量智能模型。实验结果表明,文中建立的MIV-RBF软测量模型能够实现对主蒸汽流量的实时精准检测,验证了所提出方法的有效性和可行性。更多还原

【Abstract】 In the process of municipal solid waste incineration(MSWI), the main steam flow plays a crucial role in the operation status, performance monitoring and process control of the boiler unit. Due to the fact that it is difficult to measure the main steam flow, a kind of radial basis function neural network(RBF) soft sensor model based on mean impact value(MIV) is designed for a grate incinerator. According to the technological characteristics of the boiler in the actual production process, the variables related to the main steam flow rate are selected based on the mechanism and empirical knowledge; the input variables of the soft sensor model were selected by MIV algorithm; the main steam flow intelligent modeling was established by RBF neural network. The experimental results show that the MIV-RBF model can realize the real-time accurate detection of the main steam flow, which verifies the effectiveness and feasibility of the proposed method.更多还原