基于改进型AIC的燃气调压器ARMAX模型  

Gas pressure regulator ARMAX model based on improved AIC

作　　者：刘若诚 王亚慧 LIU Ruocheng;WANG Yahui(School of Electrical and Information Engineering,Beijing University of Civil Engineering and Architecture,Beijing 100044,China)

机构地区：[1]北京建筑大学电气与信息工程学院,北京100044

出　　处：《现代电子技术》2020年第22期57-61,共5页Modern Electronics Technique

基　　金：国家自然科学基金资助项目(51471019);北京建筑大学基础研究燃气调压器故障智能诊断技术研究(H18191)。

摘　　要：目前,燃气输配系统智能化程度较落后,燃气调压器关键参数不易获取,建立机理模型难度较大,针对这一问题,提出基于受控自回归滑动平均模型(ARMAX)的燃气调压器系统辨识模型,并且在赤池信息准则(AIC)的基础上引入拟合优度的方法对模型进行阶次辨识,最后通过燃气调压器实际运行数据对传统AIC和改进型AIC所辨识模型进行对比实验。结果表明,改进型AIC方案可以降低模型阶次,并且在一定程度上可以提高模型精度,所得模型可精准描述系统动态特性,为燃气调压器的控制研究提供了一定的理论基础。The intelligence of gas transmission and distribution system is backward,the key parameters of gas pressure regulators are not easy to obtain,and the establishment of mechanism model is difficult.On this basis,an identification model of gas pressure regulator system based on controlled auto regressive moving average with eXogenous inputs(ARMAX)is established,and the method of goodness of fit is introduced based on Akaike information criterion(AIC)to identify the order of the model.The identification models of traditional AIC and improved AIC are compared by means of the actual operation data of gas regulator.The results show that the improved AIC can reduce the model′s order and improve the accuracy of the model to some extent.The obtained model can accurately describe the dynamic characteristics of the system and provide a theoretical basis for the control research of gas pressure regulator.

关 键 词：燃气调压器 ARMAX 改进型AIC 系统辨识 阶次辨识 拟合优度 

分 类 号：TN131-34[电子电信—物理电子学] TP391.9[自动化与计算机技术—计算机应用技术]