传感技术学报2018，Vol.31Issue(2)：228-233,6.DOI:10.3969/j.issn.1004-1699.2018.02.013
热电偶动态响应模型的贝叶斯推断
Bayesian Inference of Thermocouple Dynamic Response Model
摘要
Abstract
The dynamic response of the thermocouple is affected by the volume change of the joint,the change of the specific heat capacity and other factors. In order to evaluate the thermocouple dynamic response capability,the ther-mocouple is excited by the step method,and the output data of the thermocouple under different temperature excita-tion is collected. The regression model of thermocouple dynamic response is established by time series analysis method,and a Bayesian inference is used to inference the parameters. The posterior distribution of the parameters is calculated using a Hamiltonian Monte Carlo sampler with the Markov chain Monte Carlo method.In order to improve sampling efficiency,the maximum-a-posteriori point of the posterior density is estimated.Then it is used as a starting point for tuning the sampler and drawing samples. Finally,the proposed regression algorithms are used to identify dynamic response of a kind of nickel-chromium/nickel-silicon thermocouple.关键词
热电偶/动态响应/贝叶斯推断/杂交蒙特卡罗法Key words
thermocouple/dynamic response/Bayesian inference/Hamiltonian Monte Carlo分类
信息技术与安全科学引用本文复制引用
李文军,孙宏健,郑永军..热电偶动态响应模型的贝叶斯推断[J].传感技术学报,2018,31(2):228-233,6.基金项目
国家科技重大专项项目(2015ZX02101) （2015ZX02101）
