Research on dynamic compensation method of thermocouples based on autoregressive and long short term memory network hybrid model
CUI Zhiwen, LI Wenjun, YU Sisi, JIN Minjun
College of Metrology and Measurement Engineering, China Jiliang University, Hangzhou 310018, China
Abstract: Thermocouples have dynamic error in dynamic temperature measurement due to thermal inertia. In order to compensate the dynamic error of thermocouples, a compensation algorithm based on autoregressive and long short term memory network hybrid model is proposed. The dynamic responses of thermocouples were identified by autoregressive model, and then the long short term memory network was used as nonlinear compensator to correct the dynamic error. Gaussian white noise with different intensity was used to simulate noise environments, and the analog measurement dataset of thermocouples was constructed. The algorithm was verified on the analog measurement dataset. The calculation results show that the algorithm can effectively reduce the dynamic error under different noise environments. The dynamic temperature measurement experimental platform was assembled. Taking K-type Ni–Cr/Ni–Si thermocouples as the experimental object, the experimental measurement dataset was obtained. The experimental and calculation results show that the dynamic responses of the thermocouples compensated by the algorithm were improved, with an average dynamic error of 0.0028 and the standard deviation of 0.010 2.
Keywords: dynamic temperature measurement;thermocouple;dynamic error compensation;autoregressive and long short term memory network hybrid model
2023, 49(9):63-72  收稿日期: 2021-11-09;收到修改稿日期: 2021-12-03
基金项目:
作者简介: 崔志文（1997-），男，浙江杭州市人，硕士研究生，专业方向为热工参数检测和仪器
参考文献
[1] SHESTAKOV A L. Dynamic error correction method[J]. IEEE Transactions on Instrumentation and Measurement, 1996, 45(1): 250-255
[2] WANG W L, ZHANG H, DU X J, et al. PCB-integrated thin film thermocouples for transient temperature measurement[J]. Electronics Letters, 2016, 52(13): 1140-1141
[3] TANG Y Q, FANG W Z, LIN H, et al. Thin film thermocouple fabrication and its application for real-time temperature measurement inside PEMFC[J]. International Journal of Heat and Mass Transfer, 2019, 141: 1152-1158
[4] YU X, WANG S, JIANG J F, et al. Self-filtering high-resolution dual-sapphire-fiber-based high-temperature sensor[J]. Journal of Lightwave Technology, 2019, 37(4): 1408-1414
[5] WANG B T, NIU Y X, ZHENG S W, et al. A high temperature sensor based on Sapphire fiber fabry-Perot Interferometer[J]. IEEE Photonics Technology Letters, 2020, 32(2): 89-92
[6] WANG Y H, SHEN C H, CHEN S J. A new infrared radiation compensation technique for dynamic temperature measurement[J]. Sensor Letters, 2012, 10(5-6): 1099-1103
[7] AHN T H, KANG J H, JEONG J J, et al. Measurement of local wall temperature and heat flux using the two-thermocouple method for a heat transfer tube[J]. Nuclear Engineering and Technology, 2019, 51(7): 1853-1859
[8] ARWATZ G, BAHRI C, SMITS A J, et al. Dynamic calibration and modeling of a cold wire for temperature measurement[J]. Measurement Science and Technology, 2013, 24(12): 125301-125311
[9] MERCIER B, JONDEAU E, CASTELAIN T, et al. High frequency temperature fluctuation measurements by Rayleigh scattering and constant-voltage cold-wire techniques[J]. Experiments in Fluids, 2019, 60(7): 1-14
[10] 赵晨阳, 张志杰, 陈昊泽, 等. 一种用于瞬态高温测试系统动态误差修正的数字反滤波器模块化设计方法[J]. 中国测试, 2021, 47(5): 112-117+144
[11] 张龙, 张宝国, 张继军, 等. 基于PSO-LSSVM的热电偶非线性校正方法研究[J]. 中国测试, 2021, 47(3): 110-115
[12] 薛光辉, 柴敬轩. 热电偶传感器温控系统误差研究[J]. 中国测试, 2019, 45(9): 100-104
[13] 金敏俊, 李文军, 郑永军, 等. 热电偶动态响应的带外部输入自回归模型[J]. 传感技术学报, 2019, 32(6): 844-851
[14] BROWN C, KEE R J, IRWIN G W, et al. Identification applied to temperature measurement in variable velocity flows[J]. IFAC Proceedings Volumes, 2009, 42(10): 1720-1725
[15] HUNG P C, MCLOONE S, IRWIN G, et al. A total least squares approach to sensor characterization[J]. IFAC Proceedings Volumes, 2003, 36(16): 321-326
[16] JACKOWSKA-STRUMILLO L. ANN based modelling and correction in dynamic temperature measurements [C]. ICAISC 2004, 7th International Conference, 2004.
[17] WU D H, HUANG S L, ZHAO W, et al. Infrared thermometer sensor dynamic error compensation using Hammerstein neural network[J]. Sensors and Actuators A:Physical, 2009, 149(1): 152-158
[18] LI W J, CUI Z W, JIN M J, et al. Dynamic temperature measurement with a dual-thermocouple sensor based on a dual-head one-dimensional convolutional neural network[J]. Measurement, 2021: 182
[19] ZHANG C B, ZHU Y Y, DONG G Z, et al. Data-driven lithium-ion battery states estimation using neural networks and particle filtering[J]. Energy Research, 2019, 43(14): 8230-8241