为减少热惯性温度传感器的动态测温误差,提出了基于传热模型和反演算法的动态测温方法,其实质是将求解被测对象的真实温度转化为一个优化问题。该方法首先建立温度传感器非线性非定常瞬态传热正模型和有限元求解方程,并进行实验验证。在此基础上,推导了瞬态传热一阶敏度计算公式,建立了温度传感器的非稳态反演传热模型,结合快速反演算法,反求作为边界条件的被测对象真实温度。将所提出的动态测温方法应用于具有内外两层套管、传热过程复杂的复合温度传感器。实验和仿真结果均表明,正向模型的仿真值与测量值相比最大误差小于5℃,反演求解的被测对象真实温度误差不超过1%,热响应时间由548 s缩短到106 s。应用实例分析结果表明,所提出的动态测温方法具有较高的准确性和快速性。
Based on the heat transfer model and inverse algorithm,a dynamic temperature measurement method was proposed for reducing the dynamic measurement errors of temperature sensors with large thermal lag.This method is used to obtain the actual temperature of measured medium through solving optimization problems.Non-linear unsteady transient forward heat transfer models and finite element solution equations of the temperature sensor are presented.On the basis of experimental validation,the first-order sensitivity formula of transient heat transfer is derived,and the transient inverse heat transfer models are established.The actual temperature can be achieved as the boundary condition with the fast inverse algorithm.The proposed method was applied to a compound temperature sensor with internal and external two layers of sheath and complex heat transfer process.Experiment and simulation results all show that the maximum error between measurement value and forward simulation value is less than 5 ℃,the relative error between real measured temperature and inverse calculation result is within 1%,the thermal response time is reduced from 548 s to 106 s.Application results prove the accuracy and rapidity of the proposed method.