The precision detection and control method of the temperature field in the furnace chamber is the core of precision quantitative analysis. It has uncertainty characteristics of operation mode with high nonlinear, strong coupling, multiple disturbance and so on. Based on the reliability evaluation method of evidence theory, this paper presents a method to evaluate the uncertainty of the temperature field heat flux distribution in the furnace by using the information fusion technique. A method of combining heat conduction mathematical model with actual data measurement is established. Based on the Pignistic index function optimization algorithm, the Pignistic vector of the similarity measure method is established. To solve the system temperature field transient distribution of heat flux and the estimation of the thermal loss parameter, the measurement results were corrected several times by acquiring static correction factor. The experimental verification shows that the method can evaluate the uncertainty and accuracy of the measurement accuracy of the thermal analyzer.