Uncertainty dynamic evaluation method for ambient temperature measurement based on MRE and MCMC

To address the limitations of the classical GUM method in assessing measurement uncertainty for ambient temperature measurement, this study introduces the Minimum Risk Estimation and Markov Chain Monte Carlo methods, based on Bayesian assessment principles. These methods are integrated with techniques for updating prior distributions, leading to the proposal of a novel uncertainty dynamic assessment method. To validate the method's feasibility and stability, a dynamic evaluation of measurement uncertainty is conducted using computer simulation data and ambient temperature measurements in a cold storage environment. The results indicate that the enhanced Bayesian uncertainty dynamic evaluation method aligns with the actual conditions of the cold storage ambient temperature measurement system, demonstrating both feasibility and stability.