Study On The Evaluation Of Uncertainty Based On Bayesian Theory

Nowadays with the rapid development of science and technology and the expand of foreign trade, there are more requirements of the accuracy and reliability of measurement, so the evaluation of uncertainty is getting more and more attention. Based on the evaluation principle of uncertainty in GUM (Guide to the Uncertainty in Measurement), the paper studies on the evaluation of uncertainty based on the Bayesian theory.In allusion to the limit by sample size and insufficiency in using related data of Type A evaluation of uncertainty in GUM, the paper introduces Type A evaluation of measurement uncertainty based on Bayesian theory; in allusion to the poor maneuverability and quality insurance by the influence of the subjective judgments such as the cognition of appraisers, the paper introduces Type B evaluation of uncertainty based on two-step fuzzy comprehensive evaluation in fuzzy mathematics and Bayesian theory; in allusion to the combined standard uncertainty restriction in GUM by measurands and the differentiability and complexity of model, the paper introduces the evaluation of combined standard uncertainty based on MATLAB and the theory of Monte Carlo.Based on the above theory and methods, the author did some cases analysis about the evaluation of uncertainty.