研究目的
Investigating the uncertainty associated with calibration operations using a Monte Carlo procedure to compute the joint state-of-knowledge probability distribution for the parameters of a calibration function.
研究成果
The Monte Carlo procedure effectively computes the joint state-of-knowledge probability distribution for calibration function parameters, accommodating non-linear functions and non-Gaussian statistical models. The distributions obtained are conditional on the chosen models and assumptions, highlighting the method's flexibility and the subjective nature of the resulting state of knowledge.
研究不足
The procedure's effectiveness is contingent on the assumptions and choices made, including the form of the calibration function and the statistical models for data generation. The method requires numerical minimization and may not be straightforward for all calibration scenarios.
1：Experimental Design and Method Selection:
The study employs a Monte Carlo procedure aligned with Supplement 1 to the Guide to the Expression of Uncertainty in Measurement. It involves propagating the joint probability distribution of calibration quantities through a mathematical model derived from a least-squares adjustment procedure.
2：Sample Selection and Data Sources:
The example involves a particle detector calibrated with seven samples of radioactive reference materials, with activities and associated standard uncertainties provided.
3：List of Experimental Equipment and Materials:
Particle detector, radioactive reference materials with known activities.
4：Experimental Procedures and Operational Workflow:
The procedure includes drawing random samples from the joint distribution of calibration quantities, computing parameter values through a model, and repeating the process to approximate distributions.
5：Data Analysis Methods:
The Monte Carlo method is used for numerical approximation of distributions, with statistical models chosen based on data generation process characteristics.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
