基于GPS面向任务的现代不确定度评定理论及应用研究

Modern uncertainty principles and methods for task-specific uncertainty evaluation are studied based on new generation Geometrical Product Specification (GPS) and Guide to the Expression of Uncertainty in Measurement (GUM). Based on the typical tasks of Coordinate Measuring Machine (CMM), main factors that affect the uncertainty of results in the whole process of measurement are analyzed, such as measurement strategies, environments, measured objects, and so on. Mathematical model of task-specific measurement uncertainty and modular transport chain functions are carried out. Monte Carlo computer simulations and Bayesian statistical analysis are applied to study the intelligent evaluation methods of measurement uncertainty; task-specific uncertainty evaluation software and expert systems for CMM will be developed with independent intellectual property rights. The user of CMM can access accurate, complete measurement results and the compliant measurement uncertainty report when finishing the measurement through online or offline applications of this software. The reliability, operability and intelligence level of uncertainty assessment in the application will be promoted. On the basis of the research results, the standards, metrology and manufacturing can be closely linked, uncertainty evaluation, tolerance design, inspection and certification of compliance can be effectively integrated, and the common guidelines for geometric design, manufacture, testing and application will be provided.

以新一代GPS标准规范为依据，以不确定度评定指南GUM为准则，研究现代不确定度原理及面向任务的测量不确定度评定方法。结合坐标测量机常见典型任务，根据测量策略、环境条件、被测对象等情况，全面分析执行测量任务全过程中影响测量结果不确定度的主要因素，建立面向任务的测量不确定度数学模型及模块化传递链函数；应用蒙特卡罗计算机模拟和贝叶斯统计分析研究智能化测量不确定度评定方法；开发具有自主知识产权的坐标测量机面向任务的不确定度评定软件与专家系统。测量机用户通过在线或离线应用，可以在完成测量任务时获取符合规范的测量不确定度报告，得到准确完整并具有实用价值的测量结果，提高实际应用中不确定度评定的可靠性、可操作性与智能化水平。以此为基础，将标准、计量与制造紧密联系，使不确定度评定与公差设计、检验认证、合格判定有效融合，为几何产品的设计、制造、测试及应用提供共同准则。

科学合理、准确简便地进行面向任务的测量不确定度评定，是测量不确定度理论及应用研究的重要目标。本项目以新一代产品几何量技术规范（GPS）为依据，以不确定度表示指南（GUM）为准则，研究现代不确定度原理及面向任务的测量不确定度评定方法。本项目针对CMM面向任务的测量不确定度评定若干关键问题，进行系统深入研究。提出一种基于测量系统量值特性的不确定度分析法，该方法不需要进行误差溯源及误差传递，适用于各种测量模型和测量仪器，使不确定度分析简单有效，对促进测量不确定度的应用具有重要意义；重点研究基于量值特性的不确定度黑箱模型在CMM面向任务不确定度评定中的应用，有效解决CMM面向任务的不确定度建模难题；研究CMM典型任务的测量不确定度分量实验及估计方法，设计不确定度评定流程，编制不确定度评定软件，通过人机界面互动引导数据信息采集，进行测量不确定度分量估计和不确定度合成，并给出符合规范的测量结果报告，为实现CMM测量不确定度的智能化评定奠定了基础。分别研究了蒙特卡洛和贝叶斯两种现代不确定度评定方法，对科学准确评定不确定度具有重要作用。研究基于测量不确定度的产品检验合格性判定及误判风险。根据不确定度的分布函数，建立基于产品公差、检验不确定度的误判率模型，分别讨论了批量产品全数检验和抽样检验的误判率计算方法；通过在产品检验结果中融入统计受控的生产信息，合理有据地减小不确定度的影响，扩大产品合格区间；探讨了新一代GPS中符合性不确定度的量化方法，为协调产品设计与检验认证及合格判定提供科学依据。