基于证据理论的TPS诊断性能综合评定

测试程序集TPS(Test Program Set)诊断性能的评定是保证TPS质量的重要手段.针对TPS诊断性能的评定问题,研究了基于D-S(Dempster-Shafer)证据理论的综合评定方法,介绍了D-S证据理论的基本原理,提出了TPS评价验证的数学模型以及基于证据理论的TPS诊断性能综合评定的基本框架,定义了TPS诊断性能的基本概率分配函数(BPAF,Basic Probability Assignment Functions),并在此基础上提出了基于Kolmogorov-Smimov(K-S)检验的BPAF确定方法,给出了TPS诊断性能综合评定的步骤,最后对基于D-S证据理论的TPS诊断性能综合评定的有效性进行了实例分析,结果证明了该方法的有效性.

Comprehensive TPS diagnostic capability assessment methodology based on Dempster-Shafer evidence theory

The assessment of TPS(test program set) diagnostic capability is one of the key means for TPS quality assurance.According to the comprehensive assessment problem of TPS diagnostic capability,a new knowledge fusion method based on D-S(Dempster-Shafer) evidence theory is developed.The rationale of D-S theory is introduced at first.The mathematical model and framework of TPS diagnostic capability assessment is proposed.The BPAF(basic probability assignment functions) definition and decision of TPS diagnostic capability feature based on K-S test(Kolmogorov-Smimov test) is described in detail.The process of TPS diagnostic capability assessment is given.At last an example is shown to assess an avionic equipment TPS.Result shows that D-S theory can effectively consider multi-features of TPS and decrease the decision uncertainty region.