论文部分内容阅读
针对目前测试性验证试验方案样本量过大、工程上实现困难的问题,提出了基于验前信息的复杂设备的Bayes测试性验证试验方案.首先,利用Beta分布对测试性验前信息的不确定性进行描述,运用不同来源的验前信息确定验前分布超参数;然后,定义了验前分布不确定性测度和支持度作为验前信息加权因子,设计了相应的融合算法;接着,利用融合后的验前信息建立成败型装备测试性验证试验方案的Bayes决策模型;最后,通过实例分析表明,与经典验证试验方案相比,新方案减少试验样本量40%左右,又克服了传统Bayes验证试验方案的冒进.
Aiming at the problem of too large sample size and engineering difficulty in the current test verification test program, a Bayes test verification scheme based on pre-verification information is proposed.Firstly, using the Beta distribution to test the pre-test information uncertainty The pre-test distribution hyperparameters were determined using pre-test information from different sources. Then, the pre-test distribution uncertainty measure and support were defined as pre-test information weighting factors, and the corresponding fusion algorithms were designed. Then, The Bayesian decision-making model of the test-and-verify test scheme of the failed equipment is built. Finally, the case analysis shows that the new scheme reduces the sample size by about 40% compared with the classical test scheme and overcomes the traditional Bayesian verification Pilot program aggressive.