Weapons Test Analysis Methods Based On Bayes Sequential Decision-making

According to the dominant characteristics of test analysis and assessment of weapon system in small sample circumstance, the Bayesian method in small sample statistical inference and fusion theory is adopted as an important and suitable approach. Some general theoretical problems in the engineering application of Bayesian approach have been studied in this dissertation, and this paper discusses emphatically several methods assessing about the variation of impact points.First, in the composite hypothesis testing, circular dispersion error which is more practicable than horizontal and vertical error is used to discuss some problems of the dispersion of impact points; and the critical area of decision in the dispersion problem is analyzed; the specific expression of Bayesian risk is worked out, which ameliorates the method that only can get the approximate solution by calculating numerical integration. And then, the best testing number, the probability of the two types of mistakes in the decision and the expression of the prior information are specified, too.Second, this paper discusses m-truncation Bayesian sequential testing plan for the dispersion of impact points of circular dispersion error when the number of testing is prescribed; m-step look forward process is introduced to this dissertation when the truncated number is big; as the value of the risk is unlimited, the loss function is modified in this thesis to make it limited.The loss function is very important in Bayesian decision theory, and in the end, the loss functions about variation and error of impact points are put forward according to the status of test analysis and assessment of weapon system, what's more, the Bayesian risks are worked out, which indicates the functions practicable.