A New Inference Algorithm of Dynamic Uncertain Causality Graph Based on Conditional Sampling Method for Complex Cases

Dynamic Uncertain Causality Graph(DUCG) is a recently proposed model for diagnoses of complex systems. It performs well industry system such as nuclear power plants, chemical and spacecrafts. However, the variable state combination explosion in some cases still problem that may result inefficiency or even disability DUCG inference. In situation clinical diagnoses, when lot intermediate causes are unknown while downstream results known graph, appear during inference computation. Monte Carlo sampling typical algorithm to solve this problem. we facing case occurrence rate very small, e.g. $10^{-20}$, which means huge number samplings needed. This paper proposes new scheme based on conditional stochastic simulation obtains final from expectation probability loops instead counting frequency, thus overcomes As result, requires much less time than recursive presented earlier. Moreover, simple analysis convergence designed example given show advantage method. % addition, supports logic gate, cycles, parallelization, exist DUCG, also addressed paper. The reduces consumption 3 times faster old one with 2.7% error ratio practical graph Viral Hepatitis B.