Testability Evaluation by Multi-sources Bayes Method Applied in Torpedo Weapon System with Improved Jaccard Similarity
Kangkang Dou, Mengyuan Zhan, Zhen Li, Xiankang Chen · 2021 4th International Conference on Algorithms, Computing and Artificial Intelligence · 2021
Aiming at the small sample problem in the testability evaluation of complex equipment systems, this paper takes the application of a torpedo combat system as an example. Based on the Bayes theory, a hybrid pre-information weighted fusion based on the improved Jaccard coefficient to consider the similarity of historical models is proposed. This method not only solves the problem of "small samples" of complex equipment testability data, but also avoids the risk of field data being overwhelmed by the traditional Bayes method due to the different overall characteristics of multi-source information and field data. Comparative analysis of examples shows that the method used in this paper is more accurate and reasonable in test evaluation than the classical evaluation method and traditional Bayes evaluation method, and has better engineering application value.