针对传统贝叶斯网络对复杂对象建模存在的问题,分析了面向对象的概率关系模型。建立基于面向对象概率关系模型知识库的贝叶斯网络生成算法,建立战场威胁级别评估模型,并实现对威胁级别的评估。仿真结果证明了该方法的有效性,并且具有比贝叶斯网络模型更强的建模能力。 Aiming at the problems existing in the modeling of complex objects by traditional Bayesian networks, the object-oriented probabilistic relational model is analyzed. The Bayesian network generation algorithm based on object-oriented probabilistic relational model knowledge base is established, the battlefield threat level assessment model is established, and the threat level is evaluated. The simulation results prove the effectiveness of the proposed method and have better modeling ability than Bayesian network model.