由于一般优化问题的先验知识很难获取,因此在Bayesian网络学习中结合与利用先验知识一直是一个很难突破的问题。针对Bayesian优化算法(BOA)的特点,对一般优化问题如何发现和利用先验知识进行了分析讨论,把BOA中前一代种群所提供的信息作为先验知识结合到当前代Bayesian网络的学习中,提高了所学习网络的可靠性,从而提高算法的性能。仿真结果表明所提算法比传统BOA具有更强的全局寻优能力。 Because prior knowledge of general optimization problems is difficult to obtain, combining and using prior knowledge in Bayesian network learning has been a difficult problem to be solved. According to the characteristics of Bayesian optimization algorithm (BOA), this paper analyzes and discusses how to find and use the prior knowledge in general optimization problems. The information provided by the previous generation of BOA is used as the prior knowledge in the learning of current generation Bayesian networks. Improve the reliability of the learning network, thereby improving the performance of the algorithm. Simulation results show that the proposed algorithm has better global optimization ability than traditional BOA.