Abstract: Due to inadequate maintenance, the system can't function or even be paralyzed frequently. In this paper, the risk assessment method based on Bayesian network is used to analyze the causes of fire in coal gasification plant, and the risk assessment of syngas leakage and accident consequences is carried out. Combined with risk assessment, the maintenance frequency of fire fighting system is studied, and an evaluation algorithm determining the maintenance cycle of firefighting system suitable for the inherent risks of coal gasification plant is proposed, which can realize accident prevention and reduce disaster losses to a high degree.

Key words: coal gasification, Bayesian network, risk assessment, fire fighting system, maintenance frequency