本发明公开了一种基于贝叶斯网络的热力系统可靠性评估方法，包括以下步骤：S1、建立热力系统贝叶斯网络逻辑关系模型；S2、计及热惯性延迟的热力系统贝叶斯网络时序模拟过程；进行时序模拟时，假设热源供应充足并且处于正常状态，并且各元件设备均处于正常状态，则在任意时刻均能满足热负荷需求；当元件故障时，需要计及热网传输延时及其终端热惰性，以此反映热负荷点的状态；S3、对基于贝叶斯网络时序模拟的热力系统进行可靠性评估。本发明采用上述基于贝叶斯网络的热力系统可靠性评估方法，在考虑热惯性延迟的前提下采用分时段的方法，提出的推理算法不但能进行正向推理计算系统可靠性指标，且能反向推理能识别系统可靠性的薄弱环节。

摘要（英）：The invention discloses a thermodynamic system reliability evaluation method based on a Bayesian network. The method comprises the following steps: S1, establishing a thermodynamic system Bayesian network logic relation model; s2, a thermodynamic system Bayesian network time sequence simulation process considering thermal inertia delay; when time sequence simulation is carried out, it is assumed that heat source supply is sufficient and is in a normal state, and all element devices are in a normal state, the heat load requirement can be met at any moment; when an element breaks down, heat supply network transmission delay and terminal thermal inertia need to be considered, so that the state of a thermal load point is reflected; and S3, carrying out reliability evaluation on the thermodynamic system based on Bayesian network time sequence simulation. According to the thermodynamic system reliability evaluation method based on the Bayesian network, a time-phased method is adopted on the premise of considering the thermal inertia delay, and the proposed reasoning algorithm not only can perform forward reasoning to calculate the reliability index of the system, but also can perform backward reasoning to identify the weak link of the reliability of the system.