基于Markov决策过程的输变电设备最佳检修决策

发布时间：2019-07-18 21:21

【摘要】：为解决现有输变电设备检修决策很大程度依赖于决策人员经验的问题,应用Markov决策过程为检修决策提供定量、具体的决策依据。首先通过Markov过程求解设备的状态转移概率;然后运用策略迭代法对模型最优决策进行求解;最后将数据代入检修决策模型中,改变故障损失得到决策结果的变化。比较决策变化对应的设备故障损失和设备稳态概率可得知:所建立的基于状态的检修(CBM)模型适用于安装了在线监测装置的设备,而基于检测的检修(IBM)模型适用于无在线监测装置的设备;最优决策从维修成本最小的决策过渡到最后每次决策时都选择大修,整个变化过程证明了应用Markov决策可以折中维修成本和故障损失得到经济最优决策。研究结果可为检修决策人员提供定量的检修计划。
[Abstract]:In order to solve the problem that the maintenance decision of power transmission and transformation equipment depends on the experience of decision makers to a great extent, the Markov decision process is applied to provide quantitative and concrete decision basis for maintenance decision. Firstly, the state transition probability of the equipment is solved by Markov process, and then the optimal decision of the model is solved by using the strategy iterative method. Finally, the data is replaced into the maintenance decision model, and the fault loss is changed to get the change of the decision result. By comparing the equipment fault loss corresponding to the decision change and the steady state probability of the equipment, it can be seen that the established state-based maintenance (CBM) model is suitable for the equipment with on-line monitoring device, while the inspection-based maintenance (IBM) model is suitable for the equipment without on-line monitoring device. The optimal decision is transitioned from the decision with the minimum maintenance cost to the overhaul at the last decision. The whole change process proves that the economic optimal decision can be obtained by using Markov decision to compromise the maintenance cost and fault loss. The research results can provide quantitative maintenance plan for maintenance decision makers.
【作者单位】：上海交通大学电气工程系;国网山东省电力公司电力科学研究院;国网北京电力科学研究院;
【基金】：国家自然科学基金(51477100) 中国博士后基金(2015M581614) 国家高技术研究发展计划(863计划)(2015AA050204) 国家电网公司科技项目(520626140020)~~
【分类号】：TM507;TM73

【参考文献】