基于改进CBR的电力通信现场运维辅助诊断研究

Research on Auxiliary Diagnosis of Field Operation and Maintenance of Electric Power Communication Based on Improved CBR

针对电力通信现场运维的管理需要,在对基于案例推理(CBR)技术研究的基础上,提出了利用BP神经网络改进CBR检索效率的优化方法,实现现场问题诊断的自学习和自成长,有效避免了传统CBR算法存在的相似匹配度不高、收敛速度慢等问题。通过实验仿真结果表明,基于BP神经网络的CBR检索算法具有较高的查全率和查准率,可为现场运维人员提供较为可信的辅助诊断案例,具有较强的实用性。

According to the need of electric power communication on-site operation and maintenance management,based on case-based reasoning(CBR)research results,the optimization method is proposed to improve the retrieval efficiency of CBR using BP neural network to realize problem diagnosis self-learning and self-growth,effectively avoid the traditional CBR algorithm that has the similar matching degree is not high,convergence slow and other issues.The simulation results show that the CBR retrieval algorithm based on BP neural network has high recall and precision,and it can provide more reliable auxiliary diagnosis cases for field operation and maintenance personnel,and has strong practicability.