中文摘要：

针对输变电设备故障具有较强的随机性和模糊性的特点,提出了采用云模型的不确定推理预测输变电设备故障的方法。挖掘出设备当前故障发展趋势与不同年份健康状况评价数据之间的关系,构建了基于条件云发生器的云语言预测规则,在此基础上建立了基于云推理的输变电设备故障率预测方法。实例分析结果表明,云推理模型的故障率预测结果符合该地区电网实际可靠性数据统计规律;当设备健康指数＆gt;60时,该方法的预测结果比传统反演法更符合实际设备情况。此外,该方法能反映设备的实时优劣状况,具有较强的实用性。

英文摘要：

Power transmission equipment failure is characterized by strong randomness and fuzziness. Accordingly, we put forward a method for predicting the failure rate of power transmission equipment by cloud reasoning. We revealed the relevance between current equipment failure trend and annual health status evaluation data, built up a cloud language pre- diction rule based on the conditional cloud generator, and consequently established the prediction method. Analysis of practical data indicates that the prediction results of failure rate by the cloud reasoning model are consistent with the sta- tistical regularities of a regional power grid. Compared with the traditional inversion methods, when the equipment health index is higher than 60, the proposed method can predict the equipment status more suitably than the traditional inversion methods. Additionally, the proposed method is very applicable since it can reflect the real-time status of equipment.