基于指数平滑的电站设备故障时间序列预测研究

Research on the Time Series Prediction of Power Plants' Equipment Faults with Exponential Smoothing Method

发电站正常高效运行是社会发展和人民生活中一个非常重要的基础保障工作。利用数据预测技术对电站设备故障的趋势进行预测分析,发现隐藏在数据中的有用信息,有助于管理者制定更加有效的电站管理策略以及进行工作的改革和创新。根据某发电集团设备故障的统计报告,利用指数平滑的时间序列预测方法对电站设备故障的趋势进行预测。实验结果合理,对设备故障的其他研究具有一定的借鉴意义。

It is an important basic security work for social development and people's life that power plants can run normally and efficiently. If we use data prediction techniques to analyze the trend of power plants' equipment faults, the useful information hidden in the data will be found. It is helpful for managers to establish a more effective management strategy of power plant and carry out the reform and innovation. According to the equipment fault statistics report of a certain power group, the paper uses a method of time series prediction named exponential smoothing to predict the trend of power plants' equipment faults. The result of experiments is reasonable and can be a reference for other studies on equipment faults.