基于灰色滚动预测方法的变压器智能监测研究

Research on Intelligent Monitoring of Transformer Based on Grey Rolling Prediction Method

为将变压器温度保持在正常范围内,且在其发生故障前及时预测,推动智能电网建设,对灰色预测算法进行了分析。目前灰色预测算法预测连续性不好,预测精度需进一步提高。为弥补灰色预测算法上述缺点,利用MATLAB软件,编写预测程序。结合复合辛普森求积公式,对灰色预测程序进行改进,提高了预测精度。提出灰色滚动预测方法对变压器油中总烃值进行智能监测,预测变压器运行状态。通过分析,表明提出方法具备有效性和连续性,为提高变压器工作稳定性,推动智能电网建设提供方法支持。

In order to keep the temperature of the transformer within the normal range and predict its failure in time to promote the construction of smart grid,the gray prediction algorithm was analyzed.At present,the grey pre-diction algorithm has poor prediction continuity and the prediction accuracy needs to be further improved.The predic-tion program was written using MATLAB software in order to make up for the above shortcomings of the gray predic-tion algorithm.The gray prediction program had been enhanced which was combined with the composite Simpson quadrature formula,and the prediction accuracy was improved.The grey rolling prediction method was proposed to intelligently monitor the total hydrocarbon value in the transformer oil and predict the operating status of the trans-former.The analysis shows that the proposed method is of effectiveness and continuity,which provides a method for improving the working stability of the transformer and promoting the construction of smart grid.