论文部分内容阅读
基于变压器油性能参数之间的关联性,采用BP神经网络方法,在Matlab平台下研究预测多参数关联变压器油的性能,利用变压器油日常的监测数据,建立击穿电压与4个影响因素的关联模型。论文分别就常规BP算法和变学习速率、变动量因子的改进BP算法进行了比较研究,结果表明,改进BP算法模型的预测结果精度较高,预测值与实际值的相对误差在5%左右。本方法可以为变压器故障的早期诊断、预测防范和及时处理提供科学依据,具有重要的实际应用价值。
Based on the correlation between performance parameters of transformer oil, BP neural network method was used to study the performance of multi-parameter correlation transformer oil under Matlab platform. The daily monitoring data of transformer oil was used to establish the correlation between breakdown voltage and four influencing factors model. The paper makes a comparative study of the conventional BP algorithm and the improved BP algorithm with variable learning rate and variation factor. The results show that the prediction accuracy of the improved BP algorithm model is high, and the relative error between the predicted value and the actual value is about 5%. The method can provide scientific basis for early diagnosis, prediction prevention and timely treatment of transformer faults, and has important practical application value.