基于Adam算法的变压器顶层油温预测方法研究

Research on Adam-Based Prediction of Transformer Top Oil Temperature

顶层油温是反映电力变压器负载能力和绝缘老化程度的重要指标。为准确预测顶层油温,文中实现了一种基于自适应动量估计(Adaptive Momentum Estimation,Adam)算法优化的神经网络变压器顶层油温预测方法。首先运用灰色相关性分析法计算变压器各监测量与顶层油温的相关性,构建预测顶层油温的后向传播(Back Propagation,BP)神经网络模型;结合变压器状态监测历史数据,利用Adam算法训练模型和数据验证,结果表明该模型预测值与实际测量值基本一致,与Susa D热路模型和基于随机梯度下降法神经网络模型预测结果相比较,该方法预测精度相比其他两者分别提高了78.1%和33.9%;最后选择不同的变压器进行建模预测,结果表明该方法具有普遍适用能力。

Top oil temperature(TOT)is an important indicator reflecting the load capacity and insulation aging of the transformer.In order to predict the TOT accurately,this paper proposes a transformer top oil temperature prediction method based on BP(Back Propagation)neural networks optimized by Adam(Adaptive Momentum Estimation).Firstly,the neural networks prediction model(NNPM)of the TOT is established by applying the grey relational analysis method to calculate the correlation between other state variables of the transformer and the TOT.Then NNPM of TOT is trained using historical data of transformer and Adam optimization algorithm,which suggests that the prediction results of NNPM optimized by Adam of TOT are in accordance with measured results.Compared with D Susa thermal circuit model and NNPM trained by SGD,the prediction accuracy of NNPM optimized by Adam is improved by 78.1%and 33.9%respectively.Finally,different transformers are used to model and predict,and the results show that Adam-based NNPM of TOT is widely applicable.