基于T-S模型的电力变压器顶层油温预测研究

为准确估算电力变压器绕组热点温度，对变压器安全运行和寿命评估提供辅助依据，建立了一个基于Takagi- Sugeno(T-S))的变压器顶层油温预测模型。模型的前件参数由模糊C均值聚类算法确定，后件参数由加权最小二乘法离线辨识，并用现场实测数据对模型后件参数进行在线调整。通过对实测数据的仿真实验表明，该模型以简单的模糊规则实现了变压器顶层油温的预测，且模型的预测精度优于IEEE推荐的变压器顶层油温经验模型，从而提高其绕组热点温度计算的精度。

Study on Prediction of Top-oil Temperature for Power Transformer Based on T-S Model

The hottest spot winding temperature (HST) is an important factor to operation safety and insulation life assessment of power transformer. A prediction model of top-oil temperature based on Takagi-Sugeno(T-S) fuzzy model has been established to estimate HST. The antecedent parameters of Takagi-Sugeno fuzzy model are off-line identified by fuzzy C-means algorithm,its consequent part is determined by weighted least squares method. And the consequent parameters are also adjusted on-line using recursive least squares method with oblivion factor. Simulation experiment has conducted by applying this model in on-site data,the result of which indicates that the simple rules got by T-S fuzzy mode can implement the prediction of top-oil temperature of power transformer. The prediction precision of this model is higher than experiment model of top-oil calculation proposed by IEEE,accordingly the calculation precision of HST is improved.