基于灰色神经网络的变压器绕组热点温度预测

Transformer Winding Hot-spot Temperature Prediction Based On the Grey Neural Network

绕组热点温度影响着电力变压器的绝缘性能和运行寿命,为提高变压器运行的可靠性,对变压器进行三维温度场建模,通过分析变压器运行时的温度场,发现高低压绕组的温度自上而下减小,热点温度出现在端部,由此,通过监测变压器绕组上端部温度、顶层油温、中层油温、底层油温及环境温度,对绕组热点温度进行预测,为提高预测的准确性,构建绕组热点温度的灰色神经网络模型,并结合实验室变压器温度数据进行测试,结果表明:预测值与实测值之间的误差很小,灰色神经网络能对变压器绕组热点温度进行有效预测。

Winding hot-spot temperature affects the power transformer insulation performance and operating life. In order to improve the reliability of transformer operation, the three-dimensional temperature field model of the transformer is established, analysis the temperature field of the operating transformer, found that the temperature of the high and low voltage winding reduce from top to bottom, and the hot spot temperature appeared at the top. Thus, by monitoring the transformer winding temperature of upper part, top oil temperature, middle oil temperature, bottom oil temperature and environmental temperature, forecast the hot- spot temperature of the winding, in order to improve the accuracy of the prediction, build the grey neural network model of the winding hot spot temperature, and test combined with transformer temperature data of laboratory. The results showed that ： the error between the predicted values and measured values is very small, and the grey neural network is effective in prediction of transformer winding hot-spot temperature.