Overheating Risk Warning Model based on Thermal Circuit Model and Load Forecasting for Distribution Transformers

Distribution transformer is the key of power system for electrical conversion and transmission. Hot-spot temperature is caused of uneven heat dissipation during operation of transformer, it is an important indicator to reflect whether the transformer is overheated. It is difficult to directly obtain hot-spot temperature data of distribution transformers in practice. Many experts focus on improving more accurate thermal circuit model, but few people pay attention to the practical application of the model. To apply the model into practice, a combining the thermal circuit model with load forecasting to build overheating risk warning model is proposed in this paper. Load prediction is performed by using deep learning algorithm. The predicted results and future weather data are brought into the thermal circuit model to obtain the temperature data of transformers in future. This model can provide overheating risk warning to ensure the safe and stable operation of distribution transformers.