Electric heating in winter can effectively promote the consumption of new energy and reduce wind and solar abandonment. However

the harmonics generated by large-scale electric heating equipment will affect the quality of the power grid. Therefore

it is necessary to analyze and predict the electric heating load in order to facilitate grid dispatch and safe operation

and relieve the load peak and frequency regulation pressure of the power system. In order to predict the electric heating load

this paper takes the actual electric heating load in a certain area as an example

and analyzes the four factors that affect the change of the electric heating load: stable load factor

weather sensitive factor

random load factor

date type factor. Finally

it is concluded that the main influencing factors of electric heating load changes are date type factors and weather sensitive factors. Therefore

this paper establishes a prediction model that considers the date type

and is based on the average temperature and the least squares method of human comfort. The prediction results show that both models have obtained high prediction accuracy

and both can be used for regional load forecasting.