Predicting Method for number of power load users based on multi-dimensional grey model and support vector machine

With the rapid development of high-tech industries such as medical treatment, transportation and biology, the analysis and prediction of large power load users becomes very critical, which can not only make the power distribution and the system operation more reasonable in the distribution network planning, but also optimize the user's power consumption experience to a great extent. In this paper, a combined prediction method based on the improved multi-dimensional grey prediction and support vector machine is proposed, which overcomes the shortcoming that grey prediction is only applicable to small sample short-term prediction and has no selectivity to specific data, realizes the complementary advantages of different models and avoids the limitations of a single model. Given the differences in the critical demand for electricity in different industries, this paper selects four typical large users in the high-tech industry as examples to predict the change of the number of these four large users.