Abstract: Aiming at the lack of direct functional relationship between voltage-related parameters in the distribution network,it is difficult to scientifically carry out low voltage prediction.This paper proposed a low voltage prediction method based on support vector machine.The method uses a large number of distribution network voltage and voltage-related data as the training sample set,and uses the support vector machine method to find the optimal function relation in many samples,and forms a decision function to carry out the voltage prediction at the end of the low voltage side of the distribution network.The actual monitoring voltage of the distribution network in a certain area is compared with the predicted voltage of the support vector machine.The error between the predicted value of the parameter and the actual monitored value is small,which verifies the validity and feasibility of the low voltage prediction method based on vector machine in the distribution network.

Key words: distribution network, support vector machine, decision function, low voltage prediction