Open AccessDistribution Network Low Voltage Prediction Method Based on Least Squares Support Vector Machine
Author(s) -
Zhiyang Yao,
Qingren Jin,
Min Guo,
Weidong Chen
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/440/3/032127
Subject(s) - support vector machine , voltage , computer science , low voltage , least squares support vector machine , least squares function approximation , transformer , function (biology) , algorithm , control theory (sociology) , mathematics , statistics , engineering , artificial intelligence , electrical engineering , control (management) , estimator , evolutionary biology , biology
Due to the lack of a direct functional relationship between voltage and voltage related parameters in the distribution network, it is difficult to scientifically carry out low voltage prediction. Therefore, a low voltage prediction method based on least squares support vector machine is proposed. The method uses the parameters related to low voltage and voltage in the distribution network as the basic data, constructs the optimal problem equation, forms the decision function, uses the decision function to predict the predicted samples, and outputs the lowest voltage value at the end of the low voltage side of the distribution transformer. Finally, the voltage value is analyzed and the low voltage of the distribution network is predicted. The actual distribution network data is used to simulate and verify the proposed method. The results show that the predicted value and the actual value error meet the requirements of estimation accuracy. It is demonstrated that the low-voltage prediction method based on least squares support vector machine is practical.