Gray linear regression model based on adaptive particle swarm optimization power load forecasting method

2018 International Conference on Civil, Architecture and Disaster Prevention

土木建筑工程

Qi, Caijuan^1 ; Zhang, Kun^1 ; Shi, Shuhong^2 ; Zhang, Qian^2

State Grid Ningxia Electric Power Eco-Tech Research Institute, Yinchuan, Ningxia Province

750004, China^1

Beijing Kedong Electric Power Control System CO. LTD, Beijing

100192, China^2

关键词: Adaptive particle swarm optimization algorithm;    Adaptive particle swarm optimizations;    Annual electricity consumption;    Linear regression models;    Particle swarm optimization algorithm;    Power load forecasting;    Prediction accuracy;    Residual correction;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/218/1/012144/pdf
DOI  :  10.1088/1755-1315/218/1/012144

学科分类：土木及结构工程学

来源: IOP

【 摘 要 】

Aiming at the trend of "S" growth in annual electricity consumption in Ningxia, the gray linear regression combined model (GLM) load forecasting method is adopted, which can improve the lack of exponential growth trend in linear regression model and the lack of linear factors in the GM(1,1) model. However, the traditional GLM model has theoretical defects in the process of solving parameters, and there are limitations in the application range. Therefore, this paper introduces an adaptive particle swarm optimization algorithm which is more efficient than the standard particle swarm optimization algorithm, and combines it with the gray linear regression combination model. The adaptive particle swarm optimization algorithm is used to solve the parameters of the GLM model, then carry out residual correction. Adaptive particle swarm optimization gray linear regression combination model (APSO-GLM) is proposed. The example analysis shows that the model overcomes the defects of the GM(1,1) model and the linear regression model. Compared with the single model, the prediction accuracy is higher and it has a wider application range than the traditional GLM model.