基于哈里斯鹰优化器与支持向量回归的农村用电量预测方法研究，首发论文 -全讯担保网

research on rural electricity consumption prediction based on harris hawk optimizer and support vector regression

• 1、 tongzhou branch of nantong sanxin power supply service co., ltd , jiangsu 226300
• 2、college of iot engineering, hohai university, jiangsu, 213022

abstract：rural areas play an increasingly important role in economic development, and the demand for rural power consumption is also growing. rural power consumption prediction can provide decision support for managers, and its importance has become very prominent. this paper collects the data of rural power consumption panel of china\'s mainland provinces and cities from 1999 to 2018 from the national bureau of statistics, and forms rural power consumption samples, which are divided into training groups and test groups. randomly initialize the parameters of ε-support vector regression, use ε-support vector regression to train the training group data and generate multiple rural power consumption prediction models; based on the model, predict the test group and get multiple prediction accuracy values, select the optimal rural power consumption prediction model and the optimized ε- support vector regression parameter group, and use harris hawk optimizer to update the ε-support vector regression parameter group, use the updated parameter group to generate multiple rural power consumption prediction models again, cycle the above steps until the cut-off conditions are met, and get the global optimal rural power consumption prediction model, which can solve the new rural power consumption prediction problem. the feasibility and prediction performance of this method are verified by experiments.

keywords：