摘要
在综合考虑能源价格、气候变化、政策规定、经济环境及交易行情的前提下,运用皮尔森相关系数法筛选出4个影响碳交易价格的关键因素,并将其作为预测模型解释变量;其次,运用多方向螺旋搜索的海鸥算法优化随机森林模型,建立预测模型;最后,利用某地区日平均碳交易价格数据对预测模型进行验证。结果表明,所提模型预测效果明显好于其他同类型预测算法。
Taking into account energy prices,climate change,policy regulations,economic environment and trading trends,the Pearson correlation coefficient method was used to identify 4 key factors affecting carbon trading prices which are used as explanatory variables of the prediction model;secondly,the random forest model is optimized by the seagull algorithm with multi-direction spiral search,and establish the prediction model;finally,the prediction model is validated using the daily average carbon trading price data of a certain area.The results show that the prediction effect of the proposed model is obviously better than other similar prediction algorithms.
作者
李金颖
黄湘敏
LI Jinying;HUANG Xiangmin(Department of Economics and Management,North China Electric Power University,Baoding 071003,China)
出处
《电力科学与工程》
2023年第5期10-16,共7页
Electric Power Science and Engineering
基金
河北省社科基金重点项目(HB21YJ002)。
关键词
碳排放权交易
海鸥算法
随机森林模型
价格预测
carbon emission trading
seagull optimization algorithm
random forests model
price prediction