单变量
能源消耗
计算机科学
消费(社会学)
操作员(生物学)
中国
数据挖掘
数学优化
运筹学
计量经济学
数学
地理
多元统计
机器学习
工程类
电气工程
基因
转录因子
社会学
抑制因子
考古
化学
生物化学
社会科学
作者
X. Zhao,Xin Ma,Yubin Cai,Hong Yuan,Yanqiao Deng
出处
期刊:Grey systems
[Emerald (MCB UP)]
日期:2023-06-06
卷期号:13 (4): 629-656
被引量:5
标识
DOI:10.1108/gs-02-2023-0013
摘要
Purpose Considering the small sample size and non-linear characteristics of historical energy consumption data from certain provinces in Southwest China, the authors propose a hybrid accumulation operator and a hybrid accumulation grey univariate model as a more accurate and reliable methodology for forecasting energy consumption. This method can provide valuable decision-making support for policy makers involved in energy management and planning. Design/methodology/approach The hybrid accumulation operator is proposed by linearly combining the fractional-order accumulation operator and the new information priority accumulation. The new operator is then used to build a new grey system model, named the hybrid accumulation grey model (HAGM). An optimization algorithm based on the JAYA optimizer is then designed to solve the non-linear parameters θ , r , and γ of the proposed model. Four different types of curves are used to verify the prediction performance of the model for data series with completely different trends. Finally, the prediction performance of the model is applied to forecast the total energy consumption of Southwest Provinces in China using the real world data sets from 2010 to 2020. Findings The proposed HAGM is a general formulation of existing grey system models, including the fractional-order accumulation and new information priority accumulation. Results from the validation cases and real-world cases on forecasting the total energy consumption of Southwest Provinces in China illustrate that the proposed model outperforms the other seven models based on different modelling methods. Research limitations/implications The HAGM is used to forecast the total energy consumption of the Southwest Provinces of China from 2010 to 2020. The results indicate that the HAGM with HA has higher prediction accuracy and broader applicability than the seven comparative models, demonstrating its potential for use in the energy field. Practical implications The HAGM(1,1) is used to predict energy consumption of Southwest Provinces in China with the raw data from 2010 to 2020. The HAGM(1,1) with HA has higher prediction accuracy and wider applicability compared with some existing models, implying its high potential to be used in energy field. Originality/value Theoretically, this paper presents, for the first time, a hybrid accumulation grey univariate model based on a new hybrid accumulation operator. In terms of application, this work provides a new method for accurate forecasting of the total energy consumption for southwest provinces in China.
科研通智能强力驱动
Strongly Powered by AbleSci AI