Yearly Electricity Consumption Forecasting using a Nonhomogeneous Exponential Model Optimized by PSO Algorithm

被引：1

作者：

Xu, Xiaomin ^{[1
]}

Niu, Dongxiao ^{[1
]}

Meng, Ming ^{[2
]}

Shi, Huifeng ^{[3
]}

机构：

[1] North China Elect Power Univ, Sch Econ & Management, Beijing 102206, Peoples R China

[2] North China Elect Power Univ, Sch Econ & Management, Baoding 071003, Hebei, Peoples R China

[3] North China Elect Power Univ, Dept Math & Phs, Baoding 071003, Hebei, Peoples R China

来源：

APPLIED MATHEMATICS & INFORMATION SCIENCES | 2014年 / 8卷 / 03期

基金：

中国国家自然科学基金;

关键词：

Electricity Consumption Forecasting; Nonhomogeneous Exponential Model; PSO Algorithm; HIERARCHICAL NEURAL MODEL; ECONOMIC-GROWTH; LOAD; DEMAND; NETWORKS;

D O I：

10.12785/amis/080316

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

Yearly electricity consumption trends of most developing countries usually exhibit approximately exponential growth curves. An optimized nonhomogeneous exponential model (ONEM) is proposed as a method of forecasting electricity consumption by using trend extrapolation. The parameters of the nonhomogeneous exponential equation are obtained by using the inverse accumulated generating operation, discretizing the differential equation, minimizing the residual sum of squares (RSS), and accumulating the homogeneous exponential equation. Furthermore, to improve forecasting precision, particle swarm optimization (PSO) algorithm is used to optimize the equation parameters. To evaluate the forecasting performance for comparison, the said model and two other traditional methods are used to forecast the yearly electricity consumption of India. Empirical results show that this model is much better than traditional methods for each error analysis indicator.

引用

页码：1063 / 1069

页数：7

共 30 条

[11] The electricity consumption versus economic growth of the Polish economy [J].

Gurgul, Henryk ;

Lach, Lukasz .

ENERGY ECONOMICS, 2012, 34 (02) :500-510

[12] An evaluation of Bayesian techniques for controlling model complexity and selecting inputs in a neural network for short-term load forecasting [J].

Hippert, Henrique S. ;

Taylor, James W. .

NEURAL NETWORKS, 2010, 23 (03) :386-395

[13] Short-Term Load Forecasting With a New Nonsymmetric Penalty Function [J].

Kebriaei, Hamed ;

Araabi, Babak N. ;

Rahimi-Kian, Ashkan .

IEEE TRANSACTIONS ON POWER SYSTEMS, 2011, 26 (04) :1817-1825

[14]

Kennedy James., 2010, Particle Swarm Optimization, P760

[15] CO2 emissions and economic development: China's 12th five-year plan [J].

Meng, Ming ;

Niu, Dongxiao ;

Shang, Wei .

ENERGY POLICY, 2012, 42 :468-475

[16] Forecasting Monthly Electric Energy Consumption Using Feature Extraction [J].

Meng, Ming ;

Niu, Dongxiao ;

Sun, Wei .

ENERGIES, 2011, 4 (10) :1495-1507

[17] Annual electricity consumption analysis and forecasting of China based on few observations methods [J].

Meng, Ming ;

Niu, Dongxiao .

ENERGY CONVERSION AND MANAGEMENT, 2011, 52 (02) :953-957

[18] Short-term load forecasting using bayesian neural networks learned by Hybrid Monte Carlo algorithm [J].

Niu, Dong-xiao ;

Shi, Hui-feng ;

Wu, Desheng Dash .

APPLIED SOFT COMPUTING, 2012, 12 (06) :1822-1827

[19] Long-term forecasting of household and residential electric customers in Brazil [J].

Pessanha, J. F. M. ;

Leon, N. .

IEEE LATIN AMERICA TRANSACTIONS, 2012, 10 (02) :1537-1543

[20]

Petroleum B., 2008, BP STAT REV WORLD EN

← 1 2 3 →