|
Iran's Electrical Energy Demand Forecasting Using Meta-Heuristic Algorithms
|
Aliyeh Kazemi *   , Raja Bashirzadeh , Sara Aryaee |
| Tehran University , aliyehkazemi@ut.ac.ir |
|
|
Abstract: (2140 Views) |
| This study aims to forecast Iran's electricity demand by using meta-heuristic algorithms, and based on economic and social indexes. To approach the goal, two strategies are considered. In the first strategy, genetic algorithm (GA), particle swarm optimization (PSO), and imperialist competitive algorithm (ICA) are used to determine equations of electricity demand based on economic and social indexes consisted of population, gross domestic product (GDP), electricity price, and electricity consumption during the years 1968 to 2015. In this regard, linear and nonlinear models are developed. In the second strategy, artificial neural networks (ANNs) trained by meta-heuristic algorithms (GA, PSO, and ICA) are used to forecast electricity demand. The results show that nonlinear PSO with %2.85 mean absolute percentage error (MAPE) is a suitable model to forecast Iran's electrical energy demand. Iran's electricity demand would reach 324 terawatt-hours (TWh) up to the year 2025. |
|
| Keywords: Electricity demand forecasting, Artificial neural networks, Genetic algorithm, Particle swarm optimization, Imperialist competitive algorithm |
|
|
Full-Text [PDF 567 kb]
(674 Downloads)
|
Type of Study: Research |
Subject:
Others
Received: 2020/04/10 | Accepted: 2020/05/4 | Published: 2020/02/29
|
|
|
|
|
|
|
| Send email to the article author |
|
|
|
| Add your comments about this article |
|
|
|
