Energy Consumption, Demand and Price Forecasting with Artificial Intelligence
Over the last two decades, electricity demand and price forecasting has become a fundamental decision-making tool for governments and energy companies. As electricity cannot be stored like other energies such as natural gas, predictive methods to forecast their use and production needs should be developed subject to changing energy demands. All factors of supply and demand will, therefore, have an immediate impact on the price of electricity on the spot market, including profits, energy security, and energy supply risk management. In addition to energy production cost, electricity price is determined by the changing nature of the supply and the consumer demand.
This Topic will focus on energy science and engineering or related research on electricity, gas and other forms of energy consumption prediction, demand, and price forecasting with artificial intelligence. It aims to publish cutting-edge, latest research, analyses, reviews, and evaluations related to energy demand, energy price and electricity load with a strong focus on analysis, energy modelling and prediction, integrated renewable and conventional energy system, energy planning and management systems powered by artificial intelligence. The Topic welcomes original papers, extensive reviews, critical insights, or exploratory scientific studies related to energy conservation, energy efficiency, renewable energy, electricity supply and demand, predicting energy storage methods, predicting energy load in buildings, and energy economics and policy issues. We are interested in multi-agent methods to provide insights as to whether demand or prices will be above marginal and how this might influence consumer behavior.
This Topic welcomes novel research on integrated energy consumption or demand side management and price forecasting or capabilities that consider methodologies like data analytics, data science, predictive models, experimental, analysis and optimization with a verification of all methods or application challenges. The topics may include:
- Deep learning and artificial intelligence methods for electricity demand and price prediction. This includes computational intelligence (machine learning, non-parametric, non-linear statistical) methods with learning, evolutionary or fuzziness methods capable of adapting to complex dynamic energy use, price, and demand changes.
- Multi-agent, multi-agent simulation, equilibrium, game theoretic models used to simulate energy price by matching demand and supply. Methods could include cost-based (or production-cost) models, equilibrium, or game theoretic approaches (such as the Nash–Cournot framework, supply function equilibrium, strategic production-cost models, and agent-based models).
- Fundamental and structural methods: physical and economic relationship analysis for electricity production or trading including the associations between fundamental drivers such as load variations due to weather conditions; system parameter changes, etc., and fundamental inputs modeled or predicted independently using statistical, reduced form and computational intelligence.
- Reduced form models (quantitative, stochastic) to characterize statistical properties of electricity price over time for risk management. This may provide hourly or minute time-scale demand and price forecasts using main characteristics of daily or other time-scale electricity prices, marginal distributions, price dynamics or correlation between commodity prices. These include Spot price models (parsimonious representation of the dynamics of spot prices) or forward price models(pricing of derivatives in a straightforward manner but only of those written on the forward price of electricity).
- Statistical (econometric, technical) methods developed to forecast loads or price using mathematical combination of price or exogenous factors like consumption or production or weather variables.
- Complex network analysis for electricity production and demand, relations from different local areas, distribution, and planning. Study of the correlation among the different energy sources.
- Probabilistic prediction of electricity demand using the deep transformer model.
- Analysis of electricity production and demand.
The Topic welcomes papers on energy consumption, demand and price forecasting over a range of horizons such as short-term forecasting (from a few minutes to a few days ahead as being of prime importance in day-to-day market operations), medium-term forecasting (days to a few months ahead, is generally preferred for balance sheet calculations, risk management and derivatives pricing) and long-term forecasting (predicted months, quarters or even year ahead to focus on long-term investment profitability analysis or energy planning, future electricity production site and other analysis). Any new methods developed and tested for economic and financial impact analysis studies on National Electricity Markets are especially welcome.
Prof. Dr. Ravinesh Deo
Prof. Dr. Sancho Salcedo-Sanz
Dr. Sujan Ghimire
Dr. David Casillas Pérez
- electricity prediction;
- energy load prediction;
- artificial intelligence and energy demand model;
- statistical models for electricity or energy analysis;
- deep learning for electricity price, load and demand modelling;
- national electricity market;
- weather effects on electricity demand prediction
|Journal Name||Impact Factor||CiteScore||Launched Year||First Decision (median)||APC|
|-||-||2020||21.8 Days||CHF 1200|
|-||-||2020||27.2 Days||CHF 1000|
|3.2||5.5||2008||15.7 Days||CHF 2600|
|3.7||5.9||2014||20 Days||CHF 1500|
|3.9||5.8||2009||18.3 Days||CHF 2400|
Preprints is a platform dedicated to making early versions of research outputs permanently available and citable. MDPI journals allow posting on preprint servers such as Preprints.org prior to publication. For more details about reprints, please visit https://www.preprints.org.