Add Feeds
Publisher Windows
Streaming Headlines
Streaming Windows

Log out

Social Connect

OR Log into your Account below
Want One? Create New Account
Email:
Password:
How to Forecast Energy Demand

How to Forecast Energy Demand

2018-02-04 
| by Editor | Posted in Comment

Global energy demand is expected to increase by 28 percent between 2015 and 2040, according to the U.S. Energy Information Administration’s International Energy Outlook 2017 report. This type of energy demand forecasting is a relatively new area of mathematics. Policymakers became interested in understanding the nature of consumer demand and energy needs after the first oil shortage in the 1970s.

Since that time, methodologies have been developed that help guide energy decision-making around the world. A recent research article titled "An overview of energy demand forecasting methods” reports that “Demand forecasting plays a vital role in energy supply-demand management for both governments and private companies....Therefore, using models to accurately forecast the future energy consumption trends is an important issue for the power production and distribution systems.”

Lessons from the Past: Predicting the Future of Energy

Essentially, energy demand forecasting takes information from the past to predict the future. These include factors from various sources such as weather, economic activity, consumer use, price and production. Mathematicians and engineers then use this information to build a predicted future that can be a few minutes or a few years ahead.

“As long as the forecasting is performed based on shorter horizons, input data must be more accurate in terms of variations. This may be the reason that inputs such as temperature and humidity are mostly used in short-term forecasting according to the fact that the value of these variables can differ hourly,” the study explained.

Regardless of the time-frame being predicted, there are several common methods used to predict energy demand. These include a time series, an econometric analysis and end-use research. Each of these methods allows different forms of historical data to become the primary source of information.

Time Series Forecasting

Time series forecasting plots historical data on a chart and searches for any trends. If consumers in a particular area use 10 percent more electricity each year, for example, a time series graph would show the average increase as a straight line in the graph and assume that the future would continue to follow that trend.

Since time series forecasting is heavily based on statistics, these calculations and charts can be created in most commercial software. Time series graphs do not work well, however, if the historical data has a number of variables or is more complex. For example, if there is a significant drop in energy use for three to four years due to political or socioeconomic reasons, these charts will not capture the complexity of that anomaly.

Econometric Analysis

Econometric forecasting can define these more complex relationships in greater detail. These mathematical models work to make statistical sense of complex relationships. For example, an econometric model would look at the way economic activity, weather or oil prices effect energy demand. For example, extreme changes in weather often lead to increases in energy use as consumers decide to heat or cool their homes. In an econometric model, forecasters would see this causal relationship and predict that the relationship would stay the same in the future. So, an energy demand forecast would show increased energy demand at the height of winter or summer depending on the location.

Each of these factors can affect energy demand in a different way, so relationships are calculated based on affects that have occurred previously. These relationships are often calculated individually, and then mathematicians and engineers will look at all of the relationships to determine a prediction for future energy demands.

Another common forecasting method is end-use forecasting. This process reviews historical data of energy use for individual products. For example, researchers could look at energy used by power generators. They would examine historical data for residential and commercial power generator ownership, past energy usage and upcoming technologies. Based on these historical consumer trends, researchers would determine how power generator ownership and usage would change in the future and what that could mean for energy consumption. 

This type of forecasting can provide an in-depth look at specific energy use and energy efficiency. However, an immense amount of information is required to get enough data to form a solid energy usage forecast.

Energy demand forecasting can be a difficult task with any of these models. Each method requires a large amount of detailed historical data. This data can have unknown external variables, unpredictable trends and a complete lack of data from developing or war-torn countries. But it is worth the effort. These types of forecasts are invaluable when policymakers, energy producers and companies are planning for energy production, building new consumer products and recommending energy policies.

If you have a great Middle East or West African Energy industry news feed, blog, podcast or video channel that you’d like us to include on Energising The Industry Online, or if you have any ideas of new Middle East or West African Energy industry topics that you’d like us to focus on in the site, then please just contact us at energy@informa.com with details!








Post Comment
Join the conversation!Log in to leave a comment.

About Energising The Industry Online
All the latest Middle East & West African Energy industry news, all in one place. The best news sources, blogs, video channels, podcasts and social media chat for the Energy community. Curated and brought to you by Informa Exhibitions.
Contact Us
Follow Us