Is it possible to forecast yearly electrical energy consumption by the mentioned method?
Is there any online tool?
Dear Abolfazal,
Yes, you can forecast energy demand by using ARMA/ ARIMA models. You are required to get historical data for electricity consumption. This is uni variate model.
Besides ARIMA and "regression" analysis, you can also use heuristic algorithms (PSO, GA, etc) to do it, after defining your explanatory variables [input vector]. Some references are given:
M.H. Amjadi, H. Nezamabadi-pour, M.M. Farsangi (2010). Estimation of electricity demand of Iran using two heuristic algorithms, Energy Conversion and Management 51: 493–497.
M. Yousefi, M. Omid, Sh. Rafiee, S.F. Ghaderi (2013). Strategic planning for minimizing CO2 emissions using LP model based on forecasted energy demand by PSO Algorithm and ANN, International Journal of Energy, Environment and Economics 4(6):1041-1052. [Fulltext is available from the following link]
Article Strategic planning for minimizing CO2 emissions using LP mod...
If you have a regression model when the electricity demand is explained by some independent variables, you should estimate the model and after that extrapolate based on the results of estimation and using the values of the independent variables.
If you have the univariate time series of electricity, you should modelise your time series using the ARIMA models. From this you predict the values of electricity consumption or demand.
Time series method with consolidation of different variables that determine demand for energy help you to forecast future demand for energy.
You can search for time series analysis you will get a lot for Journal paper which will be related to your problem.
I found these papers which may help you
1)"Short-Term Load Forecasting with Neural Network Ensembles: A Comparative Study"
2)"Short-Term Multinodal Load Forecasting Using a Modified General Regression Neural Network"
3)"Neural-Network-Based Demand Forecasting in a Deregulated Environment"
I hope this will help you.
In order to obtain a realistic regression, you will need other data (at least another variable), as you have already been told.
Since you also asked for a tool, once you have all the required stuff for the regression, you might use the MSExcel feature for data analysis; it's quite simple, but very useful.
- Badges
- Science topic
- Similar topics
- Control Systems Engineering
- Filtering