Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks
“Energy Trilemma” has recently received an increasing concern among policy makers. The trilemma conceptual framework is based on three main dimensions: environmental sustainability, energy equity, and energy security. Energy security reflects a nation’s capability to meet current and future energy d...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
AUC Knowledge Fountain
2021
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613420059099136 |
|---|---|
| access_status_str | Open Access |
| author | Eissa, Haidy |
| author_browse | Eissa, Haidy |
| author_facet | Eissa, Haidy |
| author_sort | Eissa, Haidy |
| collection | Thesis |
| description | “Energy Trilemma” has recently received an increasing concern among policy makers. The trilemma conceptual framework is based on three main dimensions: environmental sustainability, energy equity, and energy security. Energy security reflects a nation’s capability to meet current and future energy demand. Rational energy planning is thus a fundamental aspect to articulate energy policies. The energy system is huge and complex, accordingly in order to guarantee the availability of energy supply, it is necessary to implement strategies on the consumption side. Energy modeling is a tool that helps policy makers and researchers understand the fluctuations in the energy system. Over the years, there have been many attempts to develop energy system models, which have varying degrees of success. Artificial Neural Networks (ANN) is one of the modeling techniques that shows great performance in modeling the energy consumption side, which has the time-series characteristics of complexity and nonlinearity. Static feedforward neural networks are extensively used in literature due to their simplicity. In this thesis, we propose two artificial neural network topologies: feedforward and Nonlinear Autoregressive with Exogenous Inputs (NARX) neural networks, where four separate ANN models, are formulated to study and forecast the annual final energy consumption for four different sectors in the United Kingdom till 2035: transport, domestic, services, industrial. The outputs of all models are finally summed up to yield UK’s total final energy consumption. Furthermore, in this thesis, we use the Bayesian optimization algorithm to search for the optimal network hyperparameters. Moreover, instead of arbitrarily selecting input parameters in a qualitative manner, a sequential backward selection technique is used to exclude any uninformative input variables that have no predictive power, and eventually select the optimal set of input parameters. The network performance is measured with regards to various accuracy metrics such as Root Mean Square Error (RMSE), and Mean Average Percentage Error (MAPE). The resulting forecasts are eventually compared to the final energy consumption outlook from UK’s governmental Department for Business, Energy & Industrial Strategy (BEIS). The comparisons show that the NARX model offers superior results compared to the feedforward model in terms of the accuracy metrics as well as the long-term forecasts. Moreover, the developed NARX model succeeded in having a better performance than other models developed in the literature. |
| format | Thesis |
| id | oai:fount.aucegypt.edu:etds-2711 |
| institution | American University in Cairo (Egypt) |
| last_indexed | 2026-06-10T12:35:51.500Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from AUC Knowledge Fountain — bepress |
| publishDate | 2021 |
| publishDateRange | 2021 |
| publishDateSort | 2021 |
| publisher | AUC Knowledge Fountain |
| publisherStr | AUC Knowledge Fountain |
| record_format | dspace |
| source_str | AUC Knowledge Fountain — bepress |
| spelling | oai:fount.aucegypt.edu:etds-2711 Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks Eissa, Haidy “Energy Trilemma” has recently received an increasing concern among policy makers. The trilemma conceptual framework is based on three main dimensions: environmental sustainability, energy equity, and energy security. Energy security reflects a nation’s capability to meet current and future energy demand. Rational energy planning is thus a fundamental aspect to articulate energy policies. The energy system is huge and complex, accordingly in order to guarantee the availability of energy supply, it is necessary to implement strategies on the consumption side. Energy modeling is a tool that helps policy makers and researchers understand the fluctuations in the energy system. Over the years, there have been many attempts to develop energy system models, which have varying degrees of success. Artificial Neural Networks (ANN) is one of the modeling techniques that shows great performance in modeling the energy consumption side, which has the time-series characteristics of complexity and nonlinearity. Static feedforward neural networks are extensively used in literature due to their simplicity. In this thesis, we propose two artificial neural network topologies: feedforward and Nonlinear Autoregressive with Exogenous Inputs (NARX) neural networks, where four separate ANN models, are formulated to study and forecast the annual final energy consumption for four different sectors in the United Kingdom till 2035: transport, domestic, services, industrial. The outputs of all models are finally summed up to yield UK’s total final energy consumption. Furthermore, in this thesis, we use the Bayesian optimization algorithm to search for the optimal network hyperparameters. Moreover, instead of arbitrarily selecting input parameters in a qualitative manner, a sequential backward selection technique is used to exclude any uninformative input variables that have no predictive power, and eventually select the optimal set of input parameters. The network performance is measured with regards to various accuracy metrics such as Root Mean Square Error (RMSE), and Mean Average Percentage Error (MAPE). The resulting forecasts are eventually compared to the final energy consumption outlook from UK’s governmental Department for Business, Energy & Industrial Strategy (BEIS). The comparisons show that the NARX model offers superior results compared to the feedforward model in terms of the accuracy metrics as well as the long-term forecasts. Moreover, the developed NARX model succeeded in having a better performance than other models developed in the literature. 2021-12-31T08:00:00Z thesis application/pdf https://fount.aucegypt.edu/etds/1679 https://fount.aucegypt.edu/context/etds/article/2711/viewcontent/Haidy_Mahmoud_Eissa_Thesis.pdf Theses and Dissertations AUC Knowledge Fountain Energy Consumption Forecast ANN Artificial Intelligence Optimization Neural Network Artificial Intelligence and Robotics Energy Systems Industrial Engineering |
| spellingShingle | Energy Consumption Forecast ANN Artificial Intelligence Optimization Neural Network Artificial Intelligence and Robotics Energy Systems Industrial Engineering Eissa, Haidy Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks |
| title | Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks |
| title_full | Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks |
| title_fullStr | Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks |
| title_full_unstemmed | Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks |
| title_short | Energy Planning Model Design for Forecasting the Final Energy Consumption using Artificial Neural Networks |
| title_sort | energy planning model design for forecasting the final energy consumption using artificial neural networks |
| topic | Energy Consumption Forecast ANN Artificial Intelligence Optimization Neural Network Artificial Intelligence and Robotics Energy Systems Industrial Engineering |
| url | https://fount.aucegypt.edu/etds/1679 https://fount.aucegypt.edu/context/etds/article/2711/viewcontent/Haidy_Mahmoud_Eissa_Thesis.pdf |
| work_keys_str_mv | AT eissahaidy energyplanningmodeldesignforforecastingthefinalenergyconsumptionusingartificialneuralnetworks |