Shewit Tsegaye
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Shewit Tsegaye has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Kinde Anlay Fante
Faculty of Electrical and Computer Engineering, Jimma University, Jimma, Ethiopia
3 shared publications - Biniyam Tadesse
Protection and Communication Department, Southwest Region TSOM, Ethiopian Electric Power (EEP), Addis Ababa, Ethiopia
1 shared publication - Emiyamrew Minaye Molla
Department of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan
1 shared publication - Henok Ayele Behabtu
Faculty of Electrical and Computer Engineering, Jimma University, Jimma 378, Ethiopia
1 shared publication Kasech Tsegaye Turufi
Faculty of Computing and Informatics, Jimma Institute of Technology, Jimma University (JU), Jimma, Ethiopia
1 shared publicationKifle Godana Heyi
Faculty of Electrical and Computer Engineering, Jimma Institute of Technology, Jimma University (JU), Jimma, Ethiopia
1 shared publication
Latest Publications
- Open Access
- 19 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 30627-30643
Contingency Analysis of Ethiopian Extra High Voltage Transmission Networks With Deep Neural Networks
Contingency analysis (CA) is a vital technique of assessing power system resilience during outages by detecting overloads, voltage issues, and generator faults, strengthening grid security and emergency response planning. For the Ethiopian Electric Power (EEP) network, an N‑1 contingency analysis was conducted on Extra High Voltage (EHV) transmission lines. A dual‑depth Deep Feed Forward Neural Network (DFFNN) was employed on MATLAB/MATPOWER platform to predict bus voltage magnitudes and active power flows across branches. The DFFNN achieved strong training performance, with Mean Square Error (MSE) values of 3.887 x 10-5 and 7.32 x10-3, alongside high regression values of 0.99234 and 0.99479, confirming its predictive accuracy. These predictions were used to implement a novel Hybrid Performance Index (PIhyb), which ranks contingencies according to severity. Results revealed that failure of either circuit of the Dedessa–Holeta 500 kV double-line, sourced from the Great Ethiopian Renaissance Dam (GERD), was the most severe contingency. This event produced a PIhyb of 1.798, causing voltage instability at 19 buses and severe overloading of the Dedessa–Nekemte 132 kV line. Subsequent critical contingencies were also linked to GERD, with PIhyb values of 0.473 and 0.321. Corrective and preventive measures are recommended to mitigate these risks, ensuring protection against cascading failures that could ultimately lead to a total blackout.
- 19 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 30627-30643
Open AccessContingency Analysis of Ethiopian Extra High Voltage Transmission Networks With Deep Neural Networks
Contingency analysis (CA) is a vital technique of assessing power system resilience during outages by detecting overloads, voltage issues, and generator faults, strengthening grid security and emergency response planning. For the Ethiopian Electric Power (EEP) network, an N‑1 contingency analysis was conducted on Extra High Voltage (EHV) transmission lines. A dual‑depth Deep Feed Forward Neural Network (DFFNN) was employed on MATLAB/MATPOWER platform to predict bus voltage magnitudes and active power flows across branches. The DFFNN achieved strong training performance, with Mean Square Error (MSE) values of 3.887 x 10-5 and 7.32 x10-3, alongside high regression values of 0.99234 and 0.99479, confirming its predictive accuracy. These predictions were used to implement a novel Hybrid Performance Index (PIhyb), which ranks contingencies according to severity. Results revealed that failure of either circuit of the Dedessa–Holeta 500 kV double-line, sourced from the Great Ethiopian Renaissance Dam (GERD), was the most severe contingency. This event produced a PIhyb of 1.798, causing voltage instability at 19 buses and severe overloading of the Dedessa–Nekemte 132 kV line. Subsequent critical contingencies were also linked to GERD, with PIhyb values of 0.473 and 0.321. Corrective and preventive measures are recommended to mitigate these risks, ensuring protection against cascading failures that could ultimately lead to a total blackout.
- Open Access
- 4 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 114845-114864
Deep Neural Networks in Smart Grid Digital Twins: Evolution, Challenges, and Future Outlooks
The increasing demand for reliable, sustainable, and efficient energy systems is accelerating the adoption of innovative technologies in the power sector. Among these advancements, the integration of Deep Neural Networks (DNNs) with Digital Twins (DTs) emerged as a groundbreaking innovation, particularly in smart grids (SGs). DNNs, subset of artificial intelligence, process complex datasets and are able to identify intricate patterns, proving invaluable for predictive analytics, optimization, and real-time decision-making. DTs serve as dynamic virtual representations of physical grid systems, enabling engineers to monitor, simulate, and optimize grid performance with outstanding precision. This strategic review presents the evolution of DNNs in Smart Grid DTs (SGDTs), highlighting their transformative role in enhancing grid operations, outlining the significant challenges that hindered their widespread adoption, and providing a forward-looking perspective on their future potential in revolutionizing energy systems. The scope of this study encompasses exploring evolution of smart grids, the advancements in communication systems within SGDTs, and the integration of DNNs in SGDTs by taking lessons from existing DT architectures and case studies. It also categorized challenges into physical, intelligent, and virtual layers, to emphasize hybrid DNNs’ role in addressing big data and intermittent energy issues.
- 4 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 114845-114864
Open AccessDeep Neural Networks in Smart Grid Digital Twins: Evolution, Challenges, and Future Outlooks
The increasing demand for reliable, sustainable, and efficient energy systems is accelerating the adoption of innovative technologies in the power sector. Among these advancements, the integration of Deep Neural Networks (DNNs) with Digital Twins (DTs) emerged as a groundbreaking innovation, particularly in smart grids (SGs). DNNs, subset of artificial intelligence, process complex datasets and are able to identify intricate patterns, proving invaluable for predictive analytics, optimization, and real-time decision-making. DTs serve as dynamic virtual representations of physical grid systems, enabling engineers to monitor, simulate, and optimize grid performance with outstanding precision. This strategic review presents the evolution of DNNs in Smart Grid DTs (SGDTs), highlighting their transformative role in enhancing grid operations, outlining the significant challenges that hindered their widespread adoption, and providing a forward-looking perspective on their future potential in revolutionizing energy systems. The scope of this study encompasses exploring evolution of smart grids, the advancements in communication systems within SGDTs, and the integration of DNNs in SGDTs by taking lessons from existing DT architectures and case studies. It also categorized challenges into physical, intelligent, and virtual layers, to emphasize hybrid DNNs’ role in addressing big data and intermittent energy issues.
- Open Access
- 4 November 2024
- Published by Institution of Engineering and Technology (IET) in The Journal of Engineering
- Vol. 2024 (11)
Short‐term energy forecasting using deep neural networks: Prospects and challenges
This study presents an in‐depth overview of deep neural networks (DNN) and their hybrid applications for short‐term energy forecasting (STEF). It examines DNN‐based STEF from three perspectives: basics, challenges, and prospects. The study compares recent literature using metrics like mean absolute error (MAE), mean average percentage error (MAPE), and root mean square error (RMSE). Findings indicate that combining automated data‐driven models with enhanced DNNs effectively addresses forecasting challenges. It also highlights the role of DNNs in integrating energy prosumers, renewable energy systems, microgrids, big data, and smart grids to improve STEF.
- 4 November 2024
- Published by Institution of Engineering and Technology (IET) in The Journal of Engineering
- Vol. 2024 (11)
Open AccessShort‐term energy forecasting using deep neural networks: Prospects and challenges
This study presents an in‐depth overview of deep neural networks (DNN) and their hybrid applications for short‐term energy forecasting (STEF). It examines DNN‐based STEF from three perspectives: basics, challenges, and prospects. The study compares recent literature using metrics like mean absolute error (MAE), mean average percentage error (MAPE), and root mean square error (RMSE). Findings indicate that combining automated data‐driven models with enhanced DNNs effectively addresses forecasting challenges. It also highlights the role of DNNs in integrating energy prosumers, renewable energy systems, microgrids, big data, and smart grids to improve STEF.
- Open Access
- 23 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 186856-186871
Short-Term Load Forecasting for Electrical Power Distribution Systems Using Enhanced Deep Neural Networks
The rationale for using enhanced Deep Neural Networks (DNNs) in the power distribution system for short-term load forecasting originates from a thorough analysis of current trends, the emergence of the state-of-the-art use cases and approaches in Short-Term Load Forecasting (STLF). STLF plays a crucial role in economic load dispatch, hydrothermal coordination, system security assessment, load shedding, unit commitment, and cost-effective risk management in power systems with renewable energy sources. In this study, we introduce a Long Short-Term Memory (LSTM), augmented with enhancements inspired by an Efficient and Parallel Genetic Algorithm (EPGA) for the STLF of Jimma town power distribution system. To forecast the load in the short term, the model takes into account wind direction, wind speed, humidity, temperature, season, load history, and peak load due to holidays. The optimal linear combination of inputs for determining daily load is derived using EPGA and data from Ethiopian Electric Utility (EEU). The linearly combined data is then fed into the LSTM model for load prediction. During training, this allows the LSTM model to focus on the pattern of a single time-series data rather than the best combination of many input patterns. The proposed method uses the combination of EPGA and LSTM models for accurate STLF. Thorough experimental analysis indicates that the root-mean-squared error (RMSE) achieved by the EPGA-enhanced LSTM for Jimma town power distribution system STLF is about 43.87. This represents a 7.486% improvement over the prediction obtained using only LSTM model. Additionally, the mean average percentage error (MAPE) for five sample loads used to test the EPGA-enhanced LSTM prediction method further supports the robustness of the proposed method.
- 23 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 186856-186871
Open AccessShort-Term Load Forecasting for Electrical Power Distribution Systems Using Enhanced Deep Neural Networks
The rationale for using enhanced Deep Neural Networks (DNNs) in the power distribution system for short-term load forecasting originates from a thorough analysis of current trends, the emergence of the state-of-the-art use cases and approaches in Short-Term Load Forecasting (STLF). STLF plays a crucial role in economic load dispatch, hydrothermal coordination, system security assessment, load shedding, unit commitment, and cost-effective risk management in power systems with renewable energy sources. In this study, we introduce a Long Short-Term Memory (LSTM), augmented with enhancements inspired by an Efficient and Parallel Genetic Algorithm (EPGA) for the STLF of Jimma town power distribution system. To forecast the load in the short term, the model takes into account wind direction, wind speed, humidity, temperature, season, load history, and peak load due to holidays. The optimal linear combination of inputs for determining daily load is derived using EPGA and data from Ethiopian Electric Utility (EEU). The linearly combined data is then fed into the LSTM model for load prediction. During training, this allows the LSTM model to focus on the pattern of a single time-series data rather than the best combination of many input patterns. The proposed method uses the combination of EPGA and LSTM models for accurate STLF. Thorough experimental analysis indicates that the root-mean-squared error (RMSE) achieved by the EPGA-enhanced LSTM for Jimma town power distribution system STLF is about 43.87. This represents a 7.486% improvement over the prediction obtained using only LSTM model. Additionally, the mean average percentage error (MAPE) for five sample loads used to test the EPGA-enhanced LSTM prediction method further supports the robustness of the proposed method.