JavaScript is disabled for your browser. Some features of this site may not work without it.
| dc.contributor.author | RODAS, FELEKE JEMBERIE | |
| dc.date.accessioned | 2024-12-11T06:29:29Z | |
| dc.date.available | 2024-12-11T06:29:29Z | |
| dc.date.issued | 2023-12 | |
| dc.identifier.uri | http://ir.bdu.edu.et/handle/123456789/16350 | |
| dc.description.abstract | The power system has confronted repeated and widely spread power failure due to increasing of load demand for electrification and industrialization throughout the world. Grid overloading especially during peak hours has also become common problem and can eventually lead to grid failure. For this reason, this thesis focuses on these issues and an Artificial Neural Network is used. The ANN was trained using 70 % of the extracted data, and the remaining 30 % was used for testing and validation. 1024 data was extracted from different load variation. In this thesis, an optimal energy scheduling of grid connected micro grid has been made to generate clean energy, and increase overall efficiency of the grid. The proposed micro grid included renewable energy sources. The scheduling of the system has been done using solar and wind power as distributed generators which are connected to the IEEE 14 bus test system. The extraction features of the ANN have been taken from the total loads of the test system and battery state of charge as an input, and the status of the switch to solar power, wind power, battery charging and battery discharging as an output. The PV solar system has generated 32.9 MW and covered 35 % of the critical load of the IEEE 14 bus test system which is 94 MW. While the wind system covered 65 % of this load by generating 61.1 MW using 82 wind turbines. The IEEE 14 bus test system has been simulated at normal load (100 %), under load (50-99%), and overload conditions (101-150%). The PSO based ANN training and ANN training without PSO has been compared, the percentage of improvement of PSO based ANN training is 24.21 %. The PSO based ANN training has a better performance when it compared to the training of ANN without PSO. For validation, three conditions have been presented which shows the working of the optimal energy scheduling. The modeling and simulation of energy source components of the micro grid has been done in MATLAB/SIMULINK environment. Keywords: ANN, MATLAB /Simulink, Micro grid, PID, PS, Distributed Generators | en_US |
| dc.language.iso | en_US | en_US |
| dc.subject | Electrical and Computer Engineering | en_US |
| dc.title | OPTIMAL ENERGY SCHEDULING OF GRID CONNECTED MICRO GRIDS WITH BATTERY STORAGE | en_US |
| dc.type | Thesis | en_US |
