Performance Evaluation of a Radial Distribution Network Under Emerging Load Prediction Modeling Approach and DG Integration Using a Particle Swarm Optimization Algorithm
This research presents a comprehensive performance evaluation of an 11-bus, 15 kV radial distribution network in Ethiopia, utilizing particle swarm optimization (PSO) to assess the impact of emerging load prediction models and distributed generation (DG) integration. Load forecasting is conducted us...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-01-01
|
| Series: | Journal of Electrical and Computer Engineering |
| Online Access: | http://dx.doi.org/10.1155/jece/1352068 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This research presents a comprehensive performance evaluation of an 11-bus, 15 kV radial distribution network in Ethiopia, utilizing particle swarm optimization (PSO) to assess the impact of emerging load prediction models and distributed generation (DG) integration. Load forecasting is conducted using the adaptive neuro-fuzzy inference system (ANFIS), with validation carried out through an artificial neural network (ANN). The average forecasted load predicted by ANFIS is 6,071.5 kVA, compared to 6,105.7 kVA by ANN. The accuracy of these forecasts is quantified by mean absolute error (MAE), mean absolute percentage error (MAPE), mean squared error (MSE), root mean squared error (RMSE), and the coefficient of determination (R2), where ANFIS demonstrates superior performance with a MAE = 7.7611, MAPE = 0.14401, MSE = 0.6399, RMSE = 0.79993, and R2 = 0.99993, in contrast to ANN’s MAE = 31.4114, MAPE = 1.631%, MSE = 109.55, RMSE = 10.467, and R2 = 0.98797. The study further examines the network’s operational efficiency in terms of power loss, voltage stability index (VSI), average voltage deviation index (AVDI), loss of load probability (LOLP), energy not supplied (ENS), and average energy not supplied (AENS). These performance metrics are evaluated under various load conditions, including base load and forecasted loads derived from both ANN and ANFIS predictions, incorporating DG integration. The results highlight that the PSO algorithm excels in optimizing network performance, achieving remarkable results across all evaluated parameters. Despite these promising findings, the study has certain limitations. The proposed model assumes ideal DG operation without considering uncertainties in renewable energy sources such as solar and wind power variations. Additionally, the impact of network reconfiguration and real-time control strategies for dynamic load variations is not fully explored. The computational complexity of integrating ANFIS-based forecasting with large-scale networks poses a challenge, requiring further optimization for practical applications. Future research should address these challenges by incorporating probabilistic models for DG output fluctuations, real-time network reconfiguration techniques, and hybrid optimization approaches for enhanced scalability and adaptability. |
|---|---|
| ISSN: | 2090-0155 |