Simultaneous Feeder Routing and Conductor Selection in Rural Distribution Networks Using an Exact MINLP Approach DOI Creative Commons
Brandon Cortés-Caicedo, Oscar Danilo Montoya, Luis Fernando Grisales-Noreña

et al.

Smart Cities, Journal Year: 2025, Volume and Issue: 8(2), P. 68 - 68

Published: April 15, 2025

This article addresses the optimal network expansion problem in rural distribution systems using a mixed-integer nonlinear programming (MINLP) model that simultaneously performs route selection and conductor sizing radial systems. The proposed methodology was validated on 9- 25-node test systems, comparing results against approaches based minimum spanning tree (MST) formulation metaheuristic (the sine-cosine tabu search algorithms). MINLP significantly reduced total costs. For nine-node system, cost decreased from USD 131,819.33 (MST-TSA) to 77,129.34 (MINLP), saving 54,689.99 (41.48%). Similarly, costs of energy losses dropped 111,746.73 63,764.12, reduction 47,982.61 (42.94%). In fell by over 65% 371,516.59 128,974.72, while 210,057.16 (61.06%). Despite requiring higher initial investment conductors, led substantial long-term savings due operating Unlike previous methods which separate topology design sizing, our proposal integrates both aspects, ensuring globally solutions. demonstrate its scalability effectiveness for planning complex power networks. experimental implementation carried out Julia (v1.10.2) JuMP (v1.21.1) BONMIN.

Language: Английский

Deep Learning vs. Gradient Boosting: Optimizing Transport Energy Forecasts in Thailand Through LSTM and XGBoost DOI Creative Commons
Thanapong Champahom,

Chinnakrit Banyong,

Thananya Janhuaton

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(7), P. 1685 - 1685

Published: March 27, 2025

Thailand’s transport sector faces critical challenges in energy management amid rapid economic growth, with accounting for approximately 30% of total consumption. This study addresses research gaps forecasting by comparing Long Short-Term Memory (LSTM) neural networks and XGBoost models predicting consumption Thailand. Utilizing a comprehensive dataset spanning 1993–2022 that includes vehicle registration data size category, kilometers traveled, macroeconomic indicators, this evaluates both modeling approaches through multiple performance metrics. The results demonstrate consistently outperforms LSTM, achieving an R-squared value 0.9508 test compared to LSTM’s 0.2005. Feature importance analysis reveals medium vehicles contribute 36.6% predictions, followed truck VKT (20.5%), demographic factors combined 15.2%. contributes methodological understanding practical application establishing XGBoost’s superior forecasting, quantifying the differential impact various categories on consumption, demonstrating integrating usage predictive models. findings provide evidence-based guidance prioritizing policy interventions enhance efficiency sustainability.

Language: Английский

Citations

0
Simultaneous Feeder Routing and Conductor Selection in Rural Distribution Networks Using an Exact MINLP Approach DOI Creative Commons
Brandon Cortés-Caicedo, Oscar Danilo Montoya, Luis Fernando Grisales-Noreña

et al.

Smart Cities, Journal Year: 2025, Volume and Issue: 8(2), P. 68 - 68

Published: April 15, 2025

This article addresses the optimal network expansion problem in rural distribution systems using a mixed-integer nonlinear programming (MINLP) model that simultaneously performs route selection and conductor sizing radial systems. The proposed methodology was validated on 9- 25-node test systems, comparing results against approaches based minimum spanning tree (MST) formulation metaheuristic (the sine-cosine tabu search algorithms). MINLP significantly reduced total costs. For nine-node system, cost decreased from USD 131,819.33 (MST-TSA) to 77,129.34 (MINLP), saving 54,689.99 (41.48%). Similarly, costs of energy losses dropped 111,746.73 63,764.12, reduction 47,982.61 (42.94%). In fell by over 65% 371,516.59 128,974.72, while 210,057.16 (61.06%). Despite requiring higher initial investment conductors, led substantial long-term savings due operating Unlike previous methods which separate topology design sizing, our proposal integrates both aspects, ensuring globally solutions. demonstrate its scalability effectiveness for planning complex power networks. experimental implementation carried out Julia (v1.10.2) JuMP (v1.21.1) BONMIN.

Language: Английский

Citations

0