Diatta Sene
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Diatta Sene has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Kadiatou Aissatou Barry
Laboratory for Teaching and Research in Energy, Automation and Telecommunications (LENAT), UGANC, Conakry, Guinea
1 shared publication - Mohamed Ansoumane Camara
Electrical Engineering Department of the Polytechnic Institute, Gamal Abdel Nasser University, Conakry, Guinea
1 shared publication - Mouhamadou Falilou Ndiaye
Water, Energy, Environment and Industrial Processes Laboratory (LE3PI), Cheikh Anta Diop University, Dakar, Senegal
1 shared publication - Souleymane Soumah
Water, Energy, Environment and Industrial Processes Laboratory (LE3PI), Cheikh Anta Diop University, Dakar, Senegal
1 shared publication - Vincent Sambou
Water, Energy, Environment and Industrial Processes Laboratory (LE3PI), Cheikh Anta Diop University, Dakar, Senegal
1 shared publication
Latest Publications
- Open Access
- 1 October 2026
- Published by GSC Online Press in GSC Advanced Research and Reviews
- Vol. 29 (1) , 343-360
INFLUENCE OF TANK SIZE ON THE TECHNO-ECONOMIC SIZING OF A PHOTOVOLTAIC WATER PUMPING SYSTEM: A HYBRID ARTIFICIAL INTELLIGENCE APPROACH APPLIED TO THE SAHEL (SAKAL, SENEGAL)
In rural Sahelian areas, access to water relies heavily on battery-less photovoltaic (PV) pumping, in which the tank acts as the storage element. Its volume, however, is most often set by autonomy rules of one to three days, which are unrelated to the installed power and to the variability of solar radiation. This paper quantifies the influence of this volume on techno-economic sizing using a hybrid chain that combines a Gradient Boosting power model, a climate-sequence generator based on typical day types and a Markov chain, a surrogate model (neural networks and Gradient Boosting), and a genetic algorithm with an uncertainty margin, with verification by simulation. The case study concerns Sakal (Louga, Senegal, Sahelian climate) and uses three days measured in 2020 at a 15-min time step. For a demand of 400 m³/day, a head of 40 m, and a tolerated deficit of 5 %, the optimum is a 225 m³ tank (0.56 day of autonomy) with 23 kWp, at 0.108 USD/m³. Autonomy rules of one, two, and three days raise the cost of water by 34, 112, and 183 %. Without a tank, the deficit exceeds 52 % regardless of the PV power. The optimal autonomy remains between 0.50 and 0.59 day across 17 sensitivity scenarios. These results depend on the assumed unit costs and on a limited set of measurements; their validation on multi-year local data is necessary.
- 1 October 2026
- Published by GSC Online Press in GSC Advanced Research and Reviews
- Vol. 29 (1) , 343-360
Open AccessINFLUENCE OF TANK SIZE ON THE TECHNO-ECONOMIC SIZING OF A PHOTOVOLTAIC WATER PUMPING SYSTEM: A HYBRID ARTIFICIAL INTELLIGENCE APPROACH APPLIED TO THE SAHEL (SAKAL, SENEGAL)
In rural Sahelian areas, access to water relies heavily on battery-less photovoltaic (PV) pumping, in which the tank acts as the storage element. Its volume, however, is most often set by autonomy rules of one to three days, which are unrelated to the installed power and to the variability of solar radiation. This paper quantifies the influence of this volume on techno-economic sizing using a hybrid chain that combines a Gradient Boosting power model, a climate-sequence generator based on typical day types and a Markov chain, a surrogate model (neural networks and Gradient Boosting), and a genetic algorithm with an uncertainty margin, with verification by simulation. The case study concerns Sakal (Louga, Senegal, Sahelian climate) and uses three days measured in 2020 at a 15-min time step. For a demand of 400 m³/day, a head of 40 m, and a tolerated deficit of 5 %, the optimum is a 225 m³ tank (0.56 day of autonomy) with 23 kWp, at 0.108 USD/m³. Autonomy rules of one, two, and three days raise the cost of water by 34, 112, and 183 %. Without a tank, the deficit exceeds 52 % regardless of the PV power. The optimal autonomy remains between 0.50 and 0.59 day across 17 sensitivity scenarios. These results depend on the assumed unit costs and on a limited set of measurements; their validation on multi-year local data is necessary.