Cloud Clustering Analysis and the Design of A 1mw Off-Grid Solar PV Power Plant at Molyko-Buea
Fonge Emmanuel Nchindia1, Tanyi Emmanuel Beti2
1Fonge Emmanuel Nchindia, Department of Electrical Electronics Engineering, University of Buea, Cameroon.
2Prof. Tanyi Emmanuel Beti, Department of Electrical Electronics Engineering, University of Buea, Cameroon.
Manuscript received on 20 March 2025 | First Revised Manuscript received on 27 March 2025 | Second Revised Manuscript received on 21 July 2025 | Manuscript Accepted on 15 August 2025 | Manuscript published on 30 August 2025 | PP: 13-27 | Volume-14 Issue-6, August 2025 | Retrieval Number: 100.1/ijeat.E463614050625 | DOI: 10.35940/ijeat.E4636.14060825
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Molyko is the most densely populated neighbourhood in Buea. The rise in population, in the absence of proper planning and contingency measures, has led to a poor and unreliable power supply to the inhabitants of Molyko. This project aims to provide access to clean, reliable, and affordable electricity to the inhabitants of Molyko, which will significantly improve living standards, energy equity, and security. Atmospheric variability affects solar PV power systems and the power grid to which they are connected. Since we currently have very little control over atmospheric phenomena, a mitigating measure to offset resource variability is solar forecasting. In this study, a novel ultra-shortterm PV power forecasting method based on satellite image data is proposed, which combines the spatio-temporal correlation between multiple plants with power and cloud information. Results show that the proposed method outperforms the benchmark methods, achieving higher accuracy of 4.42% and 11.44% for the two PV plant targets on a four-month validation dataset, in terms of root mean square error and mean absolute error, respectively. Based on a careful feasibility study of the site and a prudent design of the 1MW off-grid solar PV generation system at Molyko. With a solar irradiance of 5.22 kWh/m²/d and 4368 polycrystalline silicon modules, each with a peak power of 240W, we can generate 1,443,070 kWh in a year when the modules are inclined at 50 degrees towards the South. This project has a Net Present cost of 800 MFCFA, a levelized cost of energy (LCOE) of 52 CFA/kWh, and an efficiency of 81%. This energy can meet the needs of over 2000 homes in a year. A power plant model validation technique using MATLAB/Simulink was employed to verify that this design is reliable and can predictably generate energy from the system.
Keywords: Cloud Fraction, Net flux, Levelized Cost, Energy Yield Prediction.
Scope of the Article: Energy Harvesting