Monthly average daily global solar radiation data are essential for the design and study of solar energy systems. The performance and accuracy of eleven models for the estimation of monthly average global solar radiation were compared in this study. Nineteen months (Nov 2020 – May 2022) ground measurement data consisting of monthly mean daily sunshine duration, relative humidity, minimum and maximum temperatures, and global solar radiation collected from the Lawra Solar Plant were used. The models were compared using statistical indices. According to the indices, most of the models were in reasonably good agreement with the measured data. Two model equations, however, were found to have the highest accuracy and can thus be used to estimate monthly average global solar radiation in Lawra and other places with similar climatic conditions where radiation data is unavailable.

In this study, simple and multiple regression models were developed to estimate the monthly average daily global solar radiation in Lawra, Ghana using ground measurement of global horizontal irradiance (Nov 2020–May 2022) and typical meteorological year (TMY) data (Jan 2017–Dec 2019). Various predictor variables such as sunshine ratio, minimum relative humidity and maximum relative humidity ratio, minimum and maximum temperature ratio, etc. were correlated from the TMY data. Many model equations were developed with the variables ranging from one to eight. The best model from each category was chosen and compared using statistical indices to determine the overall best model. We used the JMP statistical software’s ‘All Possible Models’ functionality to select the best model from each category. The selected models where then compared using the adjusted R-squared, mean absolute percentage error, and the root mean square error statistical indices. The best model equation correlated with eight independent variables with adjusted R-squared of 0.99. The equation can be used to estimate monthly global solar radiation in Lawra and in locations with similar climatic conditions where ground measurement of radiation data is unavailable but have access to the National Solar Radiation Database’s (NSRDB) TMY data.

We investigated the predictive power of the NBEATS deep learning architecture in forecasting measured Global Horizontal Irradiance (GHI) 3 days into the future in this study.We used the NBEATS neural network model from the "NeuralForecast" Python library.Mean daily GHI data from the Lawra Solar Power Plant in the Upper West Region of Ghana was used.The NBEATS architecture could predict 3 days ahead GHI accurately, however, to accurately forecast solar irradiance more days into the future, a lot of data is required.The MAPE of the model was found to be 1.4%.We also explained GHI trend in the study site using various data visualization charts.

Data on monthly average daily global solar radiation are critical in the design and analysis of solar energy systems.Using data from the Lawra Solar Plant in Ghana's Upper West Region, a multiple regression model was developed to estimate the monthly average daily global solar radiation using the Angstrom-Garcia model.The parameters used were sunshine duration ratio, average maximum possible daily hours of sunshine and the difference between maximum and minimum temperature values.The model equation is given as ̅ ̅ 0 ⁄ = 0.15 + 0.426 ̅ ̅ ⁄ + 0.1392 ∆ ̅ ⁄ , which was found to be reliable; thus the equation can be employed for estimating global solar radiation of locations that have similar climate, latitude and altitude as Lawra.

Over 600 million people living in sub-Saharan Africa do not have access to electricity. Modern healthcare services, including vaccine refrigeration, which require electricity are therefore lacking in such energy-deprived communities. In this work, analysis has been conducted on how electricity access can help improve healthcare service delivery and rural development, with a case study on 3 different off-grid solar photovoltaic (PV) systems in community-based health planning and services (CHPS) in Ghana. Analysis from this study showed that for the 3.0 kWp solar PV systems installed at the various sites, the in-house electricity consumptions are between 4.30 and 7.58 kWh per day. It was found out that excess electricity generation of 148–304 kWh per month is available and can be used to provide other economic services including phone charging, torchlight battery charging, and small-sized cold storage services to generate income for the maintenance of the systems, which is critical for sustainability of solar PV installations in rural poor communities. The study results also showed that electrified health facilities which are able to provide basic healthcare services have potential impact on community health outcomes and rural development. Assessment conducted at the CHPS compounds revealed that, generally, there is improvement in healthcare service delivery resulting in time savings of 15-43 hours per month for the inhabitants which can potentially be used for productive work. The time savings were more significant in females and children than in males. In many rural agro-based communities in developing countries, female and children are usually the workforce engaged in various farming activities. This paper concludes that access to electricity in CHPS compounds helps to improve community health outcomes and increases time availability for women to engage in productive work that can potentially result in significant socioeconomic activities and rural development.