Department of Physics

Permanent URI for this community

https://univendspace.univen.ac.za/handle/11602/1923

Browse

Browsing Department of Physics by Author "Mulaudzi, T. S."

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    Comparison and evaluation of empirical and machine learning models in estimating global solar radiation in Limpopo province
    (2023-10-05) Murida, Thalukanyo Witney; Mulaudzi, T. S.; Maluta, N. E.; Mphephu, N.
    This study investigated the performance of machine learning techniques as compared to the empirical models to forecast the global solar radiation in Limpopo regions. The machine learning techniques used in this study are Support Vector Machines, Random Forest, and Artificial Neural Network, and the empirical models used are the Clemence and Hargreaves- Samani models. To assess the efficiences of the machine learning models against the empirical models, the researchers calculated and compared the models performance evaluation using statistical equations such as Coefficient of determination, Mean Square Error, Mean Absolute Error, and Root Mean Square Error. Calibaration was done to improve performance of the empirical models. The present study found that machine learning techniques perform better than the empirical models when estimating the global solar radiation in the selected Limpopo regions.
  • Thumbnail Image
    ItemOpen Access
    Determination of global solar radiation using temperature-based model for different climate conditions for Limpopo Province of South Africa
    (2022-07-15) Mathebe, Sampie Mphagala; Maluta, N. E.; Mulaudzi, T. S.
    The research mainly focused on the determination of global solar radiation using temperature-based model by Hargreaves and Samani for the Northern regions of Limpopo Province of South Africa. The daily maximum and minimum temperature data measured at the following six (6) stations were used: Ammondale, Mutale, Nwanedi, Roedtan, Sekgosese and Xikundu for the period 2008 – 2010. The values of empirical coefficient Kr for the Inland stations of South Africa were computed and used as an input to the model. The observed and calculated global solar radiation data were compared on the basis of the statistical error tests that is mean bias error (MBE), the mean percentage error (MPE) and the root mean square error (RMSE). Based on the statistical results the model was found suitable to estimate monthly average daily global solar radiation for the regions listed above and elsewhere with similar climatic conditions and areas where the radiation data is missing or unavailable. Hence, the study will also help to advance the state of knowledge of global solar radiation to the point where it has applications in the estimation of monthly average daily global solar radiation across.
  • Thumbnail Image
    ItemOpen Access
    Estimation of Global Solar Radiation from SAURAN stations using air temperature-based models Hargreaves and Samani and Clemence models
    (2020) Shabangu, Charlotte Beauty; Maluta, N. E.; Mulaudzi, T. S.
    Knowledge of the amount of solar radiation available in a location is important for solar energy systems, architectural designs, agronomy, and installation of pyranometers. Some developing countries do not have good quality meteorological stations that can directly measure global solar radiation. Thus, several empirical methods were developed to estimate global solar radiation. This study uses two temperature-based models which are Hargreaves - Samani and Clemence models. Four selected stations from the Southern African Universities Radiometric Network (SAURAN) for this study are University of KwaZulu–Natal, Howard college (KZH), University of Stellenbosch (SUN), Nelson Mandela University (NMU) and University of Venda (UNV). A three-year (2014-2016) temperature data for each station were sourced from SAURAN. The performance of the two models was validated using statistical analysis that is, Mean Percentage Error (MPE), Mean Bias Error (MBE), Root Mean Square (RMSE), Coefficient of Determination (R2) and t-statistical value (t). Both models obtained acceptable values of MBE, MPE, RMSE, R2 and t in KZH, NMU and UNV stations. Both models achieved the best values of MBE from 2014 to 2016, ranging from -0.0099 to 0.0147 in KZH station, followed by NMU with MBE values ranging from - 0.0293 to -0.0014, -0.0104 to 0.0330 for SUN station, 0.0241 to 0.0245 for UNV station. The models achieved MPE values between ± 10 % in all the stations. The R2 values for both models are close to 1, while the t-statistic values of one, which is less than critical value, was achieved by the models from all selected stations. This suggests that both models have got capacity to estimate global solar radiation in all the selected areas of study. However, the higher values of MBE and RMSE also revealed high level of overestimation by the models in SUN station. Therefore, this study has found evidence that both Hargreaves - Samani & Clemence models can be best recommended for estimating global solar radiation in KZH, NMU and UNV stations and areas with similar climatic and meteorological conditions.
  • Thumbnail Image
    ItemEmbargo
    Investigation of covariability between energy fluxes and CO2 exchange over a semi-arid savanna (Kruger National Park) by Eddie Covariance Technique
    (2024-09-06) Takalani, Lufuno; Mulaudzi, T. S.; Maluta, N. E.; Mateyisi, M.; Thenga, H.
    South Africa faces climate change, natural disasters, and rising temperatures due to increased levels of carbon dioxide in the atmosphere, primarily caused by deforestation, burning fossil fuels, and releasing carbon dioxide into the atmosphere without additional carbon sinks. The gap in understanding lies in studying the connection between energy flows and Net Ecosystem Exchange (NEE) at the semi-arid savanna of Kruger National Park to gain a more detailed and accurate understanding of these processes, especially in semi-arid savannas that are susceptible to changes in environmental factors. By studying energy fluxes and NEE at Kruger National Park using the eddy covariance technique, the dissertation seeks to deepen our understanding of the mechanisms driving carbon exchange in semi-arid savannas and provide insights into the impact of environmental factors on ecosystem processes. The eddy covariance technique is a powerful tool that directly measures energy and carbon dioxide exchange between the land surface and the atmosphere. The study shows that the correlation between NEE and latent heat flux (LE) and net radiation (Rn) is generally the strongest, while ground heat flux (G) and sensible heat flux (H) have little impact on NEE. The dataset provides insight into the biometeorological and flow dynamics of the Skukuza ecosystem and how it responds to climate change. The study emphasizes the importance of considering seasonality, climatic variability, and precipitation when studying the surface energy balance and its components. The findings have implications for understanding the complex interactions between ecosystem processes and environmental factors.