ABSTRACT
Most practices currently adopted in modeling the presence of people in residential houses do not reflect the complexity of the impact occupants have on lighting loads. Hence the needs to contribute a methodology, that will be able to look into such characteristics. This study entailed the use of adaptive Neuro Fuzzy inference system (ANFIS) for estimation and prediction of lighting load usage profile for energy and demand side management initiatives. Results obtained showed a better correlation analysis and root mean square error (RMSE) in contrast to the regression method model. This shows that ANFIS model has good prediction accuracy capability.
Funding
This work was supported in part by Eskom, South Africa, Centre for Energy and Electric Power, Tshwane University of Technology and Department of Higher Education and Training Research Department Grant (DHET- RDG), South Africa.