Name
Title | ||
|---|---|---|
Modelling time-series solar hot water load profile prediction using radial basis function neural network |
Abstract | ||
|---|---|---|
Power demands problem, zero reserved margin and depletion in coal production has seen the adoption of renewable energy technologies (RET) increase. Unlike conventional energy sources, RET are unpredictable and affected by non-linear, complex factors making it challenging to predict their daily/monthly supply contribution. Most available techniques seems to lack the aptitude to handle contribution complexities (e.g. ill-defined or uncertainty factors) associated with RET and its usage. This thereby results into under/over estimation or prediction of the technology demand outcomes. These factors include environmental implication, geographical location, occupant behaviour and time of use. This study proposes the use of Radial Basic Function neural network (RBFNN-based model) to improve the shortcoming that may be associated with other RET prediction technique. Ambient temperature, inlet temperature, outlet temperature, irradiance, and auxiliary contribution relatively to time and time of usage (TOU) relative to behavioural pattern are the factors considered in this investigation. The accuracy of the model was subjected to statistical measures (correlation coefficient, coefficient of determination and root mean square error). The results obtained using the investigative data and metering data showed an improved error prone capability in terms of its learning predictive skill in relation to environmental and behavioral variableness. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1109/AFRCON.2017.8095622 | 2017 IEEE AFRICON |
Keywords | Field | DocType |
RBF,RET,statistical measures,non-linear,ill-defined,prediction | Correlation coefficient,Renewable energy,Mean squared error,Load profile,Engineering,Coefficient of determination,Energy source,Statistics,Artificial neural network,Metering mode | Conference |
ISSN | Citations | PageRank |
2153-0025 | 0 | 0.34 |
References | Authors | |
2 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Mpumelelo. Mlonzi | 1 | 0 | 0.34 |
| Olawale Popoola | 2 | 0 | 2.03 |
| Josiah L. Munda | 3 | 14 | 4.47 |