Short term wind power forecasting in South Africa using neural networks
| dc.contributor.advisor | Sigauke, Caston | |
| dc.contributor.advisor | Chibaya, Colin | |
| dc.contributor.advisor | Mbuvha, Rendani | |
| dc.contributor.author | Daniel, Lucky Oghenechodja | |
| dc.date | 2020 | |
| dc.date.accessioned | 2020-10-05T07:42:02Z | |
| dc.date.available | 2020-10-05T07:42:02Z | |
| dc.date.issued | 2020-08-11 | |
| dc.description | MSc (Statistics) | en_ZA |
| dc.description | Department of Statistics | |
| dc.description.abstract | Wind offers an environmentally sustainable energy resource that has seen increasing global adoption in recent years. However, its intermittent, unstable and stochastic nature hampers its representation among other renewable energy sources. This work addresses the forecasting of wind speed, a primary input needed for wind energy generation, using data obtained from the South African Wind Atlas Project. Forecasting is carried out on a two days ahead time horizon. We investigate the predictive performance of artificial neural networks (ANN) trained with Bayesian regularisation, decision trees based stochastic gradient boosting (SGB) and generalised additive models (GAMs). The results of the comparative analysis suggest that ANN displays superior predictive performance based on root mean square error (RMSE). In contrast, SGB shows outperformance in terms of mean average error (MAE) and the related mean average percentage error (MAPE). A further comparison of two forecast combination methods involving the linear and additive quantile regression averaging show the latter forecast combination method as yielding lower prediction accuracy. The additive quantile regression averaging based prediction intervals also show outperformance in terms of validity, reliability, quality and accuracy. Interval combination methods show the median method as better than its pure average counterpart. Point forecasts combination and interval forecasting methods are found to improve forecast performance. | en_ZA |
| dc.description.sponsorship | NRF | en_ZA |
| dc.format.extent | i online resource (xv, 95 leaves : color illustrations) | |
| dc.identifier.apacitation | Daniel, L. O. (2020). <i>Short term wind power forecasting in South Africa using neural networks</i>. (). . Retrieved from http://hdl.handle.net/11602/1591 | en_ZA |
| dc.identifier.chicagocitation | Daniel, Lucky Oghenechodja. <i>"Short term wind power forecasting in South Africa using neural networks."</i> ., , 2020. http://hdl.handle.net/11602/1591 | en_ZA |
| dc.identifier.citation | Daniel, Lucky Oghenechodja (2020) Short term wind power forecasting in South Africa using neural networks. University of Venda, South Africa.<http://hdl.handle.net/11602/1591>. | |
| dc.identifier.ris | TY - Dissertation AU - Daniel, Lucky Oghenechodja AB - Wind offers an environmentally sustainable energy resource that has seen increasing global adoption in recent years. However, its intermittent, unstable and stochastic nature hampers its representation among other renewable energy sources. This work addresses the forecasting of wind speed, a primary input needed for wind energy generation, using data obtained from the South African Wind Atlas Project. Forecasting is carried out on a two days ahead time horizon. We investigate the predictive performance of artificial neural networks (ANN) trained with Bayesian regularisation, decision trees based stochastic gradient boosting (SGB) and generalised additive models (GAMs). The results of the comparative analysis suggest that ANN displays superior predictive performance based on root mean square error (RMSE). In contrast, SGB shows outperformance in terms of mean average error (MAE) and the related mean average percentage error (MAPE). A further comparison of two forecast combination methods involving the linear and additive quantile regression averaging show the latter forecast combination method as yielding lower prediction accuracy. The additive quantile regression averaging based prediction intervals also show outperformance in terms of validity, reliability, quality and accuracy. Interval combination methods show the median method as better than its pure average counterpart. Point forecasts combination and interval forecasting methods are found to improve forecast performance. DA - 2020-08-11 DB - ResearchSpace DP - Univen KW - Additive quantile regression averaging KW - Forecasts combination KW - Machine learning KW - Point and interval forecasting KW - Renewable energy KW - Wind energy LK - https://univendspace.univen.ac.za PY - 2020 T1 - Short term wind power forecasting in South Africa using neural networks TI - Short term wind power forecasting in South Africa using neural networks UR - http://hdl.handle.net/11602/1591 ER - | en_ZA |
| dc.identifier.uri | http://hdl.handle.net/11602/1591 | |
| dc.identifier.vancouvercitation | Daniel LO. Short term wind power forecasting in South Africa using neural networks. []. , 2020 [cited yyyy month dd]. Available from: http://hdl.handle.net/11602/1591 | en_ZA |
| dc.language.iso | en | en_ZA |
| dc.rights | University of Venda | |
| dc.subject | Additive quantile regression averaging | en_ZA |
| dc.subject | UCTD | en_ZA |
| dc.subject | Machine learning | en_ZA |
| dc.subject | Point and interval forecasting | en_ZA |
| dc.subject | Renewable energy | en_ZA |
| dc.subject | Wind energy | en_ZA |
| dc.subject.ddc | 621.3121360968 | |
| dc.subject.lcsh | Wind power -- South Africa | |
| dc.subject.lcsh | Power resources -- South Africa | |
| dc.subject.lcsh | Renewable energy resources -- South Africa | |
| dc.subject.lcsh | Wind energy conversion systems -- South Africa | |
| dc.subject.lcsh | Wind forecasting -- South Africa | |
| dc.subject.lcsh | Weather -- Forecasting | |
| dc.title | Short term wind power forecasting in South Africa using neural networks | en_ZA |
| dc.type | Dissertation | en_ZA |