Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications
This study presents the potential of harvesting wind energy in Sarawak, Malaysia based on the ground station and prediction models. A topographical feedforward neural network (T-FFNN) is proposed as an alternative to predict the wind speed in the areas where wind speed measurements are not done. The...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier Ltd
2017
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/14963/ http://ir.unimas.my/id/eprint/14963/1/Wind-power-generation-via-ground-wind-station-and-topographical-feedforward-neural-network-%28T-FFNN%29-model-for-small-scale-applications_2017_Journal-of-Cleaner-Production.html |
| _version_ | 1848837758519345152 |
|---|---|
| author | Salisu Muhammad, Lawan Wan Azlan, Wan Zaina Abidinl Thelaha, Bin Hj Masri Chai, Wangyin Baharun, Azhaili |
| author_facet | Salisu Muhammad, Lawan Wan Azlan, Wan Zaina Abidinl Thelaha, Bin Hj Masri Chai, Wangyin Baharun, Azhaili |
| author_sort | Salisu Muhammad, Lawan |
| building | UNIMAS Institutional Repository |
| collection | Online Access |
| description | This study presents the potential of harvesting wind energy in Sarawak, Malaysia based on the ground station and prediction models. A topographical feedforward neural network (T-FFNN) is proposed as an alternative to predict the wind speed in the areas where wind speed measurements are not done. The model has nine meteorological, geographical and topographical parameters as inputs while monthly winds speed as an output variable. The suitability of the model was assessed based on the mean absolute percentage error (MAPE). The most effective network design with lowest MAPE of 3.4% and correlation R between the predicted and the ground station wind speed of 0.91 was obtained. The study shows the characteristics of wind speed at 10–40 m heights. For the wind speed distribution, in addition to the widely applied Weibull and Rayleigh models, Gamma, Erlang and Lognormal are included. It was found that Gamma and Weibull outperform the others based on the three goodness-of-fit (GOF). An assessment of wind energy potential was performed using the measured and predicted wind speed data. The outcomes show that wind power density falls within class 1 (PD≤100 W/m2). Final results from micro-sitting investigating the performance of annual energy output (AEO) in the examined area are presented. The results indicate that the AEO differs with altitudes. In all the examined areas, the AEO values varied from about 5800–13,622 kWh/year. These results show the possibility of using wind energy for small-scale purpose. |
| first_indexed | 2025-11-15T06:44:45Z |
| format | Article |
| id | unimas-14963 |
| institution | Universiti Malaysia Sarawak |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T06:44:45Z |
| publishDate | 2017 |
| publisher | Elsevier Ltd |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | unimas-149632017-04-12T02:05:37Z http://ir.unimas.my/id/eprint/14963/ Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications Salisu Muhammad, Lawan Wan Azlan, Wan Zaina Abidinl Thelaha, Bin Hj Masri Chai, Wangyin Baharun, Azhaili TA Engineering (General). Civil engineering (General) This study presents the potential of harvesting wind energy in Sarawak, Malaysia based on the ground station and prediction models. A topographical feedforward neural network (T-FFNN) is proposed as an alternative to predict the wind speed in the areas where wind speed measurements are not done. The model has nine meteorological, geographical and topographical parameters as inputs while monthly winds speed as an output variable. The suitability of the model was assessed based on the mean absolute percentage error (MAPE). The most effective network design with lowest MAPE of 3.4% and correlation R between the predicted and the ground station wind speed of 0.91 was obtained. The study shows the characteristics of wind speed at 10–40 m heights. For the wind speed distribution, in addition to the widely applied Weibull and Rayleigh models, Gamma, Erlang and Lognormal are included. It was found that Gamma and Weibull outperform the others based on the three goodness-of-fit (GOF). An assessment of wind energy potential was performed using the measured and predicted wind speed data. The outcomes show that wind power density falls within class 1 (PD≤100 W/m2). Final results from micro-sitting investigating the performance of annual energy output (AEO) in the examined area are presented. The results indicate that the AEO differs with altitudes. In all the examined areas, the AEO values varied from about 5800–13,622 kWh/year. These results show the possibility of using wind energy for small-scale purpose. Elsevier Ltd 2017-02-01 Article PeerReviewed text en http://ir.unimas.my/id/eprint/14963/1/Wind-power-generation-via-ground-wind-station-and-topographical-feedforward-neural-network-%28T-FFNN%29-model-for-small-scale-applications_2017_Journal-of-Cleaner-Production.html Salisu Muhammad, Lawan and Wan Azlan, Wan Zaina Abidinl and Thelaha, Bin Hj Masri and Chai, Wangyin and Baharun, Azhaili (2017) Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications. Journal of Cleaner Production, 143 (1). pp. 1246-1259. ISSN 09596526 http://www.sciencedirect.com/science/article/pii/S0959652616320194 DOI: 10.1016/j.jclepro.2016.11.157 |
| spellingShingle | TA Engineering (General). Civil engineering (General) Salisu Muhammad, Lawan Wan Azlan, Wan Zaina Abidinl Thelaha, Bin Hj Masri Chai, Wangyin Baharun, Azhaili Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications |
| title | Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications |
| title_full | Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications |
| title_fullStr | Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications |
| title_full_unstemmed | Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications |
| title_short | Wind power generation via ground wind station and topographical feedforward neural network (T-FFNN) model for small-scale applications |
| title_sort | wind power generation via ground wind station and topographical feedforward neural network (t-ffnn) model for small-scale applications |
| topic | TA Engineering (General). Civil engineering (General) |
| url | http://ir.unimas.my/id/eprint/14963/ http://ir.unimas.my/id/eprint/14963/ http://ir.unimas.my/id/eprint/14963/ http://ir.unimas.my/id/eprint/14963/1/Wind-power-generation-via-ground-wind-station-and-topographical-feedforward-neural-network-%28T-FFNN%29-model-for-small-scale-applications_2017_Journal-of-Cleaner-Production.html |