Publication: Application of Extreme Learning Machine in Predicting Short- Term Wind Speed
No Thumbnail Available
Date
2020
Authors
Chen C.P.
Tiong S.K.
Koh S.P.
Nasser A.A.
Abbas D.
Chooi F.Y.
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers Inc.
Abstract
At present, wind energy is the fastest growing power generation sector with its economic advantage of being a rich, clean and environmentally sustainable resources. However, wind does not generally blow consistently which prevents wind turbines from functioning at maximum capacity and capability. However, the solution that had been put forward in this paper to overcome the aforementioned problem is able to be used to make the deterministic predictions study for the wind speed, and of which its model in this paper possesses a significant prediction accuracy and strong stability, which could be useful in predicting the randomness of short term wind speed accurately. Based on the prediction output results, the amount of power required to be generated for load dispatch planning could be calculated and used to produce a scientific basis with the purpose of designing an optimal power grid dispatching design. In this paper, Extreme Learning Machine (ELM) is used for predicting short-term wind speed, and through use of the ELM the prediction accuracy of wind speed was observed to be at 0.93 followed by the root mean square error rate at 1.9. With reference to the prediction results, the developed model is tested to be able to predict wind speed accurately. � 2020 IEEE.
Description
Electric load dispatching; Electric power transmission networks; Forecasting; Knowledge acquisition; Machine learning; Mean square error; Speed; Wind power; Developed model; Economic advantages; Extreme learning machine; Prediction accuracy; Root mean square errors; Scientific basis; Strong stability; Sustainable resources; Wind