This study contributes to proposing the improved chicken swarm algorithm optimization support vector machine (ICSO-SVM) model to predict the short-term wind power. In order to improve the prediction accuracy of the model, the following work has been done. Firstly, in order to avoid the traditional chicken swarm algorithm (CSO) getting into the local optimal solution, an improved chicken swarm algorithm is proposed. The position update formulas of hens and chicks are modified in ICSO, which makes the algorithm have better global convergence and convergence accuracy. Secondly, the wind speed prediction model of ICSO-SVM is established to predict short-term wind speed in wind farms. Finally, the wind power is predicted by the power curve conversion method, which is based on the predicted wind speed data. The predict data is compared with the real data, which shows that the novel model had a higher prediction accuracy. The experimental results show that the prediction accuracy of the proposed model meets the actual engineering requirements.
[1]
William Stafford Noble,et al.
Support vector machine
,
2013
.
[2]
Zhu Changsheng.
Regional Wind Power Forecasting System for Inner Mongolia Power Grid
,
2010
.
[3]
Ming-Lang Tseng,et al.
Renewable energy utilization method: A novel Insulated Gate Bipolar Transistor switching losses prediction model
,
2018
.
[4]
Yu Liu,et al.
A New Bio-inspired Algorithm: Chicken Swarm Optimization
,
2014,
ICSI.
[5]
Abhisek Ukil,et al.
Support Vector Machine
,
2007
.
[6]
Mao Chengxiong.
Grey predictor models for wind speed-wind power prediction
,
2010
.