基于灰色关联分析和黑洞粒子群优化算法的短期风电功率预测

Short Term Wind Power Prediction Based on Grey Relational Analysis and Black Hole Particle Swarm Optimization

短期风电功率的预测是保障风电场持续稳定运行以及电网调度的重要因素。选择最小二乘支持向量机(LSSVM)作为预测模型,使用灰色关联分析法对影响风电功率的因素进行权重比较,并使用黑洞粒子群算法(BHPSO)对LSSVM的回归性能参数进行优化,建立了基于灰色关联分析和BHPSO的LSSVM短期风电功率预测模型。对山东某风电场提供的数据进行仿真研究,并与LSSVM模型和BP神经网络模型进行对比分析。验证结果表明,基于灰色关联分析和BHPSO的LSSVM模型的预测效果最好。

Short term wind power forecasting is an important factor of ensuring continuous and stable operation of wind farm and power grid dispatching. Least squares support vector machines（ LSSVM） was selected as the predictive model. The method of grey relational analysis was used to weight the influence factors of wind power. And the black hole particle swarm optimization（ BHPSO） algorithm was used to optimize the regression performance parameters of LSSVM. The LSSVM short term wind power prediction model was established based on grey relational analysis and BHPSO algorithm. The data provided by a wind farm in Shandong was used to conduct simulation study. And the results were compared with LSSVM model and BP neural network model. The validation results show that the LSSVM predictive model based on grey relational analysis and BHPSO algorithm is better.