摘要
针对含有时间趋势性和振荡性的小样本序列的建模预测问题,提出了同时包含时间延迟参数、时间作用参数及多项式的振荡型GM(1,1,k)幂模型,采用遗传算法寻找模型中的最优非线性参数,提高了模型的预测精度。将该模型应用于我国季度风电产量的预测,结果显示,振荡型GM(1,1,k)幂模型具有良好的预测性能,为含有时间趋势性的小样本振荡序列预测提供了一种有效的建模方法和预测手段。
To solve the prediction problem for small sample sequence with time trend and oscillation characteristics,an oscillating GM(1,1,k)power model with system latency,time-varying parameters and polynomial is proposed.The genetic algorithm is used to find the optimal nonlinear parameters of the model,which improves the prediction accuracy of the model.Finally,the proposed model is applied to the forecast China's quarterly wind power output.The results show that the oscillating GM(1,l,k)power model has good predictive performance and provides an effective modeling method and forecasting means for the forecast of oscillation sequence with time trend characteristic.
作者
白雪
任洪涛
刘冬玉
刘盼
李晔
BAI Xue;REN Hong-tao;LIU Dong-yu;LIU Pan;LI Ye(College of Information and Management Science,Henan Agricultural University,Zhengzhou 450046,China)
出处
《系统工程》
CSSCI
CSCD
北大核心
2023年第4期137-144,共8页
Systems Engineering
基金
河南省软科学研究计划项目(222400410391)
教育部人文社科青年研究项目(21YJC790076)
河南省高等学校哲学社会科学创新人才支持计划项目(CXRC-2023-24)。
