首页|基于VMD-WOA混合多尺度分解的区间组合预测模型

基于VMD-WOA混合多尺度分解的区间组合预测模型

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
针对传统的点预测模型难以适用于随机性复杂系统和非线性非平稳时间序列预测的问题,提出基于VMD-WOA混合多尺度分解的区间组合预测模型.首先,引入基于鲸鱼优化(WOA)的变分模态分解(VMD)混合分解算法,得到最优区间模态子序列;其次,对各区间模态分序列使用指数平滑方法(Holt's)、支持向量回归(SVR)和BP神经网络预测,得到3 个单项预测结果,运用组合预测模型得到模态组合子序列;最后,对模态组合子序列重构,得到最终的区间组合预测序列.为了验证模型的有效性,选取AQI数据进行预测分析,实验表明所提出的基于VMD-WOA的区间组合预测方法具有较高的预测精度和良好的适应性.
Interval Combination Prediction Model Based on VMD-WOA Hybrid Multi-scale Decomposition
The traditional point prediction model is difficult to solve the prediction problem of stochastic complex system and nonlinear non-stationary time series.Therefore an interval combination prediction model based on VMD-WOA hybrid multi-scale decomposition was proposed.Firstly,the variational mode decomposition(VMD)hybrid decomposition algorithm based on whale optimization(WOA)was introduced to obtain the optimal interval modal subsequence.Secondly,Holt's(exponential smoot-hing method),support vector regression(SVR)and BP neural network were used to predict each interval modal component se-quence,and three single prediction results were obtained.The combined prediction model was used to obtain the modal combination subsequence.Finally,the final interval combination prediction sequence was obtained after the modal combination subsequence was reconstructed.In order to verify the validity of the model,AQI data was selected for prediction analysis.The experiment shows that the proposed interval combination prediction method based on VMD-WOA has high prediction accuracy and good adaptability.

hybrid multi-scale decompositionvariational mode decomposition(VMD)whale optimization algorithm(WOA)interval combination prediction modelair quality index

康晓晓、陈华友、韩冰、胡彦

展开 >

安徽大学 大数据与统计学院,安徽 合肥 230601

安徽大学 数学科学学院,安徽 合肥 230601

安徽大学 国际教育学院,安徽 合肥 230039

混合多尺度分解 变分模态分解(VMD) 鲸鱼优化(WOA) 区间组合预测 空气质量指数

国家自然科学基金项目国家自然科学基金项目

7237100171871001

2024

10.3963/j.issn.2095-3852.2024.03.017

武汉理工大学学报(信息与管理工程版)
武汉理工大学

武汉理工大学学报(信息与管理工程版)

CSTPCD
影响因子：0.37
ISSN：2095-3852
年,卷(期)：2024.46(3)