White noise

Quantum mechanics

Algorithm

Hilbert–Huang transform

Engineering

Optimization algorithm

Telecommunications

Data mining

Computer science

Mathematical optimization

Wind speed

Physics

Nonlinear system

Mathematics

Image (mathematics)

Data pre-processing

Artificial intelligence

Wind power

Meteorology

Term (time)

Artificial neural network

Noise (video)

Machine learning

Electrical engineering

Preprocessor

Applied Energy

Short-term wind speed forecasting plays an important role in wind power generation and considerably contributes to decisions regarding control and operations. In order to improve the accuracy of wind speed forecasting, a large number of prediction methods have been proposed. However, existing prediction models ignore the role of data preprocessing and are susceptible to various limitations of the single individual model that can lead to low prediction accuracy. In this study, a developed combined model is proposed, including complete ensemble empirical mode decomposition with adaptive noise—a multi-objective grasshopper optimization algorithm based on a no-negative constraint theory—and several single models, including four neural network models and a linear model, to achieve accurate prediction results. The novel combined model considers the linear and nonlinear characteristics of the sequence, successfully overcomes the limitations of the single model, and obtains accurate and stable prediction results. In order to test the performance of combined model, the wind speed sequence of a wind farm from China is used for experiments and discussions. The results of the experiments and discussions show that the novel combined model has better forecasting performance than traditional prediction models.

A combined model based on data preprocessing strategy and multi-objective optimization algorithm for short-term wind speed forecasting