A new hybrid modeling approach improves short-term wind speed prediction accuracy by combining advanced noise reduction techniques with ensemble stochastic configuration networks.
Wind speed measurements collected by sensors are often contaminated by interference from extreme weather events and equipment malfunctions. To solve this, the study introduces a noise reduction method known as Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise-Improved Wavelet Threshold Denoising (ICEEMDAN-IWTD).
The study design incorporates a multi-strategy improved coati optimization algorithm (ICOA) to refine the parameters of the ICEEMDAN process. This optimization allows the system to better decompose noisy information into distinct time-frequency scales, which is critical for distinguishing and isolating specific noise components within the raw sensor data.
By proposing a new threshold mechanism, the team avoided the common deviations associated with traditional threshold functions, allowing for a more accurate reconstruction of the underlying wind speed signals.
Following the noise reduction phase, the researchers used ensemble Stochastic Configuration Networks (ESCNs) to forecast the wind speed series. The team applied the ICOA once more to optimize the weight coefficients of this integrated model, ensuring that the prediction process maintains a high level of precision.
The effectiveness of this hybrid approach was validated using two real wind speed detection datasets collected in Chicago.
Subject of Research: Computational intelligence in wind energy forecasting
Article Title: Ensemble stochastic configuration networks modeling method for short-term wind speed prediction and noise reduction
Article References: Yu, Y., Zhou, Y., Li, K., & Han, Y. (2026). Ensemble stochastic configuration networks modeling method for short-term wind speed prediction and noise reduction. Industrial Artificial Intelligence. https://doi.org/10.1007/s44244-026-00035-7
Image Credits: AI Generated
DOI: 10.1007/s44244-026-00035-7
Keywords: Wind Energy, Wind Speed Prediction, Artificial Intelligence, Stochastic Configuration Networks, Noise Reduction, Signal Processing, Data Analysis, Renewable Energy, Optimization Algorithms, Coati Optimization Algorithm, Predictive Modeling, Time-Series Analysis, Forecasting Accuracy, Industrial Intelligence, Sensor Data
Cite Scienmag News
Faith Mcneil. (November 17, 2026). Improved Ensemble Method Enhances Short-Term Wind Speed Prediction Accuracy. Scienmag. https://scienmag.com/improved-ensemble-method-enhances-short-term-wind-speed-prediction-accuracy/
Faith Mcneil. "Improved Ensemble Method Enhances Short-Term Wind Speed Prediction Accuracy." Scienmag, 17 November 2026, https://scienmag.com/improved-ensemble-method-enhances-short-term-wind-speed-prediction-accuracy/. Accessed 17 November 2026.
Faith Mcneil. "Improved Ensemble Method Enhances Short-Term Wind Speed Prediction Accuracy." Scienmag. November 17, 2026. https://scienmag.com/improved-ensemble-method-enhances-short-term-wind-speed-prediction-accuracy/








