A Novel Multi-Scale and Adaptive Multi-Period Deep Learning with Compression-Fusion Attention for Cold Storage Load Prediction

Cai, Hao; Zhang, Yi; Zhang, Jinhong; Chen, Jie; Liu, Jiafu; Xu, Jingxuan

doi:10.3390/electronics15010160

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

A Novel Multi-Scale and Adaptive Multi-Period Deep Learning with Compression-Fusion Attention for Cold Storage Load Prediction

by

Hao Cai

^1,*

,

Yi Zhang

¹

,

Jinhong Zhang

¹,

Jie Chen

²,

Jiafu Liu

¹

and

Jingxuan Xu

¹

College of Mathematics and Computer Science, Shantou University, Shantou 515063, China

²

College of Law, Shantou University, Shantou 515063, China

^*

Author to whom correspondence should be addressed.

Electronics 2026, 15(1), 160; https://doi.org/10.3390/electronics15010160 (registering DOI)

Submission received: 9 December 2025 / Revised: 25 December 2025 / Accepted: 27 December 2025 / Published: 29 December 2025

(This article belongs to the Special Issue Applications of Machine Learning and Artificial Intelligence in Modern Power and Energy Systems, 2nd Edition)

Download Versions Notes

Abstract

Accurate load forecasting is essential for energy-efficient scheduling in cold storage facilities, where cooling demand is shaped by strong periodicity, nonlinear temporal dynamics, and irregular operational disturbances. Traditional statistical and machine-learning models struggle with these multi-scale variations, and existing deep learning approaches often rely on fixed receptive fields or fail to extract adaptive periodic structures. This study introduces MA-CFAN, a multi-scale and adaptive multi-period forecasting framework that integrates temporal decomposition, dynamic frequency-period identification, and a newly designed Compression-Fusion Attention Block (CFABlock) for cross-period representation learning. The architecture leverages FFT-derived adaptive periods to capture seasonal-trend components and employs compression-fusion attention to enhance feature discrimination across temporal scales. Furthermore, this work provides the first systematic evaluation of state-of-the-art forecasting models, including Informer, Autoformer, iTransformer, TimesNet, DLinear, and TimeMixer, to the domain of cold storage load prediction. Experiments on real operational data from a logistics center in Jinan, China, demonstrate that MA-CFAN consistently outperforms all baselines, reducing average MSE and MAE by up to 19.3% and 14.8%, respectively.

Keywords: artificial intelligence; load forecasting; time series forecasting; cold storage; multi-scale; neural networks

Share and Cite

MDPI and ACS Style

Cai, H.; Zhang, Y.; Zhang, J.; Chen, J.; Liu, J.; Xu, J. A Novel Multi-Scale and Adaptive Multi-Period Deep Learning with Compression-Fusion Attention for Cold Storage Load Prediction. Electronics 2026, 15, 160. https://doi.org/10.3390/electronics15010160

AMA Style

Cai H, Zhang Y, Zhang J, Chen J, Liu J, Xu J. A Novel Multi-Scale and Adaptive Multi-Period Deep Learning with Compression-Fusion Attention for Cold Storage Load Prediction. Electronics. 2026; 15(1):160. https://doi.org/10.3390/electronics15010160

Chicago/Turabian Style

Cai, Hao, Yi Zhang, Jinhong Zhang, Jie Chen, Jiafu Liu, and Jingxuan Xu. 2026. "A Novel Multi-Scale and Adaptive Multi-Period Deep Learning with Compression-Fusion Attention for Cold Storage Load Prediction" Electronics 15, no. 1: 160. https://doi.org/10.3390/electronics15010160

APA Style

Cai, H., Zhang, Y., Zhang, J., Chen, J., Liu, J., & Xu, J. (2026). A Novel Multi-Scale and Adaptive Multi-Period Deep Learning with Compression-Fusion Attention for Cold Storage Load Prediction. Electronics, 15(1), 160. https://doi.org/10.3390/electronics15010160

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

A Novel Multi-Scale and Adaptive Multi-Period Deep Learning with Compression-Fusion Attention for Cold Storage Load Prediction

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI