Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

D-CTNet: A Dual-Branch Channel-Temporal Forecasting Network with Frequency-Domain Correction

Shaoxun Wang, Xingjun Zhang, Kun Xia, Qianyang Li, Jiawei Cao, Zhendong Tan

TL;DR

D-CTNet tackles multivariate time-series forecasting in industrial collaboration by decoupling intra-channel temporal evolution from inter-variable dependencies and addressing non-stationarity with a frequency-domain correction. The method combines patch-based representation, a dual-branch temporal/channel modeling design, and a global patch fusion with a spectral alignment step, achieving state-of-the-art results on seven benchmarks and demonstrating robustness to distribution shifts. The work provides a practical forecasting engine for Digital Twin and IIoT applications, with ablations confirming the value of each component. It lays groundwork for efficient, robust industrial forecasting that can adapt to changing environments.

Abstract

Accurate Multivariate Time Series (MTS) forecasting is crucial for collaborative design of complex systems, Digital Twin building, and maintenance ahead of time. However, the collaborative industrial environment presents new challenges for MTS forecasting models: models should decouple complex inter-variable dependencies while addressing non-stationary distribution shift brought by environmental changes. To address these challenges and improve collaborative sensing reliability, we propose a Patch-Based Dual-Branch Channel-Temporal Forecasting Network (D-CTNet). Particularly, with a parallel dual-branch design incorporating linear temporal modeling layer and channel attention mechanism, our method explicitly decouples and jointly learns intra-channel temporal evolution patterns and dynamic multivariate correlations. Furthermore, a global patch attention fusion module goes beyond the local window scope to model long range dependencies. Most importantly, aiming at non-stationarity, a Frequency-Domain Stationarity Correction mechanism adaptively suppresses distribution shift impacts from environment change by spectrum alignment. Evaluations on seven benchmark datasets show that our model achieves better forecasting accuracy and robustness compared with state-of-the-art methods. Our work shows great promise as a new forecasting engine for industrial collaborative systems.

D-CTNet: A Dual-Branch Channel-Temporal Forecasting Network with Frequency-Domain Correction

TL;DR

D-CTNet tackles multivariate time-series forecasting in industrial collaboration by decoupling intra-channel temporal evolution from inter-variable dependencies and addressing non-stationarity with a frequency-domain correction. The method combines patch-based representation, a dual-branch temporal/channel modeling design, and a global patch fusion with a spectral alignment step, achieving state-of-the-art results on seven benchmarks and demonstrating robustness to distribution shifts. The work provides a practical forecasting engine for Digital Twin and IIoT applications, with ablations confirming the value of each component. It lays groundwork for efficient, robust industrial forecasting that can adapt to changing environments.

Abstract

Accurate Multivariate Time Series (MTS) forecasting is crucial for collaborative design of complex systems, Digital Twin building, and maintenance ahead of time. However, the collaborative industrial environment presents new challenges for MTS forecasting models: models should decouple complex inter-variable dependencies while addressing non-stationary distribution shift brought by environmental changes. To address these challenges and improve collaborative sensing reliability, we propose a Patch-Based Dual-Branch Channel-Temporal Forecasting Network (D-CTNet). Particularly, with a parallel dual-branch design incorporating linear temporal modeling layer and channel attention mechanism, our method explicitly decouples and jointly learns intra-channel temporal evolution patterns and dynamic multivariate correlations. Furthermore, a global patch attention fusion module goes beyond the local window scope to model long range dependencies. Most importantly, aiming at non-stationarity, a Frequency-Domain Stationarity Correction mechanism adaptively suppresses distribution shift impacts from environment change by spectrum alignment. Evaluations on seven benchmark datasets show that our model achieves better forecasting accuracy and robustness compared with state-of-the-art methods. Our work shows great promise as a new forecasting engine for industrial collaborative systems.

Paper Structure

This paper contains 19 sections, 8 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Deep Learning Models for Time Series Forecasting
  4. Non-Stationarity Adaptation and Distribution Correction
  5. Method
  6. Problem Formulation
  7. Overall Framework
  8. Patch Embedding and Representation Learning
  9. Dual-Branch Channel-Temporal Modeling
  10. Linear Temporal Modeling Branch
  11. Channel Attention Modeling Branch
  12. Global Inter-Patch Attention Fusion
  13. Frequency-Domain Stationarity Correction
  14. EXPERIMENTS
  15. Experimental Setup
  16. ...and 4 more sections

Figures (2)

  • Figure 1: Overall workflow of the proposed D-CTNet. It integrates four stages to model channel-temporal dependencies and mitigate non-stationarity. (1) Patch Embedding and Representation Learning. (2) Dual-Branch Channel-Temporal Modeling: Channel Attention and Temporal Featuring. (3) Global Inter-patch attention. (4) Stationarity Correction.
  • Figure 2: Visualization comparison of forecasting results.