Hierarchical bandwidth-adaptive variational mode decomposition: algorithms and applications

Area of Science:

• Signal Processing
• Machine Learning
• Mechanical Engineering

Background:

• Nonlinear and non-stationary signal analysis faces challenges like mode mixing and limited parameter adaptivity.
• Existing methods often require predefining the number of intrinsic modes, limiting flexibility.
• Accurate fault diagnosis in industrial machinery is crucial for reliability and safety.

Purpose of the Study:

• To introduce a novel hierarchical bandwidth-adaptive variational mode decomposition (HBVMD) method.
• To overcome limitations of traditional signal decomposition techniques in complex industrial applications.
• To enhance the accuracy and reliability of fault diagnosis using HBVMD.

Main Methods:

• Employs a top-down, binary recursive strategy to divide signals into low- and high-frequency components.
• Utilizes a time-varying mutual information criterion and modal energy threshold to determine decomposition depth.
• Introduces a dynamic dual-bandwidth penalty scheme with adaptive regularization and bandwidth relaxation.

Main Results:

• HBVMD successfully decomposes signals without requiring a predefined number of intrinsic modes.
• Experiments on bearing fault datasets demonstrate high diagnostic accuracy and reliability.
• The method excels in composite fault detection and feature extraction, outperforming existing approaches.

Conclusions:

• HBVMD offers a robust and interpretable framework for industrial fault diagnosis.
• The adaptive nature of HBVMD enhances its applicability to complex, real-world signals.
• This method significantly improves the detection and analysis of mechanical faults.