Compressive Sensing Approach to Harmonics Detection in the Ship Electrical Network

Area of Science:

• Electrical Engineering
• Signal Processing
• Data Science

Background:

• Ship electrical networks are susceptible to harmonic distortions.
• Harmonic detection is crucial for system stability and performance.
• Existing methods may be computationally intensive or require extensive data.

Purpose of the Study:

• To investigate the application of compressive sensing (CS) for detecting harmonics in frequency-sparse signals.
• To develop and evaluate a fast reconstruction algorithm for harmonic analysis in ship electrical networks.
• To explore the use of discrete Radon transform (DRT) and K-rank-order filters within a CS framework.

Main Methods:

• Modeling ship electrical network signals as sparse representations in the discrete Fourier transform (DFT) domain.
• Utilizing compressive sensing (CS) principles for signal reconstruction from under-sampled measurements.
• Implementing a fast reconstruction algorithm based on the inverse discrete Radon transform (DRT).
• Applying K-rank-order filters in the sparse domain to enhance convergence and acquire sub-Nyquist data.
• Employing Bernoulli matrices for measurements, diverging from typical Gaussian approaches.

Main Results:

• Preliminary numerical simulations confirm the effectiveness of the proposed CS-based algorithm for harmonic detection.
• The algorithm demonstrates limitations that require further investigation.
• The approach leverages fast Fourier transform (FFT) and inverse FFT (IFFT) for efficient analysis.
• The data processing algorithm is characterized by low memory usage and processing load.

Conclusions:

• The proposed compressive sensing technique offers a promising and efficient method for detecting harmonics in ship electrical networks.
• The integration of DRT and K-rank-order filters provides a novel approach to under-sampled signal reconstruction.
• The algorithm's speed and simplicity present significant advantages for practical implementation.