Impact localisation of composite laminates based on sparse enhancement and similarity matching

Due to strong noise interference in practical applications, some valuable damage characteristics are often obscured in the measurement data obtained from impact tests on composite laminates, making it difficult to directly identify the impact damage locations. The measurement signals contain both stress wave components generated by the impact and noise components introduced by environmental and system factors. This paper proposes a sparse enhancement and similarity matching strategy, which is a sparse denoising model with the generalised minimax-concave (GMC) penalty function to enhance the sparse features and denoise the contaminated data. Also, principal component analysis (PCA) is employed to further reduce the complexity of the impact signal data after an initial dimensionality reduction. Following this, an impact localisation matching procedure is carried out to pinpoint the impact damage location. Through low-velocity impact experiments on composite laminates, the proposed method demonstrates high levels of effectiveness and efficiency in localising the impact damage.