Self-healing is a biological phenomenon in which living organism responds to the suffered damage in a complex way. Inspired by the self-healing phenomenon in nature, various biomimetic healing methods rooted in intrinsic or extrinsic healing mechanisms have been explored. Research on novel self-healing asphalt materials with intelligent response is at the cutting-edge of materials science and offers a potential strategy for building long-life and low-carbon asphalt concrete infrastructure. This paper describes the progress of research on extrinsic self-healing asphalt materials and makes a clear distinction between intrinsic and extrinsic self-healing. The asphalt self-healing mechanism is interpreted by capillary flow theory, phase field theory, molecular diffusion theory and surface energy theory form various perspective. The extrinsic self-healing strategies including thermal induced healing and rejuvenator induced healing are proposed to enhance the healing level of cracked asphalt materials. A brief review of the methods including fracture-healing test and fatigue-healing test for assessing the efficacy of different extrinsic healing methods is presented. The thermal induced healing method bring high crack repair efficiency for asphalt concrete and the rejuvenator induced healing strategy not only improve the healing ratio of cracked asphalt concrete but also regenerate the ageing asphalt in situ. Important lessons for prospective research on the creation of novel self-healing asphalt materials are highlighted.@en

Machine intelligence fault prediction (MIFP) is crucial for ensuring complex systems safe and reliable operation. While deep learning has become the mainstream tool for MIFP due to its excellent learning abilities, its interpretability is limited, and it struggles to learn frequencies, making it challenging to understand the physical knowledge of signals at the frequency level. Therefore, this article proposes a physically interpretable wavelet-guided network (WaveGNet) with deep frequency separation for MIFP, inspired by the sound theoretical basis and physical meaning of discrete wavelet transform (DWT). WaveGNet expands the feature learning space of CNN into the frequency domain, allowing for a better understanding of the physical insights behind the frequency level. Specifically, WaveGNet involves a derivable and learnable frequency learning layer (FL-Layer) consisting of a wavelet-driven frequency decomposition module and a convolution-driven feature learning module. Multiple DWT-driven FL-Layers are used in WaveGNet to achieve deep frequency decomposition and multiresolution frequency feature learning in a coarse-to-fine manner. The effectiveness of WaveGNet was evaluated in real high-speed train wheel wear monitoring and high-speed aviation bearing fault diagnosis cases. Experimental results showed that WaveGNet outperforms cutting-edge deep learning algorithms and has excellent fault diagnosis and prediction abilities. Furthermore, an in-depth analysis of the learning mechanism of wavelet-driven CNN from the frequency domain perspective was conducted.@en