Vibration-based techniques are vital for online structural health monitoring (SHM) of rotating machines, enabling fault detection through feature analysis and threshold establishment. Rotating shafts typically exhibit non-linear dynamic behaviour, often due to misalignment or manufacturing imperfections leading to eccentricity. This non-linear behaviour is amplified after ballistic impact, leading to significant asymmetries and increased vibration loads. In this study, we develop an advanced vibration-based method to address the gap in diagnostic tools used to identify ballistic impact damage in helicopter transmission shafts. The proposed scheme employs a non-linear autoregressive model with exogenous inputs (NARX), evaluated against a long short-term memory (LSTM) model, to estimate acceleration signals from a two-sensor cluster. It then uses the estimation error arising from significant variations in signals acquired before and after ballistic impact to assess the structural integrity of the operating structure. The efficiency of the models is validated using experimental data obtained during ballistics testing. The results show that the proposed method effectively detects various types of impact damage, offering a promising tool for ballistic impact diagnosis in helicopter transmission shafts.
All-in-one image restoration aims to recover various degraded images using a unified model. To adaptively reconstruct high-quality images, recent prevalent CNN and Transformer based models incorporate learnable prompts to dynamically acquire degradation-specific knowledge for different degraded images, achieving state-of-the-art restoration performance. However, existing methods exhibit limitations, including high computational burden and inadequate modeling of long-range dependencies. To address these issues, we propose a reasoning and action prompt-driven Mamba-based image restoration model, namely RamIR. Specifically, RamIR employs the Mamba block for long-range dependencies modeling with linear computational complexity relative to the feature map size. Inspired by Chain-of-Thought (CoT) prompting, we integrate Reasoning and Action (ReAct) prompts within the Mamba block. Hence, we utilize the capability of pretrained vision language (PVL) models to generate textual reasoning prompts describing the type and severity of degradations. Simultaneously, another output from PVL acts as action prompt representing the clean image caption. These prompts, employed in a CoT manner, enhance the network’s sensitivity to degradation and elicit targeted recovery actions tailored to different reasoning prompts. Additionally, we explore the seamless interaction between Mamba blocks and prompts, introducing a novel prompt-driven module (PDM) to facilitate prompt utilization. Extensive experimental results demonstrate the superior performance of RamIR, highlighting its advantages in terms of input scaling efficiency over existing benchmark models for all-in-one image restoration.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：