Acoustic and Vibration Fusion-Based Fault Diagnosis for Wind Turbine Blades: A Review from Blade Structure, Fault Modes to Intelligent Diagnosis Methods

Wind turbine blades (WTBs) are critical for wind capture and energy conversion; their condition therefore determines turbine efficiency and safety. Exposed to sustained aerodynamic loading and harsh environments, blades commonly suffer surface erosion, cracking, and delamination that can progress to structural failure. With the rapid expansion of wind power, traditional single-modality monitoring, such as vibration or airborne acoustic sensing, has revealed limitations in information completeness and resistance to interference. Consequently, multimodal fusion diagnosis has gained attention. Because most studies emphasize a single sensing modality and lack systematic reviews linking damage mechanisms, signal responses, and fusion strategies, this paper reviews acoustic-vibration fusion fault diagnosis for WTBs. Beginning with blade material and structural features, we characterize typical damage modes and their progression. We first review single-modality detection methods and highlight their limitations under complex operating conditions, establishing the case for multimodal fusion. We then systematically summarize signal-processing techniques, diagnostic models, and fusion strategies at data, feature, and decision levels. We compare approaches in terms of feature representation capacity, resistance to interference, and engineering applicability. Finally, in light of challenges in acquiring data in complex environments and the emerging role of digital-twin-based dynamic early warning, we outline promising directions for future research.