Although condition monitoring systems (CMS) provide detailed information about the health of critical wind turbine components, their practical application in wind farm operations and maintenance (O&M) remains limited, significantly diminishing their value. The reasons for this situation are not merely technical issues, rather, a complex interplay of various factors besides the condition monitoring algorithms themselves. The purpose of the paper is to investigate these reasons and propose a potential solution. Through in-depth analysis, two gaps are identified. The first gap is the Data-Model Gap, in which operational wind turbine data do not adequately satisfy the assumptions and quality requirements of condition monitoring algorithms. The second is the Model-Decision Gap, where the CMS output fails to provide sufficient information to support maintenance decisions. Based on this, a Data-Model-Decision Gap (DMDG) framework is further developed to explain how maintenance requirements can be used to determine the CMS outputs and, subsequently, the wind turbine condition monitoring data that are needed to produce these CMS outputs. Such a CMS design logic reverses the usual data-to-model development sequence by starting from understanding what kind of CMS outputs can actually support decision-making in practical wind farm O&M. This new CMS design approach emphasizes the alignment of data, models, and decision processes through an iterative and integrated design strategy. Research findings indicate that merely improving algorithm accuracy is insufficient to enhance the practical value of CMS. Bridging the aforementioned gaps is the key to transforming the CMS design from “monitoring-oriented” to “decision-oriented” and in the end, providing “prescriptive maintenance solutions”. This change is critical for unlocking the full potential of CMS in wind farm O&M.

Rotating components in aviation equipment are subject to pronounced distribution shifts under different rotational speeds, loads, and installation conditions. As a result, conventional deep diagnostic models often suffer from performance degradation when deployed beyond the training operating condition. To address the practical constraint that target-domain samples are unavailable in advance and only one source operating condition can be used for training, this paper proposes a bearing fault diagnosis method that integrates frequency-domain statistical priors with class-conditional representation alignment. The one-dimensional vibration signal is transformed into both a frequency-domain amplitude spectrum and a two-dimensional time-frequency representation. The former is used to construct low-dimensional physical statistical descriptors, including center frequency, standard-deviation frequency, root-mean-square frequency, and kurtosis frequency, whereas the latter is fed into a convolutional network to extract deep discriminative representations. During training, a class-conditional prior-deep representation alignment constraint is introduced to pull the deep features of samples from the same class toward their corresponding frequency-domain prior representations while suppressing interference from priors of different classes. The two types of features are then concatenated and fed into a classifier for fault identification. Cross-condition experiments on a bearing dataset demonstrate that the proposed method improves diagnostic stability under unseen operating conditions without using target-domain training samples, thereby providing an interpretable modeling scheme for health monitoring of aviation equipment in high-cost and limited-source scenarios.

In mechanical systems, bearings play a vital role, yet their surfaces frequently develop defects during the production process. Hence, automated detection of such defects becomes indispensable for ensuring industrial quality control. To enhance the robustness of bearing surface defect detection, a dynamic sample matching detection network is proposed, which is guided by multi-level perceptual features. A lightweight adaptive feature extraction method is introduced to enhance defect representation while maintaining low parameter complexity and computational cost. To address diverse defect patterns, a multi-shape perception module with a dual-axis cross-weighting mechanism across multiple scales is designed to improve sensitivity to different defect types. Furthermore, a dynamic sample selection strategy with dual-label weighting is introduced to improve training sample quality. The proposed approach is evaluated on a self-developed bearing defect platform, where it outperforms mainstream detection models in complex environments and achieves superior performance and robustness.