Wang Haicheng, Zhou Juan, Chen Hu, Xu Bo, Shen Zhengxiang
College of Energy Environment and Safety Engineering and College of Carbon Metrology, China Jiliang University, Hangzhou, 310018, China.
Ningbo Special Equipment Inspection and Research Institute, Ningbo, 315000, China.
Sci Rep. 2024 Jul 7;14(1):15620. doi: 10.1038/s41598-024-66541-9.
Hydraulic systems play a pivotal and extensive role in mechanics and energy. However, the performance of intelligent fault diagnosis models for multiple components is often hindered by the complexity, variability, strong hermeticity, intricate structures, and fault concealment in real-world conditions. This study proposes a new approach for hydraulic fault diagnosis that leverages 2D temporal modeling and attention mechanisms for decoupling compound faults and extracting features from multisample rate sensor data. Initially, to address the issue of oversampling in some high-frequency sensors within the dataset, variable frequency data sampling is employed during the data preprocessing stage to resample redundant data. Subsequently, two-dimensional convolution simultaneously captures both the instantaneous and long-term features of the sensor signals for the coupling signals of hydraulic system sensors. Lastly, to address the challenge of feature fusion with multisample rate sensor data, where direct merging of features through maximum or average pooling might dilute crucial information, a feature fusion and decoupling method based on a probabilistic sparse self-attention mechanism is designed, avoiding the issue of long-tail distribution in multisample rate sensor data. Experimental validation showed that the proposed model can effectively utilize samples to achieve accurate fault decoupling and classification for different components, achieving a diagnostic accuracy exceeding 97% and demonstrating robust performance in hydraulic system fault diagnosis under noise conditions.
液压系统在机械和能源领域发挥着关键且广泛的作用。然而，在实际工况中，由于系统的复杂性、多变性、强密封性、结构的错综复杂以及故障的隐蔽性，多部件智能故障诊断模型的性能常常受到阻碍。本研究提出了一种新的液压故障诊断方法，该方法利用二维时间建模和注意力机制来解耦复合故障，并从多采样率传感器数据中提取特征。首先，为了解决数据集中一些高频传感器的过采样问题，在数据预处理阶段采用变频数据采样对冗余数据进行重采样。随后，二维卷积同时捕捉液压系统传感器耦合信号的传感器信号的瞬时特征和长期特征。最后，为了解决多采样率传感器数据的特征融合挑战，即通过最大池化或平均池化直接合并特征可能会稀释关键信息，设计了一种基于概率稀疏自注意力机制的特征融合和解耦方法，避免了多采样率传感器数据中的长尾分布问题。实验验证表明，所提出的模型能够有效利用样本，实现对不同部件的准确故障解耦和分类，诊断准确率超过97%，并在噪声条件下的液压系统故障诊断中表现出强大的性能。
