| 朱江艳,马军,吴建德,熊新.双流自适应时空特征融合网络的机械设备剩余寿命预测方法[J].电子测量与仪器学报,2026,40(3):143-154 |
| 双流自适应时空特征融合网络的机械设备剩余寿命预测方法 |
| Dual-stream adaptive temporal-spatial feature fusion network forremaining useful life prediction of mechanical equipment |
| |
| DOI： |
| 中文关键词: 剩余寿命预测 时空特征融合 卷积自注意 多尺度卷积神经网络 自适应加权 |
| 英文关键词:residual life prediction spatial-temporal feature fusion convolutional self-attention multi-scale CNN adaptive weighting |
| 基金项目:国家自然科学基金(U23A20329，62563018)、云南省基础研究计划项目 (202401AW070014)、云南省“兴滇英才支持计划’产业创新人才专项(yfgrc202402)资助 |
|
|
| Author | Institution |
| Zhu Jiangyan | Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China |
| Ma Jun | 1.Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China;
2.Yunnan Key Laboratory of Intelligent Control and Application, Kunming University of Science and Technology,
Kunming 650500, China |
| Wu Jiande | School of Engineering, Yunnan University, Kunming 650500, China |
| Xiong Xin | 1.Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China;
2.Yunnan Key Laboratory of Intelligent Control and Application, Kunming University of Science and Technology,
Kunming 650500, China |
|
| 摘要点击次数: 392 |
| 全文下载次数: 239 |
| 中文摘要: |
| 复杂系统的状态监测数据是多个传感器采集的多源时空信息，为了充分利用时间退化特征和测量变量间的空间关联信息，提出了基于双流结构的自适应时空特征融合神经网络（temporal-spatial feature fusion neural network, TSTFNN），通过并行的时间流和空间流结构分别提取时间依赖特征和空间关联特征。设计了卷积自注意机制，以弥补传统点积自注意力忽略时间序列连续性的缺点，使模型更好地捕捉时间序列的连续性和细节变化。同时，应用多尺度卷积神经网络挖掘不同测量变量之间的空间关联特征，提升模型的全局感知能力。在特征融合阶段，引入自适应加权机制，实现时间和空间特征的动态融合。为了优化模型的预测效果，构建了由约束均方误差损失和特征平衡损失组成的联合损失函数，来增强时空特征的协同学习能力。最后，基于NASA的C-MAPSS基准数据集的实验结果表明，所提方法在多源数据RUL预测精度方面优于多种SOTA模型。 |
| 英文摘要: |
| In complex systems, condition monitoring data consist of multi-source spatiotemporal information collected from multiple sensors. To effectively capture temporal degradation patterns and spatial correlations among measured variables, we propose an adaptive temporal-spatial feature fusion neural network (TSTFNN) based on a dual-stream structure. This framework incorporates parallel temporal and spatial streams to extract temporal dependencies and spatial correlations separately. To overcome the limitations of traditional dot-product self-attention, which often neglects time-series continuity, a convolutional self-attention mechanism is implemented, enhancing the capacity of model to capture sequential continuity and subtle temporal variations. A multi scale convolutional neural network further extracts spatial correlation features across variables, improving global perception capabilities. During feature fusion, an adaptive weighting mechanism enables dynamic integration of temporal and spatial features. To optimize predictive performance, a joint loss function, combining constrained mean squared error (MSE) and feature balance loss, is introduced, facilitating the collaborative learning of temporal and spatial features. Finally, experimental results based on NASA′s C-MAPSS benchmark dataset demonstrate that the proposed method outperforms various state-of-the-art (SOTA) models in terms of multi-source data RUL prediction accuracy. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
