Chair of Electronic Measurement and Diagnostic Technology, TU Berlin, Berlin, 10587, Germany.
School of Electrical Engineering and Computer Science, TU Berlin, Berlin, 10587, Germany.
Sci Rep. 2023 Apr 4;13(1):5484. doi: 10.1038/s41598-023-31532-9.
Convolutional Neural Network (CNN) has been extensively used in bearing fault diagnosis and Remaining Useful Life (RUL) prediction. However, accompanied by CNN's increasing performance is a deeper network structure and growing parameter size. This prevents it from being deployed in industrial applications with limited computation resources. To this end, this paper proposed a two-step method to build a cell-based light CNN by Neural Architecture Search (NAS) and weights-ranking-based model pruning. In the first step, a cell-based CNN was constructed with searched optimal cells and the number of stacking cells was limited to reduce the network size after influence analysis. To search for the optimal cells, a base CNN model with stacking cells was initially built, and Differentiable Architecture Search was adopted after continuous relaxation. In the second step, the connections in the built cell-based CNN were further reduced by weights-ranking-based pruning. Experiment data from the Case Western Reserve University was used for validation under the task of fault classification. Results showed that the CNN with only two cells achieved a test accuracy of 99.969% and kept at 99.968% even if 50% connections were removed. Furthermore, compared with base CNN, the parameter size of the 2-cells CNN was reduced from 9.677MB to 0.197MB. Finally, after minor revision, the network structure was adapted to achieve bearing RUL prediction and validated with the PRONOSTIA test data. Both tasks confirmed the feasibility and superiority of constructing a light cell-based CNN with NAS and pruning, which laid the potential to realize a light CNN in embedded systems.
卷积神经网络（CNN）在轴承故障诊断和剩余使用寿命（RUL）预测中得到了广泛应用。然而，随着 CNN 性能的提高，其网络结构也越来越深，参数规模也越来越大。这使得它无法部署在计算资源有限的工业应用中。为此，本文提出了一种通过神经结构搜索（NAS）和基于权重排名的模型剪枝来构建基于单元的轻量级 CNN 的两步方法。在第一步中，通过搜索最优单元构建基于单元的 CNN，并通过影响分析限制堆叠单元的数量，以减小网络规模。为了搜索最优单元，首先构建了一个具有堆叠单元的基础 CNN 模型，并采用连续松弛的方式进行可微分架构搜索。在第二步中，通过基于权重排名的剪枝进一步减少所构建的基于单元的 CNN 中的连接。使用凯斯西储大学的实验数据验证了在故障分类任务中的效果。结果表明，仅包含两个单元的 CNN 实现了 99.969%的测试精度，即使去除 50%的连接，仍保持 99.968%的精度。此外，与基础 CNN 相比，2 单元 CNN 的参数大小从 9.677MB 减少到 0.197MB。最后，经过少量修改，将网络结构适应于实现轴承 RUL 预测，并使用 PRONOSTIA 测试数据进行验证。这两个任务都证实了使用 NAS 和剪枝构建轻量级基于单元的 CNN 的可行性和优越性，这为在嵌入式系统中实现轻量级 CNN 奠定了基础。
