In general, the safety and efficiency of thermal power plants require the collaboration of multiple coal mills. However, running data from different coal mills will introduce significant inconsistent distribution, resulting in suboptimal performance or even the unavailability of conventional diagnosis methods. To this end, this article presents an advantageous discriminant analysis-aided collaborative alignment network (DA-CAN) for cross-device fault diagnosis. First, the contribution of each feature in distinguishing source and target domains is determined by assigning the adaptive updated weight, which is helpful to keep the gradient direction more stable in domain transferring. To avoid destroying the inherent data structure of different domains, we design multiple complementary class-wise discrepancy metrics to enhance the domain consistency during the domain adaption process. After that, a joint training loss term with an adjustment factor is introduced to transform the private data of individual coal mills into collective representations and smooth the conditional and marginal distribution discrepancy collaboratively. Finally, the experimental results of the real-world coal mill group indicate that the DA-CAN is more effective and practical than the state-of-the-art transfer learning methods regarding multimachine fault diagnosis.

中文翻译：

---

判别分析引导的多机故障协同学习与诊断对准网络

一般来说，火电厂的安全和效率需要多个煤厂的协作。然而，不同磨煤机的运行数据会带来明显的不一致分布，导致性能不佳甚至无法使用常规诊断方法。为此，本文提出了一种用于跨设备故障诊断的有利判别分析辅助协作对齐网络（DA-CAN）。首先，通过分配自适应更新权重来确定每个特征在区分源域和目标域中的贡献，这有助于在域转移中保持梯度方向更加稳定。为了避免破坏不同域的固有数据结构，我们设计了多个互补的类差异度量来增强域适应过程中的域一致性。之后，引入带有调整因子的联合训练损失项，将各个煤厂的私有数据转化为集体表示，并协同平滑条件和边际分布差异。最后，真实磨煤机组的实验结果表明，DA-CAN 在多机故障诊断方面比最先进的迁移学习方法更有效、更实用。