Due to the material variations of lithium-ion cells and fluctuations in their manufacturing precision, differences exist in electrochemical characteristics of cells, which inevitably lead to a reduction in the available capacity and premature failure of a battery pack with multiple cells configured in series, parallel, and series–parallel. Screening cells that have similar electrochemical characteristics to overcome the inconsistency among cells in a battery pack is a challenging problem. This paper proposes an approach for lithium -ion cell screening using convolutional neural networks (CNNs) based on two-step time-series clustering (TTSC) and hybrid resampling for imbalanced data, which takes into account the dynamic characteristics of lithium-ion cells, thus ensuring that the screened cells have similar electrochemical characteristics. In this approach, we propose the TTSC to label the raw samples and propose the hybrid resampling method to solve the sample imbalance issue, thereby obtaining labeled and balanced datasets and establishing the CNN model for online cell screening. Finally, industrial applications verify the effectiveness of the proposed approach and the inconsistency rate of the screened cells drops by 91.08%.

Domain generalization (DG) is a challenging task that aims to train a robust model with only labeled source data and can generalize well on unseen target data. The domain gap between the source and target data may degrade the performance. A plethora of methods resort to obtaining domain-invariant features to overcome the difficulties. However, these methods require sophisticated network designs or training strategies, causing inefficiency and complexity. In this paper, we first analyze and reclassify the features into two categories, i.e., implicitly disentangled ones and explicitly disentangled counterparts. Since we aim to design a generic algorithm for DG to alleviate the problems mentioned above, we focus more on the explicitly disentangled features due to their simplicity and interpretability. We find out that the shape features of images are simple and elegant choices based on our analysis. We extract the shape features from two aspects. In the aspect of networks, we propose Multi-Scale Amplitude Mixing (MSAM) to strengthen shape features at different layers of the network by Fourier transform. In the aspect of inputs, we propose a new data augmentation method called Random Shape Warping (RSW) to facilitate the model to concentrate more on the global structures of the objects. RSW randomly distorts the local parts of the images and keeps the global structures unchanged, which can further improve the robustness of the model. Our methods are simple yet efficient and can be conveniently used as plug-and-play modules. They can outperform state-of-the-art (SOTA) methods without bells and whistles.

It is difficult to detect the surface defects of a lithium battery with an aluminum/steel shell. The reflectivity, lack of 3D information on the battery surface, and the shortage of many datasets make the 2D detection method hard to apply in this field. In this paper, a cross-domain few-shot learning (FSL) approach for lithium-ion battery defect classification using an improved siamese network (BSR-SNet) is proposed. To obtain the critical 3D surface of the lithium-ion battery, a multiexposure-based structured light method is utilized. Then, the heights of the 3D cloud points are transferred to grayscale information and are saved as 8-bit 2D images. For the FSL task, the DAGM 2007 datasets are used as the source domain to pre-train the improved siamese model. To avoid negative mitigation in the target domain, batch spectral regularization (BSR) is added as a penalizer in the loss function. The accuracies of the experimental results are 93.3% for 10-shot batteries and 91.0% for 5-shot batteries, which means that our method can be used to classify the surface defects of lithium batteries well.

Machine learning-based diagnosis methods have achieved remarkable success under the assumption that the training and test data are identically distributed. However, a critical requirement of these methods is the generalization capability to unseen domains when deployed to actual diagnosis scenarios. We introduce the challenging problem of domain generalization, i.e., learning from multiple source domains to produce a model that can directly generalize to unseen domains without target information. We adopt a model-agnostic learning produce that maximizes the dot product of gradients between the source domains. Such a gradient alignment objective encourages finding a common optimization path for all source domains, which helps to focus on invariant representations. Furthermore, we propose two feature regularizations that explicitly regularize the feature space. Global feature regularization aligns class relationships between different domains to preserve the domain-invariant knowledge. Local feature regularization encourages the model to learn domain-agnostic class-specific representations with intraclass compactness and interclass separability. The effectiveness of the proposed method is demonstrated with generalization experiments on two benchmarks.

Online dynamic prediction of the hot metal silicon content in the blast furnace ironmaking process is crucial for stabilizing the furnace condition and improving the molten iron quality. However, due to the complex nonlinear correlations and time-varying time lags between silicon content and process variables, silicon content prediction is a challenging task. To tackle the problem, we propose a novel silicon content prediction method, called temporal hypergraph attention network (T-HyperGAT), which is combined the hypergraph attention network (HyperGAT) and the gated recurrent unit (GRU) network. Specifically, the Hyper-GAT is used to capture the high-order correlations of input features and perform equal-dimensional feature transformation to maintain the temporality of input features, and the GRU network is used to capture the time-series characteristics of transformed input features to overcome the time-varying time lags of process variables. Then the T-HyperGAT model can capture high-order correlations and time-series characteristics from complex industrial data. The effectiveness of the proposed T-HyperGAT method is verified by actual blast furnace ironmaking process data from a blast furnace in China.