Search

The paper proposes a method based on kernel principal component analysis (KPCA) and multi-scale temporal convolution network (MTCN) for identifying faults in lithium-ion batteries, which is crucial for ensuring the safe and stable operation of energy-storage systems. Lithium-ion batteries are the primary component of energy storage units. The method involves the following steps: First, fault data are normalized, and KPCA is used for dimensionality reduction and single fault detection to reduce computational complexity and improve data reliability. According to the different types of overcharge faults and unknown faults, KPCA is used to reduce the data from the original dimension to 2 or 4 dimensions. The KPCA model is trained using the normal data corresponding to the two groups of fault data, and the fault data are inputted as the test data. The results show that the SPE statistic and the T2 statistic considerably exceed the control limit, verifying the reliability of the data. Then, the data are labeled according to the fault type: the overcharge data are labeled as D0 (normal) and D1 (fault), and the unknown fault data as F0 (normal) and F1 (fault). The labeled data are divided into training and test sets according to a specific proportion. Afterward, the MTCN model is trained with the training dataset, and its hyperparameters are optimized with the frost algorithm to improve model accuracy. Finally, the trained MTCN model is used to classify the test dataset. The method is validated on two groups of data: overcharge fault data and unknown fault data. The results show that the frost algorithm can optimize the hyperparameters after about 20 iterations. Compared with LSTM and CNN, which are also optimized by the frost algorithm, MTCN achieves a higher classification accuracy, reaching 99.265% and 99.688%, on the overcharge fault dataset and unknown fault dataset, respectively, while maintaining comparable performance to Xception and ResNet50. Additionally, to verify the influence of the training data amount, the training and test sets are divided according to different proportions, and the three algorithms are tested. KPCA verifies the reliability of charging fault and unknown fault data, and the results show that MTCN has the highest classification accuracy, especially on the overcharge fault dataset. Owing to the low dimensionality of the original data set, LSTM and CNN exhibit poor classification performance. In contrast, MTCN can extract more temporal information, achieving high classification accuracy. These results demonstrate the effectiveness and superiority of the method in fault diagnosis of lithium-ion batteries.

The push for renewable energy emphasizes the need for energy storage systems (ESSs) to mitigate the unpre-dictability and variability of these sources, yet challenges such as high investment costs, sporadic utilization, and demand mismatch hinder their broader adoption. In response, shared energy storage systems (SESSs) offer a more cohesive and efficient use of ESS, providing more accessible and cost-effective energy storage solutions to overcome these obstacles. To enhance the profitability of SESSs, this paper designs a multi-time-scale resource allocation strategy based on long-term contracts and real-time rental business models. We initially construct a life cycle cost model for SESS and introduce a method to estimate the degradation costs of multiple battery groups by cycling numbers and depth of discharge within the SESS. Subsequently, we design various long-term contracts from both capacity and energy perspectives, establishing associated models and real-time rental models. Lastly, multi-time-scale resource allocation based on the decomposition of user demand is proposed. Numerical analysis validates that the business model based on long-term contracts excels over models operating solely in the real-time market in economic viability and user satisfaction, effectively reducing battery degradation, and leveraging the aggregation effect for SESS can generate an additional increase of 10.7% in net revenue.

With the rapid progression of Energy Storage Systems (ESSs), the capability of extensively distributed and heterogeneous ESSs to support the power grid remains largely underexplored. To better exploit the potential of these numerous ESSs and enhance their service to the power grid, this paper proposes a model for evaluating and aggregating the grid-support capability of energy storage clusters by considering the peak regulation requirements. To begin with, the proposed model employs subjective and objective combination weighting methods to establish a grid-support capability matrix between ESSs indicators and grid demand scenarios, thereby facilitating the identification of the ESSs with a strong ability to regulate peak power. Next, based on the dual-peak pattern of grid load and diverse characteristics of ESSs, the ESSs in the peak regulation cluster are evaluated by clustering again. In addition, taking into account the operational constraints of the ESSs and the peak regulation requirements, a grid-support capability aggregation model for energy storage clusters based on the revised Chino polytope is proposed. The case study results demonstrate that the proposed model not only balances computational efficiency and aggregation accuracy to a certain extent but also enhances the capability of energy storage clusters to participate in peak regulation of the power grid.

Mycotoxin deoxynivalenol (DON) produced by the Fusarium graminearum complex is highly toxic to animal and human health. During DON synthesis, the endoplasmic reticulum (ER) of F. graminearum is intensively reorganized, from thin reticular structure to thickened spherical and crescent structure, which was referred to as "DON toxisome". However, the underlying mechanism of how the ER is reorganized into toxisome remains unknown. In this study, we discovered that overproduction of ER-localized DON biosynthetic enzyme Tri4 or Tri1, or intrinsic ER-resident membrane proteins FgHmr1 and FgCnx was sufficient to induce toxisome-shaped structure (TSS) formation under non-toxin-inducing conditions. Moreover, heterologous overexpression of Tri1 and Tri4 proteins in non-DON-producing fungi F. oxysporum f. sp. lycopersici and F. fujikuroi also led to TSS formation. In addition, we found that the high osmolarity glycerol (HOG), but not the unfolded protein response (UPR) signaling pathway was involved in the assembly of ER into TSS. By using toxisome as a biomarker, we screened and identified a novel chemical which exhibited high inhibitory activity against toxisome formation and DON biosynthesis, and inhibited Fusarium growth species-specifically. Taken together, this study demonstrated that the essence of ER remodeling into toxisome structure is a response to the overproduction of ER-localized DON biosynthetic enzymes, providing a novel pathway for management of mycotoxin contamination.

In hyperspectral image classification (HSIC), every pixel of the HSI is assigned to a land cover category. While convolutional neural network (CNN)-based methods for HSIC have significantly enhanced performance, they encounter challenges in learning the relevance of deep semantic features and grappling with escalating computational costs as network depth increases. In contrast, the transformer framework is adept at capturing the relevance of high-level semantic features, presenting an effective solution to address the limitations encountered by CNN-based approaches. This article introduces a novel adaptive learnable spectral–spatial fusion transformer (ALSST) to enhance HSI classification. The model incorporates a dual-branch adaptive spectral–spatial fusion gating mechanism (ASSF), which captures spectral–spatial fusion features effectively from images. The ASSF comprises two key components: the point depthwise attention module (PDWA) for spectral feature extraction and the asymmetric depthwise attention module (ADWA) for spatial feature extraction. The model efficiently obtains spectral–spatial fusion features by multiplying the outputs of these two branches. Furthermore, we integrate the layer scale and DropKey into the traditional transformer encoder and multi-head self-attention (MHSA) to form a new transformer with a layer scale and DropKey (LD-Former). This innovation enhances data dynamics and mitigates performance degradation in deeper encoder layers. The experiments detailed in this article are executed on four renowned datasets: Trento (TR), MUUFL (MU), Augsburg (AU), and the University of Pavia (UP). The findings demonstrate that the ALSST model secures optimal performance, surpassing some existing models. The overall accuracy (OA) is 99.70%, 89.72%, 97.84%, and 99.78% on four famous datasets: Trento (TR), MUUFL (MU), Augsburg (AU), and University of Pavia (UP), respectively.

Utilizing multi-modal data, as opposed to only hyperspectral image (HSI), enhances target identification accuracy in remote sensing. Transformers are applied to multi-modal data classification for their long-range dependency but often overlook intrinsic image structure by directly flattening image blocks into vectors. Moreover, as the encoder deepens, unprofitable information negatively impacts classification performance. Therefore, this paper proposes a learnable transformer with an adaptive gating mechanism (AGMLT). Firstly, a spectral–spatial adaptive gating mechanism (SSAGM) is designed to comprehensively extract the local information from images. It mainly contains point depthwise attention (PDWA) and asymmetric depthwise attention (ADWA). The former is for extracting spectral information of HSI, and the latter is for extracting spatial information of HSI and elevation information of LiDAR-derived rasterized digital surface models (LiDAR-DSM). By omitting linear layers, local continuity is maintained. Then, the layer Scale and learnable transition matrix are introduced to the original transformer encoder and self-attention to form the learnable transformer (L-Former). It improves data dynamics and prevents performance degradation as the encoder deepens. Subsequently, learnable cross-attention (LC-Attention) with the learnable transfer matrix is designed to augment the fusion of multi-modal data by enriching feature information. Finally, poly loss, known for its adaptability with multi-modal data, is employed in training the model. Experiments in the paper are conducted on four famous multi-modal datasets: Trento (TR), MUUFL (MU), Augsburg (AU), and Houston2013 (HU). The results show that AGMLT achieves optimal performance over some existing models.

Methods based on convolutional neural networks (CNNs) have become a vital offshoot for hyperspectral image (HSI) classification. In recent years, the 3-D CNN (3-DCNN) has become dominant due to its excellent capability of extracting features. However, the high dimension and the limited training samples of HSI usually restrict the improvement of its classification accuracy. And the parameters of conventional 3-DCNN are larger so that computational complexity and running time increase. Therefore, a new model named HSSAM is proposed to solve the above problems. First, a 3-D residual-dense asymmetric convolution (3-D-RDAC) is designed to reuse the features, while reducing the parameters. Subsequently, 3-D-RDAC combined with the multiscale convolution to construct a 3-D multiscale RDAC (3-D-MRDAC) for avoiding the blind spots and unrecognized regions of receiving fields. Then, 3-D attention SimAM is applied to 3-D-MRDAC, for constituting the heterogeneous spectral-spatial attention convolutional neural (HSSAN) block, to extract spectral-spatial features of HSI adequately. Ultimately, MLP acts as the output layer of the model to better deal with the nonlinear features of HSI. Experiments in this article are carried out on four famous hyperspectral datasets: Indian Pines; Pavia University; WHU-Hi-LongKou; and WHU-Hi-HanChuan. Results show that HSSAM achieves better classification accuracy with limited training samples than several existing models. Overall accuracy reaches 96.84%, 98.85%, 98.01%, and 97.18% on the four datasets, respectively.

Due to the increasing presence of small objects in videos or images from practical applications, small object identification is currently an extremely popular topic in the field of machine vision. Additionally, small object detection is still a difficult process because to small objects’ issues with fuzzy appearance, limited information, occlusion, and noise present. Most existing methods mainly use feature pyramid networks to enrich shallow features using contextual features. However, due to the inconsistency of gradients between different layers of the feature pyramid network, the shallow features cannot be fully utilized resulting in the slow improvement of small object detection accuracy. To effectively improve the small object detection algorithm, we propose a new feature pyramid network-based small object detection algorithm, SSRDet. To effectively assign positive and negative sample labels and address the issue of sample scale imbalance, we first present RFLA. Then, to overcome the gradient inconsistency between various layers and enable the full utilization of the shallow features, we extend the feature pyramid network by including a scale enhancement module (SEM) and a scale selection module (SSM). Finally, we introduced the attention module (SPAM) to filter out the background noise in the shallow feature extraction to better extract small object features. We validated our method on VisDrone2019 and AI-TOD, and our method outperformed the state-of-the-art detectors.

Human RBM42 associates with the spliceosome complex. Here the authors show that the fungus counterpart of RBM42, FgRbp1 regulates splicing by interacting with FgU2AF23.

Purpose: We aimed to describe the clinical characteristics and outcomes of patients with transverse myelitis (TM) as a rare manifestation in systemic lupus erythematosus (SLE) and explore the risk factors and prognosis of SLE-related TM (SLE-TM). Methods: We conducted a retrospective case–control and cohort analysis. All patients with SLE-TM (58 patients) and 232 with SLE without TM, as a control group, were admitted to Peking Union Medical College Hospital between January 1993 and May 2021. Factors associated with the presence of SLE-TM and its prognosis were assessed using logistic regression and Cox proportional hazard models. Results: Multivariate analysis revealed that positive anti-Ro/Sjogren’s syndrome A (anti-Ro/ SSA) (

Heterogeneous defect prediction (HDP) aims to predict the defect tendency of modules in one project using heterogeneous data collected from other projects. It sufficiently incorporates the two characteristics of the defect prediction data: (1) datasets could have different metrics and distribution, and (2) data could be highly imbalanced. In this paper, we propose a few-shot learning based balanced distribution adaptation (FSLBDA) approach for heterogeneous defect prediction, which takes into consideration the two characteristics of the defect prediction data. Class imbalance of the defect datasets can be solved with undersampling, but the scale of the training datasets will be smaller. Specifically, we first remove redundant metrics of datasets with extreme gradient boosting. Then, we reduce the data difference between the source domain and the target domain with the balanced distribution adaptation. It considers the marginal distribution and the probability of conditional distribution differences and adaptively assigns different weights to them. Finally, we use adaptive boosting to relieve the influence caused by the size of the training dataset is smaller, which can improve the accuracy of the defect prediction model. We conduct experiments on 17 projects from 4 datasets using 3 indicators (i.e., AUC, G-mean, F-measure). Compared to three classic approaches, the experimental results show that FSLBDA can effectively improve the prediction performance.

Light Detection and Ranging (LiDAR) data are widely used for high-resolution land cover mapping, which can provide very valuable information about the height of the surveyed area for the discrimination of classes. In order to utilize the advantages of deep models for the classification of LiDAR-derived features, a new classification algorithm combined Octave Convolution (OctConv) with Capsule Network (CapsNet), is proposed here to hierarchically extract robust and discriminant features of the input data, called as OctConv-CapsNet. In the proposed approach, CapsNet captures the spatial information of the data, and OctConv processes separately for high- and low-frequency feature. OctConv is embedded in the primary capsule layer of CapsNet so that the proposed approach can make the most of both the spatial information and the high- and low-frequency information simultaneously. The proposed framework performs experiments on two LiDAR-DSM datasets (i.e. Bayview Park and Recology datasets). The results show that, compared with the traditional deep convolution model, OctConv-CapsNet can improve the classification accuracy of LiDAR-DSM data, and when the number of training samples of the experiment is 800, the classification accuracies reached 96.12% and 96.79% on Bayview Park and Recology datasets, respectively.

Antibody repertoire sequencing (Rep-seq) has been widely used to reveal repertoire dynamics and to interrogate antibodies of interest at single nucleotide-level resolution. However, polymerase chain reaction (PCR) amplification introduces extensive artifacts including chimeras and nucleotide errors, leading to false discovery of antibodies and incorrect assessment of somatic hypermutations (SHMs) which subsequently mislead downstream investigations. Here, a novel approach named DUMPArts, which improves the accuracy of antibody repertoires by labeling each sample with dual barcodes and each molecule with dual unique molecular identifiers (UMIs) via minimal PCR amplification to remove artifacts, is developed. Tested by ultra-deep Rep-seq data, DUMPArts removed inter-sample chimeras, which cause artifactual shared clones and constitute approximately 15% of reads in the library, as well as intra-sample chimeras with erroneous SHMs and constituting approximately 20% of the reads, and corrected base errors and amplification biases by consensus building. The removal of these artifacts will provide an accurate assessment of antibody repertoires and benefit related studies, especially mAb discovery and antibody-guided vaccine design.

The antibody repertoire is a critical component of the adaptive immune system and is believed to reflect an individual’s immune history and current immune status. Delineating the antibody repertoire has advanced our understanding of humoral immunity, facilitated antibody discovery, and showed great potential for improving the diagnosis and treatment of disease. However, no tool to date has effectively integrated big Rep-seq data and prior knowledge of functional antibodies to elucidate the remarkably diverse antibody repertoire. We developed a Rep-seq dataset Analysis Platform with an Integrated antibody Database (RAPID; https://rapid.zzhlab.org/), a free and web-based tool that allows researchers to process and analyse Rep-seq datasets. RAPID consolidates 521 WHO-recognized therapeutic antibodies, 88,059 antigen- or disease-specific antibodies, and 306 million clones extracted from 2,449 human IGH Rep-seq datasets generated from individuals with 29 different health conditions. RAPID also integrates a standardized Rep-seq dataset analysis pipeline to enable users to upload and analyse their datasets. In the process, users can also select set of existing repertoires for comparison. RAPID automatically annotates clones based on integrated therapeutic and known antibodies, and users can easily query antibodies or repertoires based on sequence or optional keywords. With its powerful analysis functions and rich set of antibody and antibody repertoire information, RAPID will benefit researchers in adaptive immune studies.

LiDAR data contain feature information such as the height and shape of the ground target and play an important role for land classification. The effect of convolutional neural network (CNN) for feature extraction on LiDAR data is very significant, however CNN cannot resolve the spatial relationship of features adequately. The capsule network (CapsNet) can identify the spatial variations of features and is widely used in supervised learning. In this article, the CapsNet is combined with the residual network (ResNet) to design a deep network-ResCapNet for improving the accuracy of LiDAR classification. The capsule network represents the features by vectors, which can account for the direction of the features and the relative position between the features. Therefore, more detailed feature information can be extracted. ResNet protects the integrity of information by passing input information to the output directly, which can solve the problem of network degradation caused by information loss in the traditional CNN propagation process to a certain extent. Two different LiDAR data sets and several classic machine learning algorithms are used for comparative experiments. The experimental results show that ResCapNet proposed in this article `improve the performance of LiDAR classification.

In this study, we researched the problem of self-tuning (ST) distributed fusion state estimation for multi-sensor networked stochastic linear discrete-time systems with unknown packet receiving rates, noise variances (NVs), and model parameters (MPs). Packet dropouts may occur when sensor data are sent to a local processor. A Bernoulli distributed stochastic variable is adopted to depict phenomena of packet dropouts. By model transformation, the identification problem of packet receiving rates is transformed into that of unknown MPs for a new augmented system. The recursive extended least squares (RELS) algorithm is used to simultaneously identify packet receiving rates and MPs in the original system. Then, a correlation function method is used to identify unknown NVs. Further, a ST distributed fusion state filter is achieved by applying identified packet receiving rates, NVs, and MPs to the corresponding optimal estimation algorithms. It is strictly proven that ST algorithms converge to optimal algorithms under the condition that the identifiers for parameters are consistent. Two examples verify the effectiveness of the proposed algorithms.