Accurate state-of-charge (SOC) estimation is essential for lithium-ion battery management systems, but remains challenging under battery-to-battery variation. This paper proposes a physics-guided lightweight NAS-TCN-Transformer with coulomb-counting residual learning for cross-battery SOC estimation. A five-dimensional feature representation is constructed by integrating voltage, current, temperature, voltage change rate, and cumulative discharged ampere-hour. The model predicts the residual between a coulomb-counting baseline and the target SOC rather than directly regressing SOC. Experiments on the NASA battery dataset show that the proposed method outperforms LSTM, TCN, and NAS-TCN-Transformer baselines in both single-battery and cross-battery settings, while retaining a lightweight architecture with only 18.676K parameters. The results confirm its effectiveness in improving estimation accuracy, generalization, and deployment efficiency.

Accurate State of Charge (SOC) estimation is critical for the safe and efficient operation of batteries in electric vehicles (EVs). While deep learning models like Transformers have shown promise, they often struggle with sensor noise and complex temporal dynamics. Similarly, hybrid approaches like VMD-Transformer rely on fixed basis functions that lack adaptability. To address challenges such as sensor noise, nonlinear dynamics, and complex temporal dependencies, this study proposes a novel VMD-Basisformer model that integrates Variational Mode Decomposition (VMD) with an enhanced Basisformer neural network. The key innovations of our approach include: (1) a battery-optimized VMD process for noise reduction and multi-scale feature extraction; (2) a dual-path basis generation mechanism tailored to battery temporal dynamics; (3) a hierarchical attention architecture for capturing both local and global temporal dependencies. Experimental results under Dynamic Stress Test (DST) and Urban Dynamometer Driving Schedule (UDDS) conditions show that the proposed model significantly outperforms existing methods such as Transformer, Basisformer, and VMD-Transformer. Experimental validation on LiFePO 4 battery and supercapacitor datasets under DST and UDDS conditions shows that the VMD-Basisformer outperforms benchmark models (Transformer, Basisformer, VMD-Transformer) in accuracy and robustness. Ablation studies confirm the critical role of InfoNCE loss in ensuring temporal consistency.

Flow meter is one of the most essential sensors in industrial development, energy measurement and environmental protection. Monitoring of flow meter performance can help detect anomalies early and enable timely corrective actions for critical industrial equipment in harsh operating environments. However, flow meter diagnostic models are often prone to overfitting and low accuracy caused by class-imbalanced small-sample data. To address these problems, a reinforcement learning Mahalanobis Taguchi system (RLMTS) model is proposed in this paper, which primarily consists of three modules, namely Mahalanobis space (MS) construction, threshold determination, and sample classification. In the MS module, an initial MS is constructed by selecting variables through orthogonal array design and signal-to-noise ratio analysis. Reinforcement learning is then introduced to adaptively refine the MS which is verified by the Mahalanobis distance. In the threshold determination module, a neural network algorithm is proposed to replace the traditional quality loss function for optimal threshold determination. In the sample classification module, the fault diagnosis of unknown samples is performed using the valid MS and calculated Mahalanobis distance. Experimental results show that the proposed RLMTS is not only suitable for flow meter fault diagnosis under different class-imbalance ratios with different small sample sizes, but also demonstrates a better diagnostic performance, stronger robustness, and broader applicability compared to the 19 benchmark diagnosis models. The use of RLMTS therefore guarantees stable operation of the flow meters, contributing to energy savings and environmental protection.

As a remote sensing indicator of vegetation physiology, solar-induced chlorophyll fluorescence (SIF) serves as a proxy for photosynthesis and has been widely used for monitoring stress conditions such as drought and for estimating crop yields. However, few studies systematically evaluated the capacity of SIF to indicate crop yield variations under drought and non-drought conditions across different temporal scales. This study focused on the Huang-Huai-Hai Plain (HHHP) and used the Standardized Soil Moisture Index (SSMI) to identify the spatial distribution of agricultural drought. We compared the sensitivity of SIF740, SIF683, NDVI, and NIRv to drought during different growth stages of winter wheat, and examined the associations between SIF740 and winter wheat yield under drought and non-drought conditions at various temporal scales. The results indicated that SIF740 was more sensitive to drought than SIF683, NDVI, and NIRv, particularly during the regreening stage. Under non-drought conditions, cumulative SIF740 during the jointing-anthesis stage exhibited the strongest correlation with winter wheat yield, outperforming instantaneous and growing-season scale estimates. Under drought conditions, the association between SIF740 and yield improved with increasing temporal scale, with seasonal cumulative SIF740 showing the best performance. This study further elucidated the underlying mechanisms of the association between SIF and crop yield, providing a representative example for SIF-based regional yield estimation.

While global vegetation shows widespread greening under climate change, the corresponding impacts on Net Ecosystem Production (NEP)-particularly the divergent responses between drylands and non-drylands-remain poorly understood. In this study, based on the Normalized Difference Vegetation Index (NDVI), Solar-Induced Chlorophyll Fluorescence (SIF), Gross Primary Productivity (GPP), and NEP, we analyzed the spatiotemporal dynamics of vegetation photosynthesis and NEP across Eurasia from 1982 to 2018, and explored their differential responses to environmental factors, with particular focus on the contrasting mechanisms between drylands and non-drylands. The results showed that the rate of increase in NEP standardized anomalies (0.004 yr(-1), p < 0.05) lagged behind NDVI/SIF/GPP standardized anomalies (0.009 - 0.014 yr(-1), p < 0.01), particularly after 2000, and it was more pronounced in drylands. Spatially, vegetation photosynthesis and NEP changed asynchronously in nearly one-third of Eurasia, with a slightly higher proportion in non-dryland areas and the most obvious patterns in forests and croplands. Explainable machine learning analysis revealed that this asynchrony was primarily driven by interactions among temperature, vapor pressure deficit (VPD), and evapotranspiration. Vegetation photosynthesis was mainly influenced by evapotranspiration, while NEP was constrained by increased VPD. Moreover, in drylands, NEP was highly sensitive to atmospheric moisture, whereas in non-drylands, temperature fluctuations played a dominant role. These findings offer important insights for regional carbon sequestration management and climate change adaptation strategies.

The extraction of buildings from high-resolution remote sensing imagery (HRSI) is crucial across various applications and stands as a pivotal task in the field of remote sensing. While recent methods based on convolutional neural networks exhibit superior performance in building extraction from HRSI, there are still challenges, such as incomplete and missing extractions of buildings especially the building boundaries and the small buildings. To address these issues, we propose a Supervised Attention U-Net (SAU-Net), which combines a well-designed encoder and decoder. In the encoder, we incorporate a novel residual channel attention block and a densely connected multidilated convolutional block to enhance semantic features in the channel and spatial dimensions, respectively. In addition, in the decoder, we design a supervised attention block, which reconstructs fine-grained semantic features by systematically refining features in a supervised way and efficiently integrating feature maps from both the encoding and decoding stages. The efficacy of SAU-Net is evaluated using four HRSI datasets encompassing varying scenarios. The experimental results highlight that SAU-Net exhibits superior performance in building extraction, particularly excelling in the extraction of building boundaries and small buildings.

Gearbox is a key component of mechanical equipment, which has a complex structure, harsh working conditions, and a higher probability of failure. Therefore, gearbox fault diagnosis is vital to ensure the efficient operation of the whole mechanical equipment. However, traditional gearbox fault diagnosis methods mainly rely on manual feature extraction. To address this issue, a novel end-to-end fault diagnosis model that combines convolutional neural network (CNN), long short-term memory network (LSTM) and attention mechanism (AM) is proposed. Firstly, Spatial features are extracted from the original input data using CNNs, and then temporal features are extracted even further from the spatial features using LSTMs. The attention mechanism improves the network's attention to the global key features by assigning weights, and finally, the fault diagnosis results are derived from the fully connected layer and Softmax classifier. The experimental results show that the method is able to adaptively extract features and ensure higher diagnostic accuracy, validating the effectiveness of the proposed method.

A nonhomogeneous hidden Markov model (NHMM) consists of an unobservable nonhomogeneous Markov chain and an observable stochastic process. If the Markov assumption is changed to the condition that each hidden state in each step is related to the previous m states, then the extended model is called an m-th-order NHMM. Considering the calculation of entropy rate in the case of delayed averages, the generalized entropy ergodic theorem with almost-everywhere and ℒ_1 convergence for m-th-order NHMMs is studied in this paper. The establishment of some inequalities supports the proof of convergence for two types, and the obtained results maintain strict consistency with the corresponding classical results.

Quick and accurate extraction of un-collapsed buildings from post-disaster High-resolution Remote Sensing Images (HRSIs) is imperative for emergency response. Pre-disaster HRSIs could serve as auxiliary data for training models to expedite this extraction process. However, the effectiveness of models trained directly on pre-disaster HRSIs tends to diminish when applied to post-disaster scenarios, mainly due to the notable discrepancies between these datasets. The current popular approach to mitigate this issue involves aligning features from pre- and post-disaster images using an unsupervised domain adversarial learning framework. However, conventional methods often fall short in reducing the substantial disparity between pre- and post-disaster images, due to a lack of comprehensive alignment of category-level and multi-scale features. To overcome these limitations, we propose the Multi-scale Global and Category-attention Features Alignment Network (MGCAN). This novel approach further refines feature alignment strategies by concurrently aligning both multi-scale global and category-attention features, thus effectively narrowing the gap between pre- and post-disaster HRSIs. Extensive experiments have demonstrated that MGCAN significantly improves the accuracy of un-collapsed building extraction from post-disaster HRSIs. Moreover, compared to other state-of-the-art domain adversarial networks, MGCAN exhibits superior performance in different disaster scenarios.

Entropy measures the randomness or uncertainty of a stochastic process, and the entropy rate refers to the limit of the time average of entropy. The generalized entropy rate in the form of delayed averages can overcome the redundancy of initial information while ensuring stationarity. Therefore, it has better practical value. A Hidden Markov Model (HMM) contains two stochastic processes, a stochastic process in which all states can be observed and a Markov chain in which all states cannot be observed. The entropy rate is an important characteristic of HMMs. The transition matrix of a homogeneous HMM is unique, while a Nonhomogeneous Hidden Markov Model (NHMM) requires the transition matrices to be dependent on time variables. From the perspective of model structure, NHMMs are novel extensions of homogeneous HMMs. In this paper, the concepts of the generalized entropy rate and NHMMs are defined and fully explained, a strong limit theorem and limit properties of a norm are presented, and then generalized entropy ergodic theorems with an almost surely convergence for NHMMs are obtained. These results provide concise formulas for the computation and estimation of the generalized entropy rate for NHMMs.

In the era of industrial big data, data-driven rolling bearing fault diagnosis has received widespread attention. However, most of the current methods only extract features from a single dimension such as time domain, frequency domain or time-frequency domain, and this extraction method often leads to the inability to comprehensively extract the information features. Therefore, this paper considers multi-dimensional feature extraction of information, and thus proposes a fault diagnosis model based on two-branch convolutional neural network to improve the comprehensiveness of feature extraction and the accuracy of the classifier. The experimental validation results on the Case Western Reserve University bearing dataset show that the method can effectively improve the fault diagnosis performance, with an average accuracy of 99% on four datasets.

Lithium-ion batteries have been developed as the most widely used energy storage equipment and power batteries. State-of-charge (SOC) of the battery is a key index to evaluate the remaining range of electric vehicles. The existing SOC estimation methods perform unsatisfactorily on the multivariate long-time series data produced by battery operation. In this article, a graph deviation-based autoformer is proposed to realize accurate SOC estimation. The GD-based input module utilizes the graph structure with embedding vectors to extract spatial features and detect outliers. Encoder and decoder can acquire the temporal cycle dependencies in the data, using sequence decomposition block and auto-correlation mechanism instead of self-attention mechanism. Meanwhile, the online detection method can filter out noise and fluctuations to enhance the accuracy and robustness of the estimation results. The average values of normalized root mean square error, normalized mean absolute error, and R 2 achieved in the experiments are 0.0057, 0.0042, and 0.9995 respectively, which indicates superior performance on SOC estimation compared to other state-of-the-art methods. The method also has excellent generalization capability for new driving modes and new temperatures, which shows promising potential in practical applications.

Truck air pressure system (APS) fault detection is an important part of truck fault detection. For the high-dimensional imbalanced data obtained from truck fault detection, to improve the detection rate of APS fault, this paper proposes a modified Mahalanobis-Taguchi system (MMTS) algorithm. MMTS utilizes the ReliefF algorithm and signal-to-noise ratio to select features to construct Mahalanobis space, and uses particle swarm optimization algorithm to optimize the threshold determination method. Experiments show that MMTS has higher classification performance and is more suitable for truck APS fault detection than baseline algorithms for high-dimensional imbalanced data.

Due to the complexity of hydraulic systems, the data generated is prone to be high-dimensional and imbalanced, posing a challenge for fault diagnosis. This paper proposes a fast Mahalanobis classification system (FMCS). At the dimensionality reduction stage, smymetrical uncertainty and Mahalanobis kernel principal component analysis are used to construct a two-stage fast dimensionality reduction method. At the weight distribution stage, Mahalanobis distance is used to measure the similarity between the main feature and other features to determine the feature weights. At the fault diagnosis stage, Mahalanobis-Taguchi system is improved in threshold determination by atom search optimization to achieve fault diagnosis. Experiments show that FMCS has better classification performance (accuracy = 0.92, F1 = 0.94, G-mean = 0.91) and shorter computing time (4.9045 s) compared with 24 baseline algorithms. FMCS promotes the development of intelligent fault diagnosis for complex hydraulic systems.

Accurate knowledge of urban forest patterns contributes to well-managed urbanization, but accurate urban tree canopy mapping is still a challenging task because of the complexity of the urban structure. In this paper, a new method that combines double-branch U-NET with multi-temporal satellite images containing phenological information is introduced to accurately map urban tree canopies. Based on the constructed GF-2 image dataset, we developed a double-branch U-NET based on the feature fusion strategy using multi-temporal images to obtain an accuracy improvement with an IOU (intersection over union) of 2.3% and an F1-Score of 1.3% at the pixel level compared to the U-NET using mono-temporal images which performs best in existing studies for urban tree canopy mapping. We also found that the double-branch U-NET based on the feature fusion strategy has better accuracy than the early fusion strategy and decision fusion strategy in processing multi-temporal images for urban tree canopy mapping. We compared the impact of image combinations of different seasons on the urban tree canopy mapping task and found that the combination of summer and autumn images had the highest accuracy in the study area. Our research not only provides a high-precision urban tree canopy mapping method but also provides a direction to improve the accuracy both from the model structure and data potential when using deep learning for urban tree canopy mapping.

Imbalanced data classification is a challenge in data mining and machine learning. To improve the classification performance for imbalanced data, this paper proposes an imbalanced data classification algorithm based on the optimized Mahalanobis-Taguchi system (OMTS). At the feature selection stage, important feature variables are determined by four principles, namely maximizing mutual information between features and classes, minimizing mutual information between features, maximizing the initial classification accuracy, and selecting features that produce not only the local maximum or minimum of the difference between the mean Mahalanobis distances (MDs) of normal and abnormal samples but also the largest number of features. At the threshold determination stage, using the selected features, particle swarm optimization is used to determine the optimal threshold for classifying normal and abnormal samples according to the principle of maximizing classification accuracy. At the classification and discrimination stage, the samples are divided into two classes according to their MDs and optimal threshold. Experimental results show that OMTS obtains 0.92, 0.95, 0.81, 0.88, and 0.74 in accuracy on the Forest Type Mapping UCI, Fetal Health Classification, Connectionist Bench, Wine Quality, and Oil datasets, respectively, and has better classification performance than other algorithms.

Periplaneta americana L. (PA), a type of animal medicine, has been widely used for wound healing in clinical settings. In order to further investigate the bioactive wound healing substances in PA, crude PA protein-polysaccharide complexes were further purified by cellulose DE-52 and Sephadex G100 chromatography in succession. Among these isolated fractions, two fractions eluted by 0.3 M and 0.5 M NaCl with the higher yield, respectively named PaPPc2 and PaPPc3 respectively, were chosen for the wound healing experiments. Mediated by HPGPC, amino acid and monosaccharide composition analysis, circular dichroism spectrum, glycosylation type, FT-IR, and 1H NMR analysis, the characterization of PaPPc2 and PaPPc3 was implemented. And then, the benefits of PaPPcs to promote cell proliferation, migration, and tube formation of HUVECs were determined in vitro, indicated these fractions would facilitate angiogenesis. Finally, as proof of concept, PaPPc2 and PaPPc3 were employed to accelerate the acute wounds of diabetic mice, involving in increase blood vessels and the amounts of angiogenesis-related cytokines (α-SMA, VEGF, and CD31). In short, this study provides an experimental basis to demonstrate the protein-polysaccharide complexes of Periplaneta americana L. as its wound healing bioactive substances.

Based on the study of the relationship between the performance excellence model and economic efficiency, this paper proposes a theoretical model that technological innovation as a mediating variable helps to improve the economic efficiency of enterprises, and uses the research data of 1216 government quality award-winning enterprises in Zhejiang Province from 2015 to 2019 and combines the hierarchical regression method to conduct an empirical study on the constructed mediating effect model. The results show that the implementation of the performance excellence model helps to improve the economic efficiency of enterprises, while technological innovation mediates the positive relationship between the performance excellence model and the economic efficiency of enterprises. This paper combines the performance excellence model, technological innovation, and economic efficiency to remedy the limitations of previous studies and provide new ways for enterprises to improve their economic efficiency.

ETHNOPHARMACOLOGICAL RELEVANCE:Geranium wilfordii Maxim. (family Geraniaceae) has been used as a multifunctional ethnomedicine in China for more than 600 years. There have been records about the edible and medicinal use of G.wilfordii since the Ming Dynasty. Nowadays, G.wilfordii is included in Chinese Pharmacopoeia as the Geranii Herba, and has been widely used to treat rheumatism, infectious diseases, dermatosis and tumors. In addition, more than 30 kinds of Chinese patent medicines containing G.wilfordii have been marketed in China for its high medicinal value.AIM OF THE REVIEW:The present review systematically summarized the traditional uses, phytochemistry, pharmacology, quality control and toxicology of G.wilfordii, discussed the development potential of this plant and analyzed the shortcomings of existing research, in order to provide comprehensive scientific basis for the medical application and help interested researchers discover safe and medicinal natural products from G.wilfordii.MATERIALS AND METHODS:We performed a systematic search of G.wilfordii by collecting the publications between 1995 and 2021 via PubMed, Web of Science, SciFinder, Google Scholar, Chinese National Knowledge Infrastructure database, Wanfang database and other databases using the keywords including Geranium wilfordii Maxim., botany, traditional uses, Chinese patent medicines, patents, preparations, phytochemistry, biological activity, quality and toxicity. Besides, part of the information also came from Chinese botanical and medical monographs.RESULTS:So far, more than 90 components have been identified from G.wilfordii, among which tannins as well as flavonoids have attracted widespread concerns. Due to the multi-components nature, a wide range of beneficial biological activities of G.wilfordii have been demonstrated, including anti-inflammatory, antibacterial, antiviral, anti-tumor, antioxidative, hepatoprotective, gastrointestinal protective, immunoregulatory and uric acid-lowering effects.CONCLUSION:A long history of traditional uses and abundant pharmacochemical and pharmacological studies have shown that G.wilfordii is an important natural medicine, which exerts outstanding therapeutic effects especially in anti-inflammation and anti-microorganisms. However, the present researches on the active ingredients and the mechanism of action are not in-depth enough, and more evidence-based scientific studies are needed to lend credence to the traditional uses of G.wilfordii. Additionally, limited toxicology tests cannot objectively evaluate the safety of G.wilfordii, which needs to be further supplemented. Further, more attention needs to be paid to the selection of index components of quality control research.

Extreme climate-induced vegetation greenness decline significantly affects the stability of ecosystem function. Extreme climate events have occurred frequently in the recent 20 years and the possibility of climate anomalies is forecasted to increase in the future. But currently, the spatial and temporal response of episodic local vegetation decline to climate extremes at a global scale are still unclear. In this study, the detrend NDVI data was utilized as the indicator of vegetation growth, and a spatiotemporally contiguous recognition method was proposed to identify episodic large-scale vegetation decline events globally, subsequently, the spatiotemporal characteristics of these vegetation decline events and their interannual variation trends during 2000-2019 were explored. The results showed that (1) the spatiotemporally contiguous recognition method proposed by this paper was proven to be accurate in identifying the hotspot regions of large-scale vegetation decline. A total of 243 large-scale vegetation decline events were recognized globally during 2000-2019 drived by the method. (2) The global hotspots of large-scale vegetation decline were mainly distributed in the low-elevation areas at middle and low latitudes, especially at 15°S ~ 35°S, 15°N and 35°N, where covered north-western Africa, the Sahel, the Middle East, Central Asia, western India, the border of north-eastern China and Mongolia, western and south-central United States, northern Mexico, southern Africa, Australia, and southern and north-eastern South America. (3) Recent global episodic local vegetation decline has increased significantly since 2000, at the rate of 180,000 km2 of vegetation decline areas increasing per year. Particular, the severity of vegetation decline grew significantly since 2010 at the regions where covered the latitudes of approximately 15°N, 30°N and 65°N. Additionally, the severity of vegetation decline ranging from 20°S to 30°S weakened significantly since 2010. These findings were expected to provide the valuable scientific understanding for global vegetation decline and ecosystem responses to frequent climate extremes.