Search

As a core component of electric vehicles, the state of health (SOH) of lithium-ion battery has a direct impact on vehicle performance and safety. Existing data-driven models primarily focus on feature extraction, often overlooking the processing of multi-level redundant information and the utilization of multi-stage battery features. To address the issues, this paper proposes a novel data-driven method, named multi-stage mixed attention U-Net (MMAU-Net), for SOH estimation, which is based on both the phased learning and an encoder–decoder structure. First, the geometric knee-point division method is proposed to divide the battery life cycle into multiple stages, which allows the model to learn the distinctive features of battery degradation at each stage. Second, to adeptly capture degraded features and reduce redundant information, we propose a mixed attention U-Net model for the SOH prediction task, which is constructed upon the fundamental U-Net backbone and is enhanced with time step attention and feature attention modules. The experimental results validate the proposed method's feasibility and efficacy, demonstrating an acceptable performance across a spectrum of evaluative metrics. Consequently, this study offers a research within the domain of battery health management. [ABSTRACT FROM AUTHOR]

Focusing on the problems of uncertainty and carbon emissions in the manufacturing process, this paper studies the low-carbon-emission scheduling optimization problem. Firstly, the variations in workpiece processing time and delivery date are selected as the uncertainty factors. A low-carbon-emission scheduling model for uncertain job shops is constructed with the optimization objectives of the time index, carbon emission index, and robustness index. Secondly, an improved third-generation non-dominated sorting genetic algorithm (NSGA-III) is proposed. Based on the original NSGA-III algorithm, this algorithm introduces the state transition algorithm to perform state transformation, neighborhood sampling, selection update, and alternate rotation operations on the parent population, generating new candidate solutions. Finally, the scheduling model and the improved algorithm are applied to a workshop example. Through case study computation and result analysis, the feasibility and effectiveness of the model and algorithm in addressing the low-carbon-emission job shop scheduling problem under uncertainty are further verified. [ABSTRACT FROM AUTHOR]

Ocean wave energy is a new type of clean energy. To improve the power generation and wave energy conversion efficiency of the direct-drive wave power generation system, by addressing the issue of large output errors and poor system stability commonly associated with the currently used PID (proportional, integral, and derivative) control methods, this paper proposes a maximum power control method based on BP (back propagation) neural network PID control. Combined with Kalman filtering, this method not only achieves maximum power tracking but also reduces output ripple and tracking error, thereby enhancing the system's control quality. This study uses a permanent magnet linear generator as the power generation device, establishes a system dynamics model, and predicts the main frequency of irregular waves through the Fast Fourier Transform method. It designs a desired current tracking curve that meets the maximum power strategy. On this basis, a comparative analysis of the control accuracy and stability of three control methods is conducted. The simulation results show that the BP neural network PID control method improves power generation and exhibits better accuracy and stability. [ABSTRACT FROM AUTHOR]

Early and timely fire detection within enclosed spaces notably diminishes the response time for emergency aid. Previous methods have mostly focused on singularly detecting either fire or combustible materials, rarely integrating both aspects, leading to a lack of a comprehensive understanding of indoor fire scenarios. Moreover, traditional fire load assessment methods such as empirical formula-based assessment are time-consuming and face challenges in diverse scenarios. In this paper, we collected a novel dataset of fire and materials, the Material-Auxiliary Fire Dataset (MAFD), and combined this dataset with deep learning to achieve both fire and material recognition and segmentation in the indoor scene. A sophisticated deep learning model, Dual Attention Network (DANet), was specifically designed for image semantic segmentation to recognize fire and combustible material. The experimental analysis of our MAFD database demonstrated that our approach achieved an accuracy of 84.26% and outperformed the prevalent methods (e.g., PSPNet, CCNet, FCN, ISANet, OCRNet), making a significant contribution to fire safety technology and enhancing the capacity to identify potential hazards indoors. [ABSTRACT FROM AUTHOR]

Daqing Oilfield faces increasing reservoir heterogeneity after years of polymer flooding, limiting further enhanced oil recovery. A novel preformed particle gel (PPG) was developed by Daqing Oilfield due to the limited profile control ability of polymer flooding. The preformed particle gel possesses strong deformation ability and the ability to pass through pore throats. The PPG was developed considering the in situ reservoir conditions and combined effects of plugging and flooding. Thus, a heterogeneous system was prepared by mixing polymers and the PPG. In this study, we measured the related properties of the system and assessed its profile control ability and oil displacement performance after polymer flooding. The experimental results demonstrate that typical polymer flooding cannot improve oil recovery under current reservoir conditions. Thus, deep profile control technology should be applied to remediate the highly heterogeneous reservoir issue. PPG can considerably increase the viscosity and stability of the system. The heterogeneous system has a strong plugging ability during the subsequent water flooding stage and is suitable to inject into the medium and high permeability layers. The water absorption profile of the core samples is significantly improved due to the impact of the PPG on the plugging of the layer with high permeability and the liquid flow direction. The system displays an excellent profile control effect in the core with a permeability ratio (high permeability/low permeability) of two under the conditions of a PPG concentration of 500 mg/L and an injection volume of 0.5 PV. In the oil displacement experiment, the recovery efficiency was raised by 16.56% using the polymer system, leading to significant swept volume increment and oil recovery improvement. [ABSTRACT FROM AUTHOR]

Indoor fires can easily cause property damage and especially serious casualties. Early and timely fire detection helps firefighters make scientific judgments on the cause of fires, thereby effectively controlling fire accidents. However, most of the existing computer-vision-based fire detection methods are only able to detect a single case of flame or smoke. In this paper, a tailored deep-learning-based scheme is designed to simultaneously detect flame and smoke objects in indoor scenes. We adopt the semantic segmentation architecture DeepLabv3+ as the main model, which is an encoder-decoder architecture for both the detection and segmentation of fire objects. Within this, the key module, e.g., atrous convolution, is integrated into the architecture to improve image resolution and accurately locate targets. In addition, to solve the question of an insufficient indoor fire dataset, we prepare and construct a new annotated dataset named the 'Flame and Smoke Semantic Dataset (FSSD)', which includes extensive semantic information of fire objects and is collected from real indoor scenes and other fire sources. Experiments conducted on our FSSD database and the comparisons with state-of-the-art methods (FCN, PSPNet, and DeepLabv3), confirm the high performance of the proposed scheme with 91.53% aAcc, 89.67% mAcc, and 0.8018 mIoU, respectively. [ABSTRACT FROM AUTHOR]

After polymer flooding in Daqing Oilfield, the heterogeneity of the reservoir is enhanced, leading to the development of the dominant percolation channels, a significant issue with inefficient circulation, a substantial amount of displacement agents, and elevated cost. In order to further improve oil recovery, an adaptive oil displacement system (ASP-PPG) was proposed by combining preformed particle gel (PPG) with an alkali-surfactant-polymer system (ASP). This comprehensive study aims to assess the effectiveness of the adaptive oil displacement system (ASP-PPG) in improving the recovery efficiency of heterogeneous reservoirs after polymer flooding. The evaluation encompasses various critical aspects, including static performance tests, flow experiments, microscopic experiments, profile control experiments, and flooding experiments conducted on a four-layer heterogeneous physical model. The experimental results show that the adaptive system has robust stability, enhanced mobility, effective plugging capability, and profile improvement capability. Notably, the system demonstrates the remarkable ability to successfully pass through the core and effectively block the large pores, resulting in an 18.4% recovery incremental after polymer flooding. This improvement is reflected in the reduced oil saturation values in the ultra-high permeability, high permeability, medium, and low permeability layers, which are 5.09%, 7.01%, 13.81%, and 15.45%, respectively. The adaptive system effectively recovered the remaining oil in the low and medium permeability layers, providing a promising approach for improving the recovery factors under challenging reservoir conditions. [ABSTRACT FROM AUTHOR]

No study has been reported to use machine learning methods to predict the full-range test curves of circular CFST columns. In this paper, the long short-term memory (LSTM) network was introduced to calculate the axially compressive load–strain curves of the circular CFST columns according to an experiment database of limited scale. To improve the feasibility of input data for the recurrent neural network algorithm, data preprocessing methods and data configurations were discussed. The prediction results indicate that the LSTM network provides more accurate estimations compared with the artificial neural networks, random forest and support vector regression. Meanwhile, this method can be used to calculate the mechanical properties including the elastic modulus, ultimate bearing capacity, and the ductility of the columns with acceptable accuracy for engineering practice (the prediction error within 20%). For future research, it is expected that the machine learning method will be applied to predict the structural response of different members under various loading conditions. [ABSTRACT FROM AUTHOR]

Following the application of polymer flooding in Daqing Oilfield, the heterogeneity between different layers has intensified, resulting in the formation of more favorable seepage channels and cross-flow of displacement fluids. Consequently, the circulation efficiency has decreased, necessitating the exploration of methods to enhance oil recovery. This paper focuses on experimental research utilizing a newly developed precrosslinked particle gel (PPG) combined with alkali surfactant polymer (ASP) to create a heterogeneous composite system. This study aims to improve the efficiency of heterogeneous system flooding after polymer flooding. The addition of PPG particles enhances the viscoelasticity of the ASP system, reduces the interfacial tension between the heterogeneous system and crude oil, and provides excellent stability. The heterogeneous system has high resistance and residual resistance coefficients during the migration process in a long core model, achieving an improvement rate of up to 90.1% under the permeability ratio of 9 between high and low permeability layers. Employing heterogeneous system flooding after polymer flooding can increase oil recovery by 14.6%. Furthermore, the oil recovery rate of low permeability layers can reach 28.6%. The experimental results confirm that the application of PPG/ASP heterogeneous flooding after polymer flooding can effectively plug high-flow seepage channels and improve oil washing efficiency. These findings hold significant implications for further reservoir development after polymer flooding. [ABSTRACT FROM AUTHOR]

After tertiary recovery from the oilfields, improving the production of the remaining hydrocarbon is always challenging. To significantly improve oil recovery, a heterogeneous composite flooding system has been developed with preformed particle gels (PPG) and polymers according to the technical approach of plugging and flooding combination. In addition, an oil saturation monitoring device and a large-scale 3D physical model were designed to better evaluate the performance of the technique. The evaluation results show that the viscosity, stability, and elasticity of the heterogeneous composite flooding system are better than the single polymer system. In addition, both systems exhibit pseudoplastic fluid characteristics and follow the principle of shear thinning. The results of seepage experiments showed that PPG migrates alternately in porous media in the manner of "piling plugging-pressure increasing-deformation migration". The heterogeneous composite system can migrate to the depths of the oil layer, which improves the injection profile. In the visualization experiment, the heterogeneous composite system preferentially flowed into the high-permeability layer, which increased the seepage resistance and forced the subsequent fluid to flow into the medium and low permeability layers. The average saturation of the high, medium, and low permeability layers decreased by 4.74%, 9.51%, and 17.12%, respectively, and the recovery factor was further improved by 13.56% after the polymer flooding. [ABSTRACT FROM AUTHOR]

Ground penetration radar (GPR) technology has received in-depth analysis and rapid development in the field of civil engineering. GPR data analysis is one of the basic and challenging problems in this field. This research aims to conduct a comprehensive survey of the progress from 2015 to the present in GPR scanning tasks. More than 130 major publications are cited in this research covering different aspects of the research, including advanced data processing methods and a wide variety of applications. First, it briefly introduces the data collection of the GPR system and discusses the signal complexity in simulated/real scenes. Then, it reviews the main signal processing techniques used to interpret the GPR data. Subsequently, the latest GPR surveys are considered and divided according to four application domains, namely bridges, road pavements, underground utilities, and urban subsurface risks. Finally, the survey discusses the open challenges and directions for future research. [ABSTRACT FROM AUTHOR]

In China, the world's largest producer and consumer of lead-acid batteries (LABs), more than 3.6 million tons of waste lead-acid batteries (WLABs) are generated every year, yet only 30% of them can be recycled in a well-regulated manner, while the remaining 70% are recycled through informal channels, resulting in serious waste of resources and environmental pollution. More than half of the country's LAB consumers are e-car and e-bike owners. Based on the theoretical model of unified theory of acceptance and use of technology (UTAUT), this study examines and investigates the factors that affect consumers' participation in the regulated recycling of WLABs and finds that consumers' performance expectancy, social influence, and facilitating conditions can significantly increase their willingness to participate in regulated recycling, while effort expectancy can reduce such willingness. In addition, this paper also includes an analysis of moderating variables such as age and education. Finally, the paper points out the current lack of consumer-oriented recycling management measures in China and proposes policy recommendations in three aspects: system, channel, and incentive ones, to provide references for promoting the regulated recycling and sustainable use of WLABs. [ABSTRACT FROM AUTHOR]

Wave energy is a new type of clean energy. Aiming at a low wave energy density and small wave height in China's coastal areas, a tubular permanent magnet linear generator (PMLG) with a short stroke, small volume, and high power density is designed for wave power generation. Firstly, the generator's electromagnetic parameters are analyzed by the analytical method, and the magnetic circuit topology and basic structure of the generator are analyzed by the equivalent magnetic circuit method (EMCM). Then, the finite element method (FEM) is used to analyze the influence law of the generator's basic structural parameters on the output electromotive force (EMF) and its sinusoidal characteristics. The multi-factor and multi-level analysis is carried out based on the orthogonal test method to study the size parameters of the above analysis, and the optimal structure parameter combination for the generator is obtained. Finally, the prototype is trial-produced and tested for steady-state and transient performance to confirm the accuracy of the simulation calculations, and the output performance under no-load and load conditions is examined. The results show that both the optimized prototype's power density and the output EMF's sinusoidal properties have been improved under the proposed scheme. [ABSTRACT FROM AUTHOR]

Fusarium head blight (FHB) is a serious wheat disease caused by Fusarium graminearum (Fg) Schwabe. FHB can cause huge loss in wheat yield. In addition, trichothecene mycotoxins produced by Fg are harmful to the environment and humans. In our previous study, we obtained two mutants TPS1− and TPS2−. Neither of these mutants could synthesize trehalose, and they produced fewer mycotoxins. To understand the complex interaction between Fg and wheat, we systematically analyzed the metabolic responses of FHB-susceptible and -resistant wheat to ddH2O, the TPS− mutants and wild type (WT) using NMR combined with multivariate analysis. More than 40 metabolites were identified in wheat extracts including sugars, amino acids, organic acids, choline metabolites and other metabolites. When infected by Fg, FHB-resistant and -susceptible wheat plants showed different metabolic responses. For FHB-resistant wheat, there were clear metabolic differences between inoculation with mutants (TPS1−/TPS2−) and with ddH2O/WT. For the susceptible wheat, there were obvious metabolic differences between inoculation with mutant (TPS1−/TPS2−) and inoculation with ddH2O; however, there were no significant metabolic differences between inoculation with TPS− mutants and with WT. Specifically, compared with ddH2O, resistant wheat increased the levels of Phe, p-hydroxy cinnamic acid (p-HCA), and chlorogenic acid in response to TPS− mutants; however, susceptible wheat did not. Shikimate-mediated secondary metabolism was activated in the FHB-resistant wheat to inhibit the growth of Fg and reduce the production of mycotoxins. These results can be helpful for the development of FHB-resistant wheat varieties, although the molecular relationship between the trehalose biosynthetic pathway in Fg and shikimate-mediated secondary metabolism in wheat remains to be further studied. [ABSTRACT FROM AUTHOR]

Cavity under urban roads has increasingly become a huge threat to traffic safety. This paper aims to study cavity morphology characteristics and proposes a deep learning (DL)-based morphology classification method using the 3D ground-penetrating radar (GPR) data. Fine-tuning technology in DL can be used in some cases with relatively few samples, but in the case of only one or very few samples, there will still be overfitting problems. To address this issue, a simple and general framework, few-shot learning (FSL), is first employed for the cavity classification tasks, based on which a classifier learns to identify new classes given only very few examples. We adopt a relation network (RelationNet) as the FSL framework, which consists of an embedding module and a relation module. Furthermore, the proposed method is simpler and faster because it does not require pre-training or fine-tuning. The experimental results are validated using the 3D GPR road modeling data obtained from the gprMax3D system. The proposed method is compared with other FSL networks such as ProtoNet, R2D2, and BaseLine relative to different benchmarks. The experimental results demonstrate that this method outperforms other prior approaches, and its average accuracy reaches 97.328% in a four-way five-shot problem using few support samples. [ABSTRACT FROM AUTHOR]

The presence of liver cancer stem cells (LCSCs) is one of the reasons for the treatment failure of hepatocellular carcinoma (HCC). For LCSCs, one of their prominent features is metabolism plasticity, which depends on transporters and ion channels to exchange metabolites and ions. The K+ channel protein KCNN4 (Potassium Calcium-Activated Channel Subfamily N Member 4) has been reported to promote cell metabolism and malignant progression of HCCs, but its influence on LCSC stemness has remained unclear. Here, we demonstrated that KCNN4 was highly expressed in L-CSCs by RT-PCR and Western blot. Then, we illustrated that KCNN4 promoted the stemness of HC-C cells by CD133+CD44+ LCSC subpopulation ratio analysis, in vitro stemness transcription factor detection, and sphere formation assay, as well as in vivo orthotopic liver tumor formation and limiting dilution tumorigenesis assays. We also showed that KCNN4 enhanced the glucose metabolism in LCSCs by metabolic enzyme detections and seahorse analysis, and the KCNN4-promoted increase in LCSC ratios was abolished by glycolysis inhibitor 2-DG or OXPHOS inhibitor oligomycin. Collectively, our results suggested that KCNN4 promoted LCSC stemness via enhancing glucose metabolism, and that KCNN4 would be a potential molecular target for eliminating LCSCs in HCC. [ABSTRACT FROM AUTHOR]

Achieving deep learning-based bearing fault diagnosis heavily relies on large labeled training samples. However, in real industry applications, labeled data are scarce or even impossible to obtain. In this study, we addressed a challenging few-shot bearing fault diagnosis problem with few or no training labeled samples of novel categories. To tackle this problem, we considered a semi-supervised prototype network based on few-shot bearing fault diagnosis with pseudo-labels. The existing prototypical networks with pseudo-label methods train a pseudo label model to label unlabeled samples using high-dimensional labeled data, which cannot eliminate the instability of the pseudo-label model caused by dimensional labeled features. To mitigate this issue, we used kernel principal component analysis to reduce the dimensions of and remove redundant information from high-dimensional data. Specifically, we used the pseudo-label prediction algorithm with probability distance to label unlabeled samples, aiming to improve the labeling accuracy. We applied two well-known bearing data sets for the validation experiments with symmetry parameters. The findings illustrated that the classification accuracy of the proposed method is higher than that of other existing methods. [ABSTRACT FROM AUTHOR]

Featured Application: The sensor developed in the paper can be used for high sensitivity temperature detection of power system, biomedicine, aerospace, and so on. A high temperature (up to 950 °C) sensor was proposed and demonstrated based on a micro taper in-line fiber Mach–Zehnder interferometer (MZI) structure. The fiber MZI structure comprises a single mode fiber (SMF) with two micro tapers along its longitudinal direction. An annealing at 1000 °C was applied to the fiber sensor to stabilize the temperature measurement. The experimental results showed that the sensitivity was 0.114 nm/°C and 0.116 nm/°C for the heating and cooling cycles, respectively, and, after two days, the sensor still had a sensitivity of 0.11 nm/°C, showing a good stability of the sensor. A probe-type fiber MZI was designed by cutting the sandwiched SMF, which has good linear temperature responses of 0.113 nm/°C over a large temperature range from 89 to 950 °C. The probe-type fiber MZI temperature sensor was independent to the surrounding refractive index (RI) and immune to strain. The developed sensor has a wide application prospect in the fields of high temperature hot gas flow, as well as oil and gas field development. [ABSTRACT FROM AUTHOR]

With the increasingly stringent CO2 emission regulations, the degree of strengthening of the engines is increasing. Under high-pressure conditions, the airway throat parts of the intake and exhaust systems have a great influence on the flow loss of the diesel engine. The reasonable distribution of the throat area of the intake and exhaust ports in the limited cylinder headspace is key to improving the performance of supercharged engines. This study took a large-bore, high-pressure ratio diesel engine as the research object. Firstly, the three-dimensional (3D) flow simulation method was used to reveal the influence law of different throat areas on the engine intake and exhaust flow under steady-state conditions, and a steady-flow test bench was built to verify the accuracy of the simulation model and law. Secondly, based on the 3D steady-state calculation and test results, a more accurate one-dimensional simulation model was constructed, and a joint optimization simulation platform was established based on the dynamic data link library. On this basis, the mathematical description of the multi-objective optimization of airway throat size was established using machine learning methods, such as a genetic algorithm, the design domain and boundary conditions of variable parameters were clarified, and the collaborative optimization objective of integrated flow coefficient and flow loss is proposed to achieve the fast and accurate optimization of intake and exhaust throat diameters. [ABSTRACT FROM AUTHOR]

Electromagnetic linear actuators, as key executive components, have a vital impact on the performance of fully flexible variable valve trains. Considering that the conventional moving coil electromagnetic linear actuator (MCELA) has the disadvantages of low force density and a lack of end-passive self-holding ability, a novel composited electromagnetic linear actuator (CELA) is proposed by combining the performance advantages of MCELA and moving iron electromagnetic linear actuator (MIELA) in this work. Firstly, the structure and magnetic circuit design scheme of the proposed actuator are introduced and the finite element simulation model is established. The magnetic field distribution and force characteristics of the actuators are assessed by finite element simulation. Secondly, the construction of the prototype of the actuator is outlined, based on which the feasibility of the design scheme and the steady-state performance of the actuator are verified. Finally, the coordinated control strategy is proposed to realize the multi motion coordination control of the actuator. The research results show that the maximum starting force of the CELA with the end-passive self-holding ability is 574.92 N while the holding force can approach 229.25 N. Moreover, the CELA is proven to have excellent dynamic characteristics and control precision under different motion modes and to have an improved adaptability to the complex working conditions of internal combustion engines. [ABSTRACT FROM AUTHOR]

Large deep foundation pits are usually in a complex environment, so their surface deformation tends to show a stable rising trend with a small range of fluctuation, which brings certain difficulty to the prediction work. Therefore, in this study we proposed a nonlinear autoregressive exogenous (NARX) prediction method based on empirical wavelet transform (EWT) pretreatment is proposed for this feature. Firstly, EWT is used to conduct adaptive decomposition of the measured deformation data and extract the modal signal components with characteristic differences. Secondly, the main components affecting the deformation of the foundation pit are analyzed as a part of the external input. Then, we established a NARX prediction model for different components. Finally, all predicted values are superpositioned to obtain a total value, and the result is compared with the predicted results of the nonlinear autoregressive (NAR) model, empirical mode decomposition-nonlinear autoregressive (EMD-NAR) model, EWT-NAR model, NARX model, EMD-NARX model and EWT-NARX model. The results showed that, compared with the EWT-NAR and EWT-NARX models, the EWT-NARX model reduced the mean square error of KD25 by 91.35%, indicating that the feature of introducing external input makes NARX more suitable for combining with the EWT method. Meanwhile, compared with the EMD-NAR and EWT-NAR models, the introduction of the NARX model reduced the mean square error of KD25 by 78.58% and 95.71%, indicating that EWT had better modal decomposition capability than EMD. [ABSTRACT FROM AUTHOR]