Poor Versatility Research Articles

A Combined Data-Driven and Model-Based Algorithm for Accurate Battery Thermal Runaway Warning.

With the increasingly widespread application of large-scale energy storage battery systems, the demand for battery safety is rising. Research on how to detect battery anomalies early and reduce the occurrence of thermal runaway (TR) accidents has become particularly important. Existing research on battery TR warning algorithms can be mainly divided into two categories: model-driven and data-driven methods. However, the common model-driven methods are often of high complexity, with poor versatility and low early warning capability; and the common data-driven methods are mostly based on neural networks, requiring substantial training costs, with better early warning capabilities but higher false alarm probabilities. To address the limitations of existing works, this paper proposes a combined data-driven and model-based algorithm for accurate battery TR warnings. Specifically, the K-Means algorithm serves as the data-driven module, capturing outliers in battery data, and the Bernardi equation serves as the model-driven module used to evaluate battery temperature. Ultimately, the outputs of the weighted model-driven module and data-driven module are combined to comprehensively assess whether the battery is abnormal. The proposed algorithm combines the advantages of model-driven and data-driven approaches, achieving a 25 min advance warning for thermal runaway, with a significantly reduced probability of false alarms.

Optical information hiding for different surface images.

Optical hiding often requires the selection of specific artificial optical components as carriers, which results in poor versatility of the carriers and high costs for the hiding system. To conceal secret information on different surfaces such as metal, wood, and paper, we propose an optical information hiding method. In this method, we use images of surfaces, whose grayscale histograms have the characteristic of symmetric distribution. Based on this characteristic, we first scramble the surface image, and then adjust part of the gray value of the surface image to the complementary value to embed the secret information into a scrambled surface image to generate a key image. In the extraction process, a projector is used to reproduce the scrambled surface image and the key image, which are then incoherently superimposed to extract the secret information using the human visual system. The extraction process does not require complex optical knowledge and is simple and feasible. Simulation experiments and optical experiments indicate that this method is applicable in practice and possesses good security and imperceptibility. Furthermore, we prove the reliability of this method by embedding secret information in different surface images, demonstrating the potential application of more surface images in the field of optical information hiding. Finally, we discuss the applicability of surface information images and analyze the imperceptibility of key images.

Design and Experiment of Automatic Transport System for Planting Plate in Plant Factory

In the plant factories using stereoscopic cultivation systems, the cultivation plate transport equipment is an essential component of production. However, there are problems, such as high labor intensity, low levels of automation, and poor versatility of existing solutions, that can affect the efficiency of cultivation plate transport processes. To address these issues, this study designed a cultivation plate transport system that can automatically input and output cultivation plates, and can flexibly adjust its structure to accommodate different cultivation frame heights. We elucidated the working principles of the transport system and carried out structural design and parameter calculation for the lift cart, input actuator, and output actuator. In the input process, we used dynamic simulation technology to obtain an optimum propulsion speed of 0.3 m·s−1. In the output process, we used finite element numerical simulation technology to verify that the deformation of the cultivation plate and the maximum stress suffered by it could meet the operational requirements. Finally, operation and performance experiments showed that, under the condition of satisfying the allowable amount of positioning error in the horizontal and vertical directions, the horizontal operation speed was 0.2 m·s−1, the maximum positioning error was 2.87 mm, the vertical operation speed was 0.3 m·s−1, and the maximum positioning error was 1.34 mm. Accordingly, the success rate of the transport system was 92.5–96.0%, and the operational efficiency was 176–317 plates/h. These results proved that the transport system could meet the operational requirements and provide feasible solutions for the automation of plant factory transport equipment.

Improvement of the enantioselectivities of Mandyphos‐Pd complexes by introducing groups with different steric and electronic properties

AbstractBACKGROUNDMandyphos‐Pd has been proven to be a good chiral extractant to enantioseparate 3‐chlorophenylglycine enantiomers with a separation factor of 2.64. However, this extractant is not suitable to enantioseparate other amino acids with separation factors of <2.0. The poor versatility has limited its practical application in chiral extraction.RESULTSMandyphos derivatives with different steric and electronic properties were synthesized and their enantioseparation properties toward amino acid and mandelic acid enantiomers were investigated. The results suggested that substituent groups with different steric properties affect the extraction properties, probably by adjusting the angles of tetra‐coordinated palladium complexes. And the substituent groups with different electronic properties affect the extraction properties mainly by adjusting the π‐π interactions/steric hindrance effects between Mandyphos‐Pd and substrates. After optimization, the highest separation factors for phenylalanine, homophenylalanine, 3‐chloro‐phenylglycine, 4‐nitro‐phenylalanine, mandelic acid and 2‐Cl‐mandelic acid were 2.10, 2.77, 3.51, 2.42, 2.24 and 5.19, respectively. The enantioseparation mechanisms revealed that the coordination interactions between metal ions and COO−/NH2, π‐π interactions or steric hindrance effects between aryls are the main forces acting in chiral recognition.CONCLUSIONThe enantioselectivities of the Mandyphos‐Pd complex are improved by introducing groups with suitable steric and electronic properties. Structural modification is a useful method to improve the enantioseparation efficiencies of chiral extractants. The results of this work will supply important theoretical guidance and reference for the design and modification of novel diphosphine ligands in future work. © 2024 Society of Chemical Industry (SCI).

Rice leaf disease detection based on enhanced feature fusion and target adaptation

AbstractIntelligent rice disease recognition methods based on deep neural networks can predict the degree of disease on the basis of, for example, the number of disease spots on an image, so that preventive measures can be taken. Currently, intelligent recognition methods for rice diseases suffer from the disadvantages of poor versatility and low accuracy. This paper uses eight common image classification networks to classify and identify four rice diseases. ResNet50 was selected as the feature extraction network and an enhanced feature fusion and target adaptive network (EFFTAN), referred to as EFFTAN, is proposed. The EFFTAN was used to detect four rice spot diseases in the rice leaf disease image samples dataset; the mean average precision of the final detection was 95.3%, and effective detection was also achieved for the dense spot features.

Electrolytic capacitor surface defect detection based on deep convolution neural network

The existing methods for detecting surface defects in electrolytic capacitors are typically based on conventional machine vision, with limited feature extraction capabilities, poor versatility, slow detection speed, and the inability to achieve accurate and real-time defect detection. In this study, a real-time object detection algorithm based on an improved single shot multibox detector (SSD) is proposed to achieve omnidirectional surface defect detection of electrolytic capacitors. First, an electrolytic capacitor surface image acquisition device was established to capture omnidirectional surface images of the capacitors, and an electrolytic capacitor surface defect dataset was created. Next, the visual geometry group (VGG)-16 network structure was replaced with the MobileNetv2 network structure, effectively reducing the model’s parameter count and improving inference speed. Moreover, the Multibox Loss function was replaced with the Focal Loss function to increase the model’s attention to difficult-to-classify samples and improve model accuracy. Additionally, a transfer learning network model was designed to apply the model to electrolytic capacitors of different colors using small sample learning. Finally, the performance of the improved network model was tested on a dataset of electrolytic capacitor surface defects. The experimental results demonstrate that the parameters quantity of improved model is 3.50 M, the mAP value reaches 92.67 %, which is improved by 2.54 %, and the Macro-F1 value reaches 92.15 %, which is 11.32 % higher than that before improvement. Thus, the proposed improved SSD model provides a theoretical basis and technical prerequisites for automated and intelligent surface defect detection in electrolytic capacitors.

Suitable-Matching Areas’ Selection Method Based on Multi-Level Saliency

Scene-matching navigation is one of the essential technologies for achieving precise navigation in satellite-denied environments. Selecting suitable-matching areas is crucial for planning trajectory and reducing yaw. Most traditional selection methods of suitable-matching areas use hierarchical screening based on multiple feature indicators. However, these methods rarely consider the interrelationship between different feature indicators and use the same set of screening thresholds for different categories of images, which has poor versatility and can easily cause mis-selection and omission. To solve this problem, a suitable-matching areas’ selection method based on multi-level saliency is proposed. The matching performance score is obtained by fusing several segmentation levels’ salient feature extraction results and performing weighted calculations with the sub-image edge density. Compared with the hierarchical screening methods, the matching performance of the candidate areas selected by our algorithm is at least 22.2% higher, and it also has a better matching ability in different scene categories. In addition, the number of missed and wrong selections is significantly reduced. The average matching accuracy of the top three areas selected by our method reached 0.8549, 0.7993, and 0.7803, respectively, under the verification of multiple matching algorithms. Experimental results show this paper’s suitable-matching areas’ selection method is more robust.

Recent Patents on Red Jujube Harvester

Abstract: As the demand for jujubes grows, so does the acreage of jujubes. However, the main method of jujube picking is traditional manual picking, which is the most time-consuming and labor- intensive in the production chain. Through the analysis and discussion of the patents on the jujube harvester, some valuable conclusions are drawn to predict the research and development of the jujube harvester in the future. An overview of various representative patents of the current jujube harvester is provided in this paper. A large number of patents of jujube harvesting devices have been summarized, the characteristics of these patents have been analyzed and compared, and the further development trend of jujube harvesting machines has been developed. The research on the multifunction, intelligence and versatility of the jujube harvester is helpful to solve the problems of low efficiency, low quality, and poor versatility of the jujube harvester.

Preparation strategy of bimetallic MOF hollow photocatalysts for hydrogen evolution

Hollow structure materials have been proved to be excellent photocatalysts due to their advantages in the separation of electron-hole pairs. However, the synthesis of such materials is complex, time-consuming, and poor versatility. In this study, Cu-DHTP/ZIF-67 (DHTP = 2,5-dihydroxyterephthalic acid) hollow materials containing both Co and Cu were designed and synthesized by a simple one-step etching and ligand exchange strategy. It is worth noting that the bimetallic hollow catalyst can be synthesized in the presence of both Cu(Ⅱ) and DHTP within 5 min at room temperature. The synthesized hollow bimetallic catalyst has improved photoresponse performance and electron-hole pair separation ability, which results in the best catalytic performance compared with DHTP/ZIF-67, Cu/ZIF-67, and ZIF-67 when used in the photocatalytic water splitting reaction. The hydrogen evolution yield of Cu-DHTP/ZIF-67 can reach 39.7 mmol g−1, which is 6 times than that of pure ZIF-67. This study demonstrates the promising potential of Cu-DHTP/ZIF-67 photocatalyst in the field of photocatalytic water reduction and hydrogen evolution, and provides a new idea for designing and preparing more hollow structure bimetallic photocatalysts with excellent performance.

The Advent of Domain Adaptation into Artificial Intelligence for Gastrointestinal Endoscopy and Medical Imaging.

Artificial intelligence (AI) is a subfield of computer science that aims to implement computer systems that perform tasks that generally require human learning, reasoning, and perceptual abilities. AI is widely used in the medical field. The interpretation of medical images requires considerable effort, time, and skill. AI-aided interpretations, such as automated abnormal lesion detection and image classification, are promising areas of AI. However, when images with different characteristics are extracted, depending on the manufacturer and imaging environment, a so-called domain shift problem occurs in which the developed AI has a poor versatility. Domain adaptation is used to address this problem. Domain adaptation is a tool that generates a newly converted image which is suitable for other domains. It has also shown promise in reducing the differences in appearance among the images collected from different devices. Domain adaptation is expected to improve the reading accuracy of AI for heterogeneous image distributions in gastrointestinal (GI) endoscopy and medical image analyses. In this paper, we review the history and basic characteristics of domain shift and domain adaptation. We also address their use in gastrointestinal endoscopy and the medical field more generally through published examples, perspectives, and future directions.

Electroactive nanoinjection platform for intracellular delivery and gene silencing

BackgroundNanoinjection—the process of intracellular delivery using vertically configured nanostructures—is a physical route that efficiently negotiates the plasma membrane, with minimal perturbation and toxicity to the cells. Nanoinjection, as a physical membrane-disruption-mediated approach, overcomes challenges associated with conventional carrier-mediated approaches such as safety issues (with viral carriers), genotoxicity, limited packaging capacity, low levels of endosomal escape, and poor versatility for cell and cargo types. Yet, despite the implementation of nanoinjection tools and their assisted analogues in diverse cellular manipulations, there are still substantial challenges in harnessing these platforms to gain access into cell interiors with much greater precision without damaging the cell’s intricate structure. Here, we propose a non-viral, low-voltage, and reusable electroactive nanoinjection (ENI) platform based on vertically configured conductive nanotubes (NTs) that allows for rapid influx of targeted biomolecular cargos into the intracellular environment, and for successful gene silencing. The localization of electric fields at the tight interface between conductive NTs and the cell membrane drastically lowers the voltage required for cargo delivery into the cells, from kilovolts (for bulk electroporation) to only ≤ 10 V; this enhances the fine control over membrane disruption and mitigates the problem of high cell mortality experienced by conventional electroporation.ResultsThrough both theoretical simulations and experiments, we demonstrate the capability of the ENI platform to locally perforate GPE-86 mouse fibroblast cells and efficiently inject a diverse range of membrane-impermeable biomolecules with efficacy of 62.5% (antibody), 55.5% (mRNA), and 51.8% (plasmid DNA), with minimal impact on cells’ viability post nanoscale-EP (> 90%). We also show gene silencing through the delivery of siRNA that targets TRIOBP, yielding gene knockdown efficiency of 41.3%.ConclusionsWe anticipate that our non-viral and low-voltage ENI platform is set to offer a new safe path to intracellular delivery with broader selection of cargo and cell types, and will open opportunities for advanced ex vivo cell engineering and gene silencing.Graphical abstract

Multi-person Action Recognition Based on the HARNet Network in the Park Environment

Action recognition is crucial to public security, playing an important role in pedestrian protection. Recently, numerous approaches based on deep learning models have been reported for action recognition, which mainly has problems such as poor versatility, not good robustness, low recognition effects for low resolution, and occlusion in outdoor environments. This paper addresses this issue and aims to develop an approach for multi-person action recognition in the park environment. Frame extraction operation is first used to extract images from surveillance video in real-time, then YOLOv5 is adopted to detect and segment human targets from images. Thirdly, the AlphaPose algorithm is used to estimate the keypoint heatmap of the human body according to human targets. Finally, the heatmaps are preprocessed and then fused and classified by deep learning networks to complete the multi-person action recognition. In addition, an optimization method, i.e., Heatmap-based Action Recognition Network (HARNet), for ResNet is proposed in order to overcome the difficulty of information extraction from heatmaps, where the attention mechanism is considered. The approach presented here was tested on park surveillance video with multiple different scenarios, and the corresponding results verified its performance.

Cushioning Performance of the Biomimetic Cobweb Cushioning Silicone Pad.

At present, the packing method of "plastic bag-buffer packing-packing paper box" is adopted for bearing packaging. However, the common packing method has a poor packing effect and poor versatility. In this study, a new biomimetic cobweb cushion is proposed to solve the problem of insufficient cushioning capacity of high-precision bearing cushion packaging pads. First, according to the nature of cobweb form, the cobweb cushion structure configuration is determined. Next, based on the structure of the cushion and the relationship between the parameters of radial thread and spiral thread, a mechanical and target optimization model is established. The stress nephogram of bearing and the cobweb cushion are analyzed under three drop heights of 381, 610, and 700 mm, in the finite element simulation software to ensure that the maximum bearings stress is not beyond the material yield strength. Via the 3D printing technology, a cobweb cushion shell cast is made. Drop tests of the bearing were performed, and the results were verified with the finite element simulation analysis. This research can provide technical support for the protection of high-precision bearings from accidental drops during transportation.

Covalent Organic Framework Nanosheets for Fluorescence Quantification of Peptide.

Rapid and sensitive quantification of peptides plays an important role in clinical diagnosis. Fluorescence assay is one of the most promising peptide detection tools, but it relies on intrinsic fluorescence or additional derivatization, resulting in poor versatility. Covalent organic frameworks (COFs) have shown a good application prospect in the field of fluorescence detection, but their application scope is limited to heavy metal ions and some small polar organic molecules. Herein, we report the application of COFs nanosheet for fluorescence detection of peptides. Fluorescent sp2 acrylonitrile-linked COFs nanosheets (TTAN-CON) were prepared by water-assisted ultrasonic exfoliation which performed with excellent fluorescence properties with Stokes shifts of 146 nm and fluorescence quantum yield of up to 24.45%. Compared to the bulk fluorescent COFs, exfoliated CONs films performed with better stability of fluorescence signal in solution. We found the fluorescence of TTAN-CON can be effectively quenched by hydrophobic peptides at a very rapid rate (less than 5 min per sample). TTAN-CON presented good sensitivity and selectivity for hydrophobic peptides detection via the static and dynamic joint quenching mechanism. TTAN-CON was further used to detect NLLGLIEAK and ProGRP31-98, two target peptide fragments of lung cancer biomarker ProGRP. The fluorescence intensities of TTAN-CON were negative linearly correlated with the amounts of hydrophobic NLLGLIEAK over the range of 5-1000 ng/mL with the correlation coefficients over 0.99, and the limit of detection was 1.67 ng/mL, displaying higher sensitivity and convenience than traditional optical methods. What's more, the quantification of ProGRP31-98 was achieved by the quantification of hydrophobic peptides in its enzyme hydrolysis products. We anticipate COFs nanosheets to be a universal fluorescence detection work-box for peptides biomarkers with clinical significance.

Microcontroller Unit Chip Temperature Fingerprint Informed Machine Learning for IIoT Intrusion Detection

Physics-informed learning for industrial Internet is essential especially to safety issues. Consequently, various methods have been developed to conduct Industrial Internet of Things (IIoT) intrusion detection. However, the conventional methods usually require the help of auxiliary equipment (e.g., spectrum analyzers, log-periodic antennas), which proves to be unsuitable for general IIoT systems due to their poor versatility. Facing the dilemma mentioned above, this article proposes a microcontroller unit (MCU) chip temperature fingerprint informed machine learning method, called MTID, for IIoT intrusion detection. Specifically, first, the node's MCU temperature sequence is recorded and the relationship between the temperature sequence and the computational complexity of the node is analyzed. Then, we calculate the temperature residuals and construct a temperature residuals dataset. Finally, to identify the security status of the nodes, a self-encoder-based intrusion detection model is constructed. Furthermore, to ensure the model's applicability under the diversified deployment environment of IIoT systems, an online incremental training method is developed and applied. In the end, we use the Raspberry Pi 4B for experimental analysis when testing the performance of MTID. The results show that the accuracy of MTID for intrusion detection reaches 89%, which also demonstrates the feasibility of the intrusion detection method based on MCU temperature.

Research on Signal Transmission Device of Multi Adaptive Coiled Tube Measurement and Control System

As an important technology for oilfield development and oil and gas reservoir reconstruction, the demand for coiled tubing logging, perforating, fracturing and acidizing and other process operations has increased dramatically, and downhole testing and control operations must be completed through downhole data acquisition and transmission and control signals. However, at present, the special cable coiled tubing roller device and logging roller operating machine mainly used at home and abroad have problems such as single function, high cost, complex structure, and poor versatility. Therefore, based on the structural characteristics of the conventional coiled tubing machine, through the analysis of the existing coiled tubing testing and control data signal transmission system, the research and design of the signal transmission device for the multi adaptive coiled tubing measurement and control system was completed, and a kind of signal transmission device for the multi adaptive coiled tubing signal testing, acquisition, transmission and feedback control system with strong versatility and high transmission efficiency was studied and designed, It eliminates the high cost of traditional equipment customized processing and manufacturing and the high labor intensity of operation and maintenance operations caused by the single adaptability of operation process, which are caused by complicated structures. It is of great significance for realizing the transformation and upgrading of the multi-function of intelligent measurement and control integration of coiled tubing machine, and is extremely important for the innovation of the digital network real-time monitoring system for future oilfield development.

BEARNet: A Novel Buildings Edge-Aware Refined Network for Building Extraction From High-Resolution Remote Sensing Images

Accurately extracting buildings from high-resolution remote sensing images is important to obtain urban information, and promote the development of smart cities. At present, the knowledge-driven building extraction methods mostly rely on building prior knowledge of manual design, resulting in low automation and poor versatility. The data-driven methods usually rely too much on training samples, resulting in insufficient pertinence for the features of building extraction and low generalization ability of the model. Therefore, this study proposes a novel buildings edge-aware refined deep learning network (BEARNet) for building extraction from high-resolution remote sensing images. The network takes the building edge as a priori knowledge, learns the building edge features by decoupling the building body and edge, and further optimizes the network by combining the multi-objective loss function to strengthen the pertinence of building edge features extraction. Experimental results show that on the WHU building dataset, which is less difficult to extract buildings, compared with other methods, BEARNet has the highest Precision, F1, IoU and OA values of 97.70%, 97.42%, 95.3% and 98.67%. On the Massachusetts building dataset, where building extraction is difficult, BEARNet has the highest Precision, Recall, F1, IoU and OA compared to other methods, with values of 84.92%, 85.27%, 85.09%, 75.82% and 93.99%, respectively. Our proposed method is more accurate in extracting complex shapes and dense small-scale buildings, and the building edges are more refined and complete.

A Temperature Monitoring Method For Power Electronic Converter Based on Infrared Image and Object Detection Algorithm

Power electronic converters are more and more widely used, and abnormal temperature of converter components is the most important factor in converter failure. To improve the reliability of the converter design, it is necessary to monitor the temperature of key components in the converter during the prototype test stage. The temperature measurement method of infrared thermal images has rich temperature information, wide detection range, and does not affect the original circuit design. However, in the current automatic temperature measurement methods, it is necessary to manually establish a standard matching template for the infrared thermal image of the circuit to be tested, which indicates a large workload and poor versatility. This paper proposes a method for fully automatic temperature monitoring of converter components. This method is based on a deep learning object detection algorithm, which can automatically identify the type of converter components, obtain partial infrared thermal images of components through heterogeneous image registration, achieve accurate component temperature monitoring and facilitate the converter state monitoring and fault detection. The advantages of this method are: 1) there is no need to manually establish a standard template for each converter; 2) it can monitor the converter temperature without manual intervention; 3) combining the temperature information and circuit prior knowledge, the state monitoring and fault diagnosis can further be realized. The experimental results also verify the feasibility and accuracy of this method.

Lossless enrichment of trace analytes in levitating droplets for multiphase and multiplex detection

Concentrating a trace amount of molecules from liquids, solid objects, or the gas phase and delivering them to a localized area are crucial for almost any trace analyte detection device. Analytes within a liquid droplet resting on micro/nanostructured surfaces with liquid-repellent coatings can be concentrated during solvent evaporation. However, these coatings suffer from complex manufacturing procedures, poor versatility, and limited analyte enrichment efficiency. Here, we report on the use of an acoustic levitation platform to losslessly concentrate the analyte molecules dissolved in any volatile liquid, attached to solid objects, or spread in air. Gold nanoparticles can be simultaneously concentrated with the analytes in different phases, realizing sensitive, surface-enhanced Raman scattering detection even at attomolar (10−18 mol/L) concentration levels. The acoustic levitation platform-enabled, lossless analyte enrichment can significantly increase the analytical performance of many conventional microsensing techniques.

A fast sampling based evolutionary algorithm for million-dimensional multiobjective optimization

With their complexity and vast search space, large-scale multiobjective optimization problems (LSMOPs) challenge existing multiobjective evolutionary algorithms (MOEAs). Recently, several large-scale multiobjective evolutionary algorithms have been developed to tackle LSMOPs. Unlike conventional MOEAs that concentrate on selection operations in the objective space, large-scale MOEAs emphasize operations in the decision space, such as offspring generation, to tackle the large number of decision variables. Nevertheless, most present large-scale MOEAs experience difficulty effectively and efficiently solving LSMOPs with tens of thousands or more decision variables or exhibit poor versatility in solving different LSMOPs. We propose a fast large-scale MOEA framework with reference-guided offspring generation, named FLEA, aiming at these issues. Generally, FLEA constructs several reference vectors in the decision space to steer the sampling of promising solutions during offspring generation. A parameter is used to allocate computation resources between the convergence and diversity of the offspring population adaptively. Without computationally expensive problem reformulation or decision variable analysis techniques, the proposed method can significantly accelerate the search speed of conventional MOEAs in solving LSMOPs. FLEA is examined on various LSMOPs with up to 1.6 million decision variables, demonstrating its superior effectiveness, efficiency, and versatility in large-scale multiobjective optimization.