Abstract A defect detection approach based on the BiFormer + MPDIoU’s YOLOv8 (BM-YOLOv8) model is proposed which addresses challenges of low accuracy and efficiency in detecting tiny defects inner-wall outer surface automotive Anti-lock Brake Systems (ABS) brake master cylinder. This method constructs an imaging process parameters such as speed inspection required during production ABS On this basis, it employs dynamic sparse self-attention mechanism to build a network for feature extraction...

This study proposes a novel defect detection method to address the low accuracy and insufficient efficiency encountered during surface on aeroengine turbine blades (ATBs). The proposed approach employs LDconv model adjust size shape of convolutional kernels dynamically, integrates deformable attention mechanism (DAT) capture minute features effectively, uses Focaler-CIoU optimize bounding box loss function network. Our approaches collectively provide precise defects ATBs. results show that...

Thermal runaway in lithium-ion batteries is a primary safety concern electric vehicles (EVs). Herein, numerical thermal abuse model proposed that integrates ordinary differential equations (ODEs), heat-transfer partial (PDEs), natural convection computational fluid dynamics (CFD), and radiation to investigate propagation battery pack. A three-dimensional geometric of high-energy pack comprising 18,650 format cells was constructed analyze the characteristics using finite element method (FEM)....

Increasing the coverage rate of ultrasonic path to flow field is helpful improve measurement results' accuracy transit-time flowmeters while asymmetric or turbulence exists. In this paper, a 3D isosceles triangular presented considering width signal and maximum field. To evaluate feasibility effectiveness proposed, computational fluid dynamics simulations three pipeline configurations, including straight pipeline, single elbow, contracted upstream flowmeter <sub...

The monitoring of structural integrity pipeline tapered thread connections is great significance in terms safe operation the industry. In order to detect effectively loosening degree connection, an active sensing method using piezoceramic transducers was developed based on time reversal technique this paper. As piezoeramic can be either as actuators or sensors generate stress waves, energy transmission for connection analyzed. Subsequently, detection principle introduced. Finally, inherent...

ABSTRACT The finite deformation of rubber under multiaxial stress will finally result in its fatigue failure. ability to predict the effects complex strain histories on life is a critical need. cracking energy density (CED) distribution characteristics and estimated by CED criterion are investigated. Then influences crack orientation angle θ principal stretch ratio λ relationship between (SED) obtained. Finally, results used for predicting material compared experimental values. indicate that...

Drones have emerged as a critical tool for the detection of high-altitude glass curtain cracks. However, their utility is often compromised by vibrations and other environmental factors that can induce motion blur, compromising image quality accuracy crack detection. This paper presents novel GAN-based enhanced U-shaped Transformer network, named GlassCurtainCrackDeblurNet, designed specifically deblurring drone-captured images To optimize performance our proposed method this application, we...

The monitoring of structural integrity for pipes is one potential health (SHM) application using piezoelectric sensor networks. A dynamic cooperative identification method (DCIM) pipe SHM with lead zirconate titanate (PZT) proposed based on synergetics in this paper. model a cooperation–competition–update process was developed and its kinetic equation established. With the governing principle, order parameter function were acquired. On basis synergetic effect adjoint vector obtained from...

To solve the problems of 'irreversibility', 'error accumulation' and randomicity classification order in multi-class support vector machines based on binary tree (BT-SVM), paper proposes a method double (DBT-SVM). According to method, each sub-classifier BT-SVM is modified. After unknown samples are classified by modified BT-SVM, negative output its final can be again adding an Auxiliary so that misclassified mixed correctly. Experiment results show accuracy earlier improved using DBT-SVM...

A magnet stably suspending in magnetic fluid is influenced by several factors. The key to analyze prerequisites of suspension immersed body fluids the establishment an exact mathematical model. magnetomechanical modeling spherical encased vessel established. influence on force conductivity, remnant magnetism and radius magnet, container off-centering analyzed. conclusion has universal character. calculation results indicate that various parameters inerratic force, conditions for...

Clark-Wilson model is widely used to protect commercial information against unauthorized modification. However, due only single-level data protection support and no attention paid on difference of importance level protected data, the does not apply many scenarios. An enhanced put forward in this paper provide flexible protection. By using traditional integrity as basis for modification objects regarding procedures special items, upgradeable software provided. The size access control triple...

Abstract Aiming to the present problems of vehicle operating safe state monitoring technology, study on parameter and measurement model is proposed. Vehicle parameters including vehicle's motion attitude parameter(MAP), dynamic load parameter(DLP) braking performance parameter(BPP) three key are put forward, which more comprehensive scientific than ever. By establishing using WEIS realize monitoring. Analyzed connection among MAP, DLP BPP information in reducing unnecessary, repetitive...

The characteristics of the combustion temperature, flow velocity, CO distribution and NO x emissions a 10 MW gas burner at different primary to secondary air ratios are numerically studied using computational fluid dynamics software Fluent. results indicate that ratio in determines quality through influencing some parameters directly, such as efficiency, intensity, profile stability flame well emission . Then two evaluation indexes summarized after analyzing structure flow. detailed reported...

Semantic segmentation with convolutional neural networks under a complex background using the encoder-decoder network increases overall performance of online machine vision detection and identification. To maximize accuracy semantic background, it is necessary to consider response values objects components their mutually exclusive relationship. In this study, we attempt improve low component segmentation. The basic encoder selected for segmentation, UPerNet modified based on analysis module....

Direct Current (DC) electrical resistivity is a material property that sensitive to temperature changes. In this paper, the relationship between and local inside steel shell battery cells (two commercial 10 Ah 4.5 lithium-ion cells) innovatively studied by Electrical Resistance Tomography (ERT). The Schlumberger configuration in ERT applied divide cell body into several blocks distributed different levels, where apparent resistivities are measured multi-electrode surface probes. investigated...

OPNET Modeler is an important network simulation tool, which can be used for designing and analyzing of time synchronization protocol. Local clock necessary to protocols such as NTP, SNTP while OPENT only provides a precise that recording time. This paper presents design model drifting by client node in protocol using tools.

In order to enhance the stability of RBF neural network under constant changing conditions, we propose a new scheme forecasting method based on single model structure. The can avoid inherent defects double structure and improve network's online self-correcting ability. Self-correcting is realized by continuous learning adjustment according changes monitoring environment. New samples are used approximation precision during work time. shows remarkable adaptability slow quick environment from...

Vertical tanks are commonly used appliances for liquids, and its capacity is very important quantitative liquid ratio trade. In order to measure the of vertical more conveniently, this paper proposes a tank measurement method based on Monte Carlo Method. The arranges plurality sensor points inner surface tank, then performs tests by generating large number random sample points, finally calculates counting that meet criterion. criterion whether point located in which core issue, established...

Objectives The prevalence of hypothyroidism in systemic lupus erythematosus (SLE) is significantly higher than that the common public. While SLE itself can affect multiple organs, abnormal thyroid function may exacerbate organ damage patients with SLE. We aimed to predict and examine associated factors machine learning approaches.Methods In a cross-sectional study, 255 diagnosed at rheumatology department Xiangya Hospital between June 2012 December 2016 were investigated. Feature engineering...

After analyzing the influencing factors of flexible workpiece path(FWP) process deformation, this article proposes basic conception deformation intelligent forecasting and compensation, start from modeling method Takagi-Sugeno fuzzy neural network, to modify classic FNN model construct multiple input/output TS-FNN for FWP control; with LMS law steepest descent method, antecedent network membership function parameter adjustment study is deduced; finally carry on comprehensive simulation...