The detection of wind turbine blades (WTBs) damage is crucial for improving power generation efficiency and extending the lifespan turbines. However, traditional methods often suffer from false positives missed detections, they do not adequately account complex weather conditions such as fog snow. Therefore, this study proposes a WTBs model based on an improved YOLOv8, named AUD-YOLO. Firstly, ADown module integrated into YOLOv8 backbone to replace some conventional convolutional...

Variable amplitude loading is frequently applied to welded structures in practical engineering, and fatigue failure a prevalent problem. In recent years, machine learning useful technique for predicting life. However, it challenging acquire sufficient number of reliable training samples tests under variable loading. The models' accuracy generalization capabilities are impacted by this. This work introduces novel data augmentation approach utilizing physics-informed Generative Adversarial...

The bearing system of an alternating current (AC) motor is a nonlinear dynamics system. working state rolling bearings directly determines whether the machine in reliable operation. Therefore, it very meaningful to study fault diagnosis and prediction bearings. In this paper, new method based on variational mode decomposition (VMD), Hilbert transform (HT), broad learning model (BLM), called VHBLFD proposed for method, VMD used decompose vibration signals obtain intrinsic functions (IMFs). HT...

Photosynthetic responses of potato (Solanum tuberosum L. cv. Chunzao) were examined during virus Y (PVYNTN) infection. PVYNTN infection significantly reduced net photosynthetic rate and stomatal conductance, but had little influence on intercellular CO2 concentration. As the disease developed, maximum carboxylation velocity ribulose-1,5-bisphosphate carboxylase/oxygenase electron transport contributing to regeneration gradually decreased, followed by substantial reductions in relative...

Abstract Phytochemical study on the root tubers of Ophiopogon japonicus resulted in isolation and identification 13 homoisoflavonoids, including three new compounds, 8‐formyl‐7‐hydroxy‐5,4′‐dimethoxy‐6‐methylhomoisoflavone ( 1 ), 6‐formylisoophiopogonone B 2 8‐formylophiopogonanone 4 ten known homoisoflavonoids 3 , 5 – . The absolute configurations ) 8‐formyl‐7‐hydroxy‐5,4′‐dimethoxy‐6‐methylhomoisoflavanone were confirmed by time‐dependent density‐functional‐theory (TD DFT) calculations...

We report the latest development in tunable transmitter optical sub-assembly (TOSA) and small form-factor pluggable (SFP) transceivers based on simple compact V-cavity laser (VCL), aiming for low-cost deployment metro, access, data center networks. The VCL employs a half-wave coupler to achieve high side-mode suppression ratio (SMSR) Vernier effect wide wavelength tuning range. Full C-band from 1529.55- 1566.31-nm with SMSR above 36 dB is demonstrated. Since does not involve any grating or...

In this paper, an improved fish swarm algorithm for neighborhood rough set reduction (IFSANRSR) is proposed. IFSANRSR, by introducing adaptive function to control the visual and step size of artificial fish, problem inconsistent convergence speed existed in a traditional (FSA) avoided. The movement swarming following behavior shorten running time algorithm. searching enhance local search ability without changing global By mechanism extinction rebirth, worst solution eliminated rebirth takes...

Wind turbine blades are easily affected by the working environment and often show damage features such as cracks surface shedding. An improved convolution neural network, ED Net, is proposed to identify their features. EAC block based on asymmetric introduced which strengthens feature extraction during convolution. A DPCI_SC block, attention module, embedded enhance ability obtain spatial location information of damage. GELU used activation function. The loss function smoothed labeled...

License plate detection is a challenging problem due to the large visual variations in complex environments, such as motion blur, occlusion, and lighting changes. An advanced discriminative model needed accurately segment license plates from backgrounds. However, effective models for tend be computationally prohibitive. To address these two conflicting challenges, we propose detect based on CNNs, shallow CNN deep CNN. The used quickly remove most of background regions reduce computation...

In order to obtain comprehensive assessment of the factors influencing fatigue life and further improve accuracy prediction welded joints, soft computing methods, including entropy-based neighborhood rough set reduction algorithm, particle swarm optimization (PSO) algorithm support vector regression machine (SVRM) are combined construct a model titanium alloy joints. By using an joints such as joint type, plate thickness, etc. analyzed results obtained. Fatigue characteristic domains...

Abstract To meet the lightweight needs of modern machinery equipment, welded structure is widely used in practical engineering. During service period, often subjected to multilevel variable amplitude loading, and fatigue failure easy occur. Due complexity stochasticity process, traditional life prediction models cannot demand In this work, a novel model based on whale optimization algorithm (WOA) together with support vector regression (SVR), namely, WOA‐SVR model, established. It combines...

Abstract Defining asset poverty as insufficiency of assets to satisfy household basic needs for a limited period time, the study examines asset-poverty rates in urban China using 2002 survey data from Chinese Household Income Project (CHIP). We find that are lower than those developed countries, part due households’ strong commitment precautionary savings and low standards. However, liquid rate is five times income-poverty China. Notably, asset-poverty-gap ratio shows most households have...

The Manson-Halford (M-H) nonlinear cumulative damage model is widely applied for fatigue life analysis problems under multi-level loading. In this model, the influence of loading sequence on can be better considerer, but interaction effect ignored. An improved whale optimization algorithm (IWOA) by integrating multiple strategies proposed. ability global search and local exploitation balanced through convergence factor, adaptive weighting factors Cauchy reverse learning strategies. order to...

Wind turbine blades (WTBs) are prone to damage from their working environment, including surface peeling and cracks. Early effective detection of defects on WTBs can avoid complex costly repairs serious safety hazards. Traditional object methods have disadvantages insufficient capabilities, extended model inference times, low recognition accuracy for small objects, elongated strip within WTB datasets. In light these challenges, a novel named DCW-YOLO is proposed in this research, which...

Abstract Two new cycloartane triterpenoids, (23 R )‐21,23:23,27‐diepoxycycloarta‐1,24‐diene‐3,27‐dione ( 1 ) and (3 α )‐( ‐ L ‐arabinopyranosyloxy)‐1 ‐hydroxy‐23‐oxocycloartan‐28‐oic acid 2 ), together with six known pentacyclic 3 – 8 , five C 29 steroids, 9 13 were isolated from Kleinhovia hospita. The structures of these compounds determined by analysis their spectroscopic data. Moreover, the absolute configuration was confirmed quantum‐chemical TDDFT calculation its ECD spectrum. All...

Wind turbine blades are readily damaged by the workplace environment and frequently experience flaws such as surface peeling cracking. To address problems of cumbersome operation, high cost, harsh application conditions with traditional damage identification methods, to cater wide mobile terminal devices unmanned aerial vehicles, a novel lightweight asymmetric convolution neural network is proposed. The introduces module based on improved convolution, which applies depthwise separable...

By reversing the pillars formed on SiO2 mask sidewalls, it is possible to fabricate deep-submicron slots with width down 240nm by standard UV-lithography. Based this newly developed process, a single-mode slotted Fabry-Perot laser and wavelength tunable periodically distributed are designed, fabricated characterized. Numerical analysis shows low-loss advantage of slots. Experimentally, showed low threshold current 22mA grating exhibited maximum side mode suppression ratio (SMSR) 43dB...