An Improved Deep Neural Network Model of Intelligent Vehicle Dynamics via Linear Decreasing Weight Particle Swarm and Invasive Weed Optimization Algorithms.

We propose an improved DNN modeling method based on two optimization algorithms, namely the linear decreasing weight particle swarm optimization (LDWPSO) algorithm and invasive weed optimization (IWO) algorithm, for predicting vehicle's longitudinal-lateral responses. The proposed improved method can restrain the solutions of weight matrices and bias matrices from falling into a local optimum while training the DNN model. First, dynamic simulations for a vehicle are performed based on an efficient semirecursive multibody model for real-time data acquisition. Next, the vehicle data are processed and used to train and test the improved DNN model. The vehicle responses, which are obtained from the LDWPSO-DNN and IWO-DNN models, are compared with the DNN and multibody results. The comparative results show that the LDWPSO-DNN and IWO-DNN models predict accurate longitudinal-lateral responses in real-time without falling into a local optimum. The improved DNN model based on optimization algorithms can be employed for real-time simulation and preview control in intelligent vehicles.

Adaptive Contrastive Learning with Label Consistency for Source Data Free Unsupervised Domain Adaptation.

Unsupervised domain adaptation, which aims to alleviate the domain shift between source domain and target domain, has attracted extensive research interest; however, this is unlikely in practical application scenarios, which may be due to privacy issues and intellectual rights. In this paper, we discuss a more challenging and practical source-free unsupervised domain adaptation, which needs to adapt the source domain model to the target domain without the aid of source domain data. We propose label consistent contrastive learning (LCCL), an adaptive contrastive learning framework for source-free unsupervised domain adaptation, which encourages target domain samples to learn class-level discriminative features. Considering that the data in the source domain are unavailable, we introduce the memory bank to store the samples with the same pseudo label output and the samples obtained by clustering, and the trusted historical samples are involved in contrastive learning. In addition, we demonstrate that LCCL is a general framework that can be applied to unsupervised domain adaptation. Extensive experiments on digit recognition and image classification benchmark datasets demonstrate the effectiveness of the proposed method.

A New Deep Model for Detecting Multiple Moving Targets in Real Traffic Scenarios: Machine Vision-Based Vehicles.

When performing multiple target detection, it is difficult to detect small and occluded targets in complex traffic scenes. To this end, an improved YOLOv4 detection method is proposed in this work. Firstly, the network structure of the original YOLOv4 is adjusted, and the 4× down-sampling feature map of the backbone network is introduced into the neck network of the YOLOv4 model to splice the feature map with 8× down-sampling to form a four-scale detection structure, which enhances the fusion of deep and shallow semantics information of the feature map to improve the detection accuracy of small targets. Then, the convolutional block attention module (CBAM) is added to the model neck network to enhance the learning ability for features in space and on channels. Lastly, the detection rate of the occluded target is improved by using the soft non-maximum suppression (Soft-NMS) algorithm based on the distance intersection over union (DIoU) to avoid deleting the bounding boxes. On the KITTI dataset, experimental evaluation is performed and the analysis results demonstrate that the proposed detection model can effectively improve the multiple target detection accuracy, and the mean average accuracy (mAP) of the improved YOLOv4 model reaches 81.23%, which is 3.18% higher than the original YOLOv4; and the computation speed of the proposed model reaches 47.32 FPS. Compared with existing popular detection models, the proposed model produces higher detection accuracy and computation speed.

Damage Detection for Conveyor Belt Surface Based on Conditional Cycle Generative Adversarial Network.

The belt conveyor is an essential piece of equipment in coal mining for coal transportation, and its stable operation is key to efficient production. Belt surface of the conveyor is vulnerable to foreign bodies which can be extremely destructive. In the past decades, much research and numerous approaches to inspect belt status have been proposed, and machine learning-based non-destructive testing (NDT) methods are becoming more and more popular. Deep learning (DL), as a branch of machine learning (ML), has been widely applied in data mining, natural language processing, pattern recognition, image processing, etc. Generative adversarial networks (GAN) are one of the deep learning methods based on generative models and have been proved to be of great potential. In this paper, a novel multi-classification conditional CycleGAN (MCC-CycleGAN) method is proposed to generate and discriminate surface images of damages of conveyor belt. A novel architecture of improved CycleGAN is designed to enhance the classification performance using a limited capacity images dataset. Experimental results show that the proposed deep learning network can generate realistic belt surface images with defects and efficiently classify different damaged images of the conveyor belt surface.

Artificial Intelligence-Based Hole Quality Prediction in Micro-Drilling Using Multiple Sensors.

The prevalence of micro-holes is widespread in mechanical, electronic, optical, ornaments, micro-fluidic devices, etc. However, monitoring and detection tool wear and tool breakage are imperative to achieve improved hole quality and high productivity in micro-drilling. The various multi-sensor signals are used to monitor the condition of the tool. In this work, the vibration signals and cutting force signals have been applied individually as well as in combination to determine their effectiveness for tool-condition monitoring applications. Moreover, they have been used to determine the best strategies for tool-condition monitoring by prediction of hole quality during micro-drilling operations with 0.4 mm micro-drills. Furthermore, this work also developed an adaptive neuro fuzzy inference system (ANFIS) model using different time domains and wavelet packet features of these sensor signals for the prediction of the hole quality. The best prediction of hole quality was obtained by a combination of different sensor features in wavelet domain of vibration signal. The model's predicted results were found to exert a good agreement with the experimental results.

Performance investigations for sustainability assessment of Hastelloy C-276 under different machining environments.

Hastelloy is categorized as difficult to cut superalloy widely used in aerospace, nuclear reactor components and chemical industry because of its magnificent strength and higher heat efficiency. Since, the machining of this material is quite difficult and hence suitable cooling systems are required to achieve sustainable manufacturing goals. The present investigation has been focused on the machining performance and sustainability assessment of turning Hastelloy C-276 in dry, flood and minimum quantity lubrication (MQL) environments. Taguchi L-9 array has been utilized to conduct and record the experimental output along with TOPSIS approach to evaluate the sustainability. The output responses viz. cutting forces, surface roughness, cutting temperature, energy consumption and carbon emission have been recorded at various levels of input variables. The experimental results revealed that MQL has minimized the cutting forces, surface roughness and temperature by margin of 20-38%. Likewise, energy expenditure and carbon emission was declined by 9-27% respectively compared to other conditions. Sustainability analysis explored best performance index during equal weightage criteria at 125 m/min, 0.246 and 0.8 mm doc under MQL. However, implementing assigned weightage system evaluated best condition for dry machining as 88 m/min and 0.246 mm/rev having same doc. SEM analysis of insert reported mainly abrasion and adhesion type of tool wear at all parametric range and machining conditions.

Measuring Shape Parameters of Pearls in Batches Using Machine Vision: A Case Study.

To solve the problem of low precision of pearl shape parameters' measurement caused by the mutual contact of batches of pearls and the error of shape sorting, a method of contacting pearls' segmentation based on the pit detection was proposed. Multiple pearl images were obtained by backlit imaging, the quality of the pearl images was improved through appropriate preprocessing, and the contacted pearl area was extracted by calculating the area ratio of the connected domains. Then, the contour feature of the contact area was obtained by edge tracking to establish the mathematical model of the angles between the edge contour points. By judging the angle with a threshold of 60° as the candidate concave point, a concave point matching algorithm was introduced to get the true concave point, and the Euclidean distance was adopted as a metric function to achieve the segmentation of the tangent pearls. The pearl shape parameters' model was established through the pearl contour image information, and the shape classification standard was constructed according to the national standard. Experimental results showed that the proposed method produced a better segmentation performance than the popular watershed algorithm and morphological algorithm. The segmentation accuracy was above 95%, the average loss rate was within 4%, and the sorting accuracy based on the shape information was 94%.

Measuring Liquid Droplet Size in Two-Phase Nozzle Flow Employing Numerical and Experimental Analyses.

The flavoring process ensures the quality of cigarettes by endowing them with special tastes. In this process, the flavoring liquid is atomized into particles by a nozzle and mixed with the tobacco in a rotating drum. The particle size of the flavoring liquid has great influence on the atomization effect; however, limited research has addressed the quantitation of the liquid particle size in two-phase nozzle flow. To bridge this research gap, the authors of this study employed numerical and experimental techniques to explore the quantitative analysis of particle size. First, a simulation model for the flavoring nozzle was established to investigate the atomization effect under different ejection pressures. Then, an experimental test is carried out to compare the test results with the simulation results. Lastly, the influencing factors of liquid particle size in two-phase nozzle flow were analyzed to quantify particle size. The analysis results demonstrated that there was a cubic correction relationship between the simulation and experiment particle size. The findings of this study may provide a reliable reference when evaluating the atomization effect of flavoring nozzles.

A New Kinect V2-Based Method for Visual Recognition and Grasping of a Yarn-Bobbin-Handling Robot.

This work proposes a Kinect V2-based visual method to solve the human dependence on the yarn bobbin robot in the grabbing operation. In this new method, a Kinect V2 camera is used to produce three-dimensional (3D) yarn-bobbin point cloud data for the robot in a work scenario. After removing the noise point cloud through a proper filtering process, the M-estimator sample consensus (MSAC) algorithm is employed to find the fitting plane of the 3D cloud data; then, the principal component analysis (PCA) is adopted to roughly register the template point cloud and the yarn-bobbin point cloud to define the initial position of the yarn bobbin. Lastly, the iterative closest point (ICP) algorithm is used to achieve precise registration of the 3D cloud data to determine the precise pose of the yarn bobbin. To evaluate the performance of the proposed method, an experimental platform is developed to validate the grabbing operation of the yarn bobbin robot in different scenarios. The analysis results show that the average working time of the robot system is within 10 s, and the grasping success rate is above 80%, which meets the industrial production requirements.

Belt Tear Detection for Coal Mining Conveyors.

The belt conveyor is the most commonly used conveying equipment in the coal mining industry. As the core part of the conveyor, the belt is vulnerable to various failures, such as scratches, cracks, wear and tear. Inspection and defect detection is essential for conveyor belts, both in academic research and industrial applications. In this paper, we discuss existing techniques used in industrial production and state-of-the-art theories for conveyor belt tear detection. First, the basic structure of conveyor belts is discussed and an overview of tear defect detection methods for conveyor belts is studied. Next, the causes of conveyor belt tear are classified, such as belt aging, scratches by sharp objects, abnormal load or a combination of the above reasons. Then, recent mainstream techniques and theories for conveyor belt tear detection are reviewed, and their characteristics, advantages and shortcomings are discussed. Furthermore, image dataset preparation and data imbalance problems are studied for belt defect detection. Moreover, the current challenges and opportunities for conveyor belt defect detection are discussed. Lastly, a case study was carried out to compare the detection performance of popular techniques using industrial image datasets. This paper provides professional guidelines and promising research directions for researchers and engineers based on the leading theories in machine vision and deep learning.

Identification of localized defects and fault size estimation of taper roller bearing (NBC_30205) with signal processing using the Shannon entropy method in MATLAB for automobile industries applications.

Rotating machine is a common class of machinery in most of the industry and the main root cause of machinery failure is a faulty bearing. Bearings are most widely used in various types of machine elements ranging from small to heavy machinery and the common cause of machinery failure is a fault in bearings. Bearing faults can be external or internal which mainly depends on different operating conditions and these faults may cause severe damage to rotating components in machinery. Signal processing methods have traditionally been used to diagnose faults in tapered roller element bearings. A wavelet transform is the most common and effective tool for understanding and analyzing the vibration signal of bearings as it is responded quickly and observed sudden changes along with the transient impulses in the signal caused by faults in the different parts of bearing elements. In this article, localized fault's position and size on the outer ring of tapered roller bearing were investigated. Three different real values wavelets (DB2, Meyer, and Morlet) are analyzed as per Simple Sensitivity index criteria. Finally, experiments are carried out with four sets of bearing having fault on outer racing of bearing, and for the estimation of fault size, the setup was misaligned at ranging (0.00mm-1.50 mm) with a uniform deviation of 0.50 mm for each experiment. Shannon entropy was calculated for the identification of localized size of the faults with wavelets nomenclature, the result of DB2, Morlet, and Meyer wavelets at high-frequency zone are presented.The scanning electron microscope (SEM) has been taken for the estimation of size of the fault. The proposed method has been successfully implemented for measuring defect width and size. Also, it has been observed that with increased magnification level from 0.00 mm to 0.50 mm, the crack width of the faulty bearing was increased by 0.813 mm, and whenever on further increase in magnification level of 0.50, 1.00 mm and 1.50 mm the crack width of the faulty bearing was increased by 2.568 mm and 3.856 respectively.

Recent developments in the design, development, and analysis of the influence of external magnetic-field on gas-metal arc welding of non-ferrous alloys: review on optimization of arc-structure to enhance the morphology, and mechanical properties of welded joints for automotive applications.

"External magnetic-field (EMF)" has been proved as an additional process parameter like voltage and current affecting the weld arc form, molten metal-flow, microstructure, and characteristics of the weld joint. This article analyzed the research work that has been done to promote EMF application in welding and discussed the recent development trends and research in the design and fabrication of EMF setup to the controlled arc welding process. It is found that even after the successful application of EMF in welding. Still, there is no mass level initiation to integrate EMF with welding machines that hinder researchers and manufacturers to accept it as a regular process parameter to control weld quality.

Progressive developments and challenges in dissimilar laser welding of steel to various other light alloys (Al/Ti/Mg): A comprehensive review.

In recent days, the utilization of lightweight alloys for various applications has been increased massively. Starting from the automobile industry, aerospace industry, and even in the biomedical field, there is a need for dissimilar precise joining of steel to other light alloys (magnesium alloy, aluminum alloy, titanium alloy). However, those alloys are characterized by different melting temperatures, machinability, strength, thermal conductivity, and oxygen reactivity. Considering this welding to challenge ongoing laser welding efforts to improve laser welding quality by altering the welding techniques, modes, proper use of shielding gasses, using suitable process parameters, and even proper joint and surface preparations are discussed. The feasibility of implementing all those things in the industrial setup can be understood only after analyzing recent works. Changes in microstructure and the defects (solidification cracking, intermetallic components formation, porosity) arrived during and after laser welding of these materials are reviewed. The paper also highlights the effect of shielding gas, welding speed, laser power, defocusing position, etc. during laser welding of lightweight materials. The critical issues related to dissimilar laser welding of these combinations and some remedial measures are discussed. The purpose of this review is to emphasize and understand the recent trends of dissimilar laser welding and explore the scope of industry level applications.

Multiobjective Optimization of Chemically Assisted Magnetic Abrasive Finishing (MAF) on Inconel 625 Tubes Using Genetic Algorithm: Modeling and Microstructural Analysis.

The demand for the surface integrity of complex structures is drastically increasing in the field of aerospace, marine and automotive industry. Therefore, Inconel alloy, due to its superior attributes, has a wide scope for the improvement in surface integrity. To achieve the precise surface finish and enhance the process performance, process optimization is necessary. In current paper, chemically assisted MAF process parameters were optimized using the genetic algorithm (GA) approach during finishing of Inconel 625 tubes. Regression models were developed for improvement in internal surface finish (PIISF), improvement in external surface finish (PIESF), and material removal (MR) using Design expert software. Then, the surface microstructure of Inconel 625 tubes was analyzed using scanning electron microscopy (SEM). ANOVA analysis predicts that processing time and abrasive size have the highest percentage contribution in improving the surface finish and material removal. Multioptimization results suggested to set the level of processing time (A) at 75 min, surface rotational speed (B) at 60 RPM, weight % of abrasives (C) at 30%, chemical concentration (D) at 500 gm/lt and abrasive size (E) at 40 microns to obtain optimal parameters for PIISF, PIESF and MR responses.

3D Parametric and Nonparametric Description of Surface Topography in Manufacturing Processes.

Surface topography has a profound influence on the function of a surface [...].

Surface Roughness Evaluation in Thin EN AW-6086-T6 Alloy Plates after Face Milling Process with Different Strategies.

Lightweight alloys made from aluminium are used to manufacture cars, trains and planes. The main parts most often manufactured from thin sheets requiring the use of milling in the manufacturing process are front panels for control systems, housing parts for electrical and electronic components. As a result of the final phase of the manufacturing process, cold rolling, residual stresses remain in the surface layers, which can influence the cutting processes carried out on these materials. The main aim of this study was to verify whether the strategy of removing the outer material layers of aluminium alloy sheets affects the surface roughness after the face milling process. EN AW-6082-T6 aluminium alloy thin plates with three different thicknesses and with two directions relative to the cold rolling process direction (longitudinal and transverse) were analysed. Three different strategies for removing the outer layers of the material by face milling were considered. Noticeable differences in surface roughness 2D and 3D parameters were found among all machining strategies and for both rolling directions, but these differences were not statistically significant. The lowest values of Ra = 0.34 µm were measured for the S#3 strategy, which asymmetrically removed material from both sides of the plate (main and back), for an 8-mm-thick plate in the transverse rolling direction. The highest values of Ra = 0.48 µm were measured for a 6-mm-thick plate milled with the S#2 strategy, which symmetrically removed material from both sides of the plate, in the longitudinal rolling direction. However, the position of the face cutter axis during the machining process was observed to have a significant effect on the surface roughness. A higher surface roughness was measured in the areas of the tool point transition from the up-milling direction to the down-milling direction (tool axis path) for all analysed strategies (Ra = 0.63-0.68 µm). The best values were obtained for the up-milling direction, but in the area of the smooth execution of the process (Ra = 0.26-0.29 µm), not in the area of the blade entry into the material. A similar relationship was obtained for analysed medians of the arithmetic mean height (Sa) and the root-mean-square height (Sq). However, in the case of the S#3 strategy, the spreads of results were the lowest.

Multiple-Criteria Decision-Making and Sensitivity Analysis for Selection of Materials for Knee Implant Femoral Component.

Total knee replacement (TKR) is a remarkable achievement in biomedical science that enhances human life. However, human beings still suffer from knee-joint-related problems such as aseptic loosening caused by excessive wear between articular surfaces, stress-shielding of the bone by prosthesis, and soft tissue development in the interface of bone and implant due to inappropriate selection of TKR material. The choice of most suitable materials for the femoral component of TKR is a critical decision; therefore, in this research paper, a hybrid multiple-criteria decision-making (MCDM) tactic is applied using the degree of membership (DoM) technique with a varied system, using the weighted sum method (WSM), the weighted product method (WPM), the weighted aggregated sum product assessment method (WASPAS), an evaluation based on distance from average solution (EDAS), and a technique for order of preference by similarity to ideal solution (TOPSIS). The weights of importance are assigned to different criteria by the equal weights method (EWM). Furthermore, sensitivity analysis is conducted to check the solidity of the projected tactic. The weights of importance are varied using the entropy weights technique (EWT) and the standard deviation method (SDM). The projected hybrid MCDM methodology is simple, reliable and valuable for a conflicting decision-making environment.

Concurrent Synthesis and Immobilization of Ag Nanoparticles over TiO2 via Plasma Reduction for Photocatalytic Treatment of Methyl Blue in Water.

Pure TiO2 nanoparticles (TiO2NPs) were produced via the sol-gel method and then coated with silver nanoparticles (AgNPs) to reduce their optical band gap. The concurrent synthesis and immobilization of AgNPs over TiO2NPs was achieved through the interaction of an open-air argon plasma jet with a solution of silver nitrate/stabilizer/TiO2NPs. The one-pot plasma synthesis and coating of AgNPs over TiO2NPs is a more straightforward and environmentally friendly method than others. The plasma-produced Ag/TiO2 nanocomposites were characterized and tested for their photocatalytic potential by degrading different concentrations of methyl blue (MB) in water. The dye concentration, oxidant dose, catalyst dose, and reaction time were also optimized for MB degradation. XRD results revealed the formation of pure AgNPs, pure TiO2NPs, and Ag/TiO2 nanocomposites with an average grain size of 12.36 nm, 18.09 nm, and 15.66 nm, respectively. The immobilization of AgNPs over TiO2NPs was also checked by producing SEM and TEM images. The band gap of AgNPs, TiO2NPs, and Ag/TiO2 nanoparticles was measured about 2.58 eV, 3.36 eV, and 2.86 eV, respectively. The ultraviolet (UV) results of the nanocomposites were supportive of the degradation of synthetic dyes in the visible light spectrum. The AgNPs in the composite not only lowered the band gap but also obstructed the electron-hole recombinations. The Ag/TiO2 composite catalyst showed 90.9% degradation efficiency with a 5 ppm dye concentration after 120 min of light exposure.

Special Issue: Advances in Hard-to-Cut Materials: Manufacturing, Properties, Process Mechanics and Evaluation of Surface Integrity.

The rapid growth of a modern industry results in a growing demand for construction materials with excellent operational properties. However, the improved features of these materials can significantly hinder their manufacturing, therefore they can be defined as hard-to-cut. The main difficulties during the manufacturing/processing of hard-to-cut materials are attributed to their high hardness and abrasion resistance, high strength at room or elevated temperatures, increased thermal conductivity, as well as their resistance to oxidation and corrosion. Nowadays the group of hard-to-cut materials includes the metallic materials, composites, as well as ceramics. This special issue, "Advances in Hard-to-Cut Materials: Manufacturing, Properties, Process Mechanics and Evaluation of Surface Integrity" provides a collection of research papers regarding the various problems correlated with hard-to-cut materials. The analysis of these studies reveals primary directions regarding the developments in manufacturing methods, and the characterization and optimization of hard-to-cut materials.

On the Chip Shaping and Surface Topography When Finish Cutting 17-4 PH Precipitation-Hardening Stainless Steel under Near-Dry Cutting Conditions.

This study describes the surface topography of the 17-4 PH stainless steel machined under dry, wet and near-dry cutting conditions. Cutting speeds of 150-500 m/min, feeds of 0.05-0.4 mm/rev and 0.5 mm depth of cutting were applied. The research was based on the 'parameter space investigation' method. Surface roughness parameters, contour maps and material participation curves were analysed using the optical Sensofar S Neox 3D profilometer and the effect of feed, cutting speed and their mutual interaction was noticed. Changes in chip shape depending on the processing conditions are shown. Compared to dry machining, a reduction of Sa, Sq and Sz parameters of 38-48% was achieved for near-dry condition. For lower feeds and average cutting speeds valleys and ridges were observed on the surface machined under dry, wet and near-dry conditions. For higher feeds and middle and higher cutting speeds, deep valleys and high ridges were observed on the surface. Depending on the processing conditions, different textures of the machined surface were registered, particularly anisotropic mixed, periodic and periodically determined. In the Sa range of 0.4-0.8 µm for dry and wet conditions the surface isotropy is ~20%, under near-dry conditions it is ~60%.