Search results

Multi-domain convolutional neural network (MDCNN) model has been widely used in object recognition and tracking in the field of computer vision. However, if the objects to be tracked move rapid or the appearances of moving objects vary dramatically, the conventional MDCNN model will suffer from the model drift problem. To solve such problem in tracking rapid objects under limiting environment for MDCNN model, this paper proposed an auto-attentional mechanism-based MDCNN (AA-MDCNN) model for the rapid moving and changing objects tracking under limiting environment.

First, to distinguish the foreground object between background and other similar objects, the auto-attentional mechanism is used to selectively aggregate the weighted summation of all feature maps to make the similar features related to each other. Then, the bidirectional gated recurrent unit (Bi-GRU) architecture is used to integrate all the feature maps to selectively emphasize the importance of the correlated feature maps. Finally, the final feature map is obtained by fusion the above two feature maps for object tracking. In addition, a composite loss function is constructed to solve the similar but different attribute sequences tracking using conventional MDCNN model.

In order to validate the effectiveness and feasibility of the proposed AA-MDCNN model, this paper used ImageNet-Vid dataset to train the object tracking model, and the OTB-50 dataset is used to validate the AA-MDCNN tracking model. Experimental results have shown that the augmentation of auto-attentional mechanism will improve the accuracy rate 2.75% and success rate 2.41%, respectively. In addition, the authors also selected six complex tracking scenarios in OTB-50 dataset; over eleven attributes have been validated that the proposed AA-MDCNN model outperformed than the comparative models over nine attributes. In addition, except for the scenario of multi-objects moving with each other, the proposed AA-MDCNN model solved the majority rapid moving objects tracking scenarios and outperformed than the comparative models on such complex scenarios.

This paper introduced the auto-attentional mechanism into MDCNN model and adopted Bi-GRU architecture to extract key features. By using the proposed AA-MDCNN model, rapid object tracking under complex background, motion blur and occlusion objects has better effect, and such model is expected to be further applied to the rapid object tracking in the real world.

This study aims to explore the evolving field of physics-informed neural networks (PINNs) through an analysis of 996 records retrieved from the Web of Science (WoS) database from 2019 to 2022.

WoS database was analyzed for PINNs using an inhouse python code. The author’s collaborations, most contributing institutes, countries and journals were identified. The trends and application categories were also analyzed.

The papers were classified into seven key domains: Fluid Dynamics and computational fluid dynamics (CFD); Mechanics and Material Science; Electromagnetism and Wave Propagation; Biomedical Engineering and Biophysics; Quantum Mechanics and Physics; Renewable Energy and Power Systems; and Astrophysics and Cosmology. Fluid Dynamics and CFD emerged as the primary focus, accounting for 69.3% of total publications and witnessing exponential growth from 22 papers in 2019 to 366 in 2022. Mechanics and Material Science followed, with an impressive growth trajectory from 3 to 65 papers within the same period. The study also underscored the rising interest in PINNs across diverse fields such as Biomedical Engineering and Biophysics, and Renewable Energy and Power Systems. Furthermore, the focus of the most active countries within each application category was examined, revealing, for instance, the USA’s significant contribution to Fluid Dynamics and CFD with 319 papers and to Mechanics and Material Science with 66 papers.

This analysis illuminates the rapidly expanding role of PINNs in tackling complex scientific problems and highlights its potential for future research across diverse domains.

The purpose of this paper is to develop a decision model to help decision makers with selection of the appropriate supplier.

Supplier selection is a multi‐criteria decision‐making process encompassing various tangible and intangible factors. Both risks and benefits of using a vendor in supply chain are identified for inclusion in the evaluation process. Since these factors can be objective and subjective, a hybrid approach that applies to both quantitative and qualitative factors is used in the development of the model. Taguchi loss functions are used to measure performance of each supplier candidate with respect to the risks and benefits. Analytical hierarchy process (AHP) is used to determine the relative importance of these factors to the decision maker. The weighted loss scores are then calculated for each supplier by using the relative importance as the weights. The composite weighted loss scores are used for ranking of the suppliers. The supplier with the smallest loss score is recommended for selection.

Inclusion of both risk and benefit categories in the evaluation process provides a comprehensive decision tool.

The proposed model provides guidelines for supply chain managers to make an informed decision regarding supplier selection.

Combining Taguchi loss function and AHP provides a novel approach for ranking of potential suppliers for outsourcing purposes.

Specific chemical environments step out in the industry objects. Portland cement composites (concrete and mortar) were impregnated by using the special polymerized sulfur and technical soot as a filler (polymer sulfur composite). Sulfur and technical soot were applied as the industrial waste. Portland cement composites were made of the same aggregate, cement and water. The durability of prepared cement composite samples was tested in 5 per cent solution of HCl and 5 per cent solution of H₂SO₄ as a function of immersion time. The changes in mechanical strength and mass of the samples were periodically measured. Cement composites impregnated with sulfur composite exhibited limited mechanical strength and mass loss, whereas physico-mechanical properties of Portland cement concrete regressed rapidly. The loss in weight of ordinary concrete impregnated with sulfur composite, kept in aqueous solutions of acids, hydroxides, salts and in water for a year was determined using 100 × 100 × 100 mm samples. The same samples were then used in compressive strength tests.

Specific chemical environments affect industrial objects. Portland cement composites (concrete and mortar) were impregnated with a special polymerized sulfur and technical soot as a filler (polymer sulfur composite). Sulfur and technical soot were applied as industrial waste. Portland cement composites were made of the same aggregate, cement and water. The durability of the prepared cement composite samples was tested in 5 per cent solution of HCl and 5 per cent solution of H₂SO₄ as a function of immersion time. The changes in mechanical strength and mass of the samples were periodically measured. Cement composites impregnated with sulfur composite exhibited limited mechanical strength and mass loss, whereas the physico-mechanical properties of the Portland cement concrete regressed rapidly. The loss in weight of ordinary concrete impregnated with sulfur composite, kept in aqueous solutions of acids, hydroxides, salts and in water for a year was determined using 100 × 100 × 100 mm samples. The same samples were then used in compressive strength tests. The image analysis used for surface destruction monitoring, performed by scanning microscopy for the determination of damaged surface area and the original surface area before acid resistance testing, showed similar results. Based on the image analysis results, a model for predicting the degradation of mechanical strength during durability testing was established. The fact that the calculated and experimental strength values were not vastly different proved the validity of the proposed model. A brief summary of new products related to the special sulfur composite is given as follows: impregnation, repair, overlays and precast polymer concrete will be presented. Sulfur composite as a polymer coating impregnation, which has received little attention in recent years, currently has some very interesting applications.

Author comments: The article is original. The article has been written by the stated authors who are all aware of its content and approve its submission. 3. The article has not been published previously. 4. The article is not under consideration for publication elsewhere. 5. No conflict of interest exists, or if such conflict exists, the exact nature must be declared. 6. If accepted, the article will not be published elsewhere in the same form, in any language, without the written consent of the publisher.

Author comments: 1. The article is original. 2. The article has been written by the stated authors who are all aware of its content and approve its submission. 3. The article has not been published previously. 4. The article is not under consideration for publication elsewhere. 5. No conflict of interest exists, or if such conflict exists, the exact nature must be declared. 6. If accepted, the article will not be published elsewhere in the same form, in any language, without the written consent of the publisher.

Selection of logistics service provider (LSP) (also known as Third-party logistics (3PL) is a critical decision, because logistics affects top and bottom line as well. Companies consider logistics as a cost driver and at the time of LSP selection decision, many important decision criteria’s are left out. 3PL selection is multi-criteria decision-making process. The purpose of this paper is to develop an integrated approach, combining quality function deployment (QFD), and Taguchi loss function (TLF) to select optimal 3PL.

Multiple criteria are derived from the company requirements using house of quality. The 3PL service attributes are developed using QFD and the relative importance of the attributes are assessed. TLFs are used to measure performance of each 3PL on each decision variable. Composite weighted loss scores are used to rank 3PLs.

QFD is a better tool which connects attributes used in a decision problem to decision maker’s requirements. In total, 15 criteria were used and TLF provides performance on these criteria.

The proposed model provides a methodology to make informed decision related to 3PL selection. The proposed model may be converted into decision support system.

Proposed approach in this paper is a novel approach that connects the 3PL selection problem to practice in terms of identifying criteria’s and provides a single numerical value in terms of Taghui loss.

This study aims to use deep neural networks (DNNs) to learn the conduction heat transfer physics and estimate temperature distribution images in a physical domain without using any physical model or mathematical governing equation.

Two novel DNNs capable of learning the conduction heat transfer physics were defined. The first DNN (U-Net autoencoder residual network [UARN]) was designed to extract local and global features simultaneously. In the second DNN, a conditional generative adversarial network (CGAN) was used to enhance the accuracy of UARN, which is referred to as CGUARN. Then, novel loss functions, introduced based on outlier errors, were used to train the DNNs.

A UARN neural network could learn the physics of heat transfer. Within a few epochs, it reached mean and outlier errors that other DNNs could never reach after many epochs. The composite outlier-mean error as a loss function showed excellent performance in training DNNs for physical images. A UARN could excellently capture local and global features of conduction heat transfer, whereas the composite error could accurately guide DNN to extract high-level information by estimating temperature distribution images.

This study offers a unique approach to estimating physical information, moving from traditional mathematical and physical models to machine learning approaches. Developing novel DNNs and loss functions has shown promising results, opening up new avenues in heat transfer physics and potentially other fields.

The purpose of this paper is to produce Al6061 metal matrix composites reinforced with silicon carbide (SiC) and graphite particulates and study their wear behavior and also to develop artificial neural network model to predict the mass loss of hybrid composites.

The hybrid composites were produced by using stir casting process. The experiments were conducted based on the central composite rotatable design matrix using pin‐on‐disc wear testing machine. The set of data collected from the experimental values were used to train a back propagation (BP) learning algorithm with one hidden layer network. In artificial neural network (ANN) training module, four input vectors were used in the construction of proposed network namely, weight percentage of SiC particles, weight percentage of graphite particles, applied load and sliding distance. Mass loss was the output to be obtained from the proposed network. After training process, the test data collected from the experimental values were used to check the accuracy of proposed ANN model.

The results show that the well trained one hidden layer network have smaller training errors and much better generalization performance and can be successfully used for the prediction of mass loss of hybrid aluminium metal matrix composites.

In this paper the ANN method was adopted to predict the mass loss of hybrid composites. It was found that artificial neural network can be successfully used for prediction of mass loss of composites.

This paper presents the development of a procedure for the determination of the local magnetic loss distribution in transformer cores. An efficient identification method of the parameters of the Jiles‐Atherton model is first described. This method uses nonlinear optimization techniques and several experimental loops with different magnitudes, or measurements obtained with a low frequency supply signal, for a precise determination of the hysteresis model parameters. It is validated by the identification of two different kinds of magnetic materials: a standard laminated material made of 1008 steel and a soft magnetic composite Atomet‐EM1. The implementation of the hysteresis Jiles‐Atherton model in a 2D field calculation tool is detailed. The field calculation procedure is illustrated by two application examples involving single phase tranformers with cores made of the soft magnetic composite Atomet‐EM1.

The purpose of this paper is to investigate the influence of most predominant heat-treatment parameters on the wear behavior of Al6061 hybrid composite reinforced with 10 weight per cent SiC and 2 weight per cent graphite particles.

The aluminum hybrid composite was produced using stir casting process. Wear testing of heat-treated samples was carried out using a pin-on-disc apparatus. Experiments were conducted by applying design of experiments (DOE) technique. The experimental values were used for formulation of a mathematical model. The wear surfaces of composite specimens were analyzed using scanning electron microscope (SEM).

The volume loss of heat-treated composite initially decreased with increasing aging duration. This was followed by the attainment of a minimum and then a reversal in the trend at longer aging times. SEM micrographs of the wear surfaces of the composite show that the wear mechanisms were abrasion, delamination and adhesion.

In this paper, the hybrid composite was produced using stir casting route, and its wear properties after heat treatment were tested using pin-on-disc apparatus. It was found that heat treatment had a profound effect on the wear behaviour of the developed composite.

Additive manufacturing (AM) of thermoplastic polymers for powder bed fusion processes typically requires each layer to be fused before the next can be deposited. The purpose of this paper is to present a volumetric AM method in the form of deeply penetrating radio frequency (RF) radiation to improve the speed of the process and the mechanical properties of the polymer parts.

The focus of this study was to demonstrate the volumetric fusion of composite mixtures containing polyamide (nylon) 12 and graphite powders using RF radiation as the sole energy source to establish the feasibility of a volumetric AM process for thermoplastic polymers. Impedance spectroscopy was used to measure the dielectric properties of the mixtures as a function of increasing graphite content and identify the percolation limit. The mixtures were then tested in a parallel plate electrode chamber connected to an RF generator to measure the heating effectiveness of different graphite concentrations. During the experiments, the surface temperature of the doped mixtures was monitored.

Nylon 12 mixtures containing between 10% and 60% graphite by weight were created, and the loss tangent reached a maximum of 35%. Selective RF heating was shown through the formation of fused composite parts within the powder beds.

The feasibility of a novel volumetric AM process for thermoplastic polymers was demonstrated in this study, in which RF radiation was used to achieve fusion in graphite-doped nylon powders.