Search Results (8)

A YOLOv5 aluminum profile defect detection algorithm that integrates attention and multi-scale features is proposed in this paper to address the issues of the low detection accuracy, high false detection rates, and high missed detection rates that are caused by the large-scale variation of surface defects, inconspicuous small defect characteristics, and a lack of concentrated feature information in defect areas. Firstly, an improved CBAM (Channel-Wise Attention Module) convolutional attention module is employed, which effectively focuses on the feature information of defect areas in the aluminum defect dataset with only a small amount of spatial dimension. Secondly, a bidirectional weighted feature fusion network is utilized, incorporating a multi-scale feature fusion network with skip connections to aggregate various high-resolution features, thus enriching the semantic expression of features. Then, new size feature maps that have not been fused are introduced into the detection layer network to improve the detection effect of small target defects. Experimental results indicate that an average detection accuracy (mAP) of 82.6% was achieved by the improved YOLOv5 algorithm on the aluminum surface defect dataset. An improvement of 6.2% over the previous version was observed. The current defect detection requirements of aluminum profile production sites are met by this enhanced algorithm. Full article

The large-scale ingot of the 7xxx-series aluminum alloys fabricated by direct chill (DC) casting often suffers from foundry defects such as cracks and cold shut due to the formidable challenges in the precise controlling of casting parameters. In this manuscript, by using the integrated computational method combining numerical simulations with machine learning, we systematically estimated the evolution of multi-physical fields and grain structures during the solidification processes. The numerical simulation results quantified the influences of key casting parameters including pouring temperature, casting speed, primary cooling intensity, and secondary cooling water flow rate on the shape of the mushy zone, heat transport, residual stress, and grain structure of DC casting ingots. Then, based on the data of numerical simulations, we established a novel model for the relationship between casting parameters and solidification characteristics through machine learning. By comparing it with experimental measurements, the model showed reasonable accuracy in predicting the sump profile, microstructure evolution, and solidification kinetics under the complicated influences of casting parameters. The integrated computational method and predicting model could be used to efficiently and accurately determine the DC casting parameters to decrease the casting defects. Full article

One of the main challenges that face the reliable use of photovoltaic solar systems in hot arid regions is the prevailing high temperatures during the day. To overcome this issue, Phase Change Materials (PCM) are used for passive cooling providing different options to attain sufficient thermal management solutions for different applications. Passive cooling can be achieved by adjusting a heat sink to the solar PV module. This can be realized by attaching a PCM layer or sensible heat storage to the backside of PV panels. Few studies have reported on simplified modeling and numerical procedures using the apparent heat capacity formulation and volume averaging technique as a robust approach to solving such sophisticated problems with minimal computational efforts, high accuracy, and in a short period of time. However, there is still a need to bridge the large-scale gap between the macroscale within the PCM layer, with a moving melting front, and the length scale of PV modules. Hence, this work focuses on modeling and simulating PCM-Matrix Absorbers (PCM-MA) that consist of fibrous aluminum cellular structure filled with PCM for passive thermal management of PV panels using apparent heat capacity formulation and homogenization based on volume averaging technique. COMSOL Multiphysics FEM software was used for the numerical simulation of the phase change problem by using a Smoothed Heaviside step function to overcome the singularity of PCM challenge that arises with sharp melting temperatures. To validate the proposed model, it has been compared with a benchmark analytical solution for an ice melting problem, i.e., the Stefan problem, in a semi-finite slab, i.e., Neumann’s solution under the same assumptions and boundary conditions. The specific characteristics of phase change and evolution of melting front with time, heat capacity change with the temperature at different times, and with locations along the slab height are presented. As the phase change is modeled to take place over a mushy region, i.e., a narrow temperature interval rather than a sharp melting point, the results show a good coincidence of the heat capacity profile and its peak at different times and locations. The validated model can be used for the optimization of PCM-MA for any specific geographical location and other applications such as the passive cooling of buildings with PCM integrated with the outer envelope. To this end, the results of the simulation in this work are shown to be in agreement with those obtained from the analytical solution. Full article

In response to problems such as low recognition rate, random distribution of defects and large-scale differences in the detection of surface defects of aluminum profiles by other state-of-the-art algorithms, this paper proposes an improved MS-YOLOv5 model based on the YOLOv5 algorithm. First, a PE-Neck structure is proposed to replace the neck part of the original algorithm in order to enhance the model’s ability to extract and locate defects at different scales. Secondly, a multi-streamnet is proposed as the first detection head of the algorithm to increase the model’s ability to identify distributed random defects. Meanwhile, to overcome the problem of inadequate industrial defect samples, the training set is enhanced by geometric variations and image-processing techniques. Experiments show that the proposed MS-YOLOv5 model has the best mean average precision (mAP) compared to the mainstream target-detection algorithm for detecting surface defects in aluminium profiles, whereas the average single image recognition time is within 19.1FPS, meeting the real-time requirements of industrial inspection. Full article

The internal coupled electromagnetic melt treatment (ICEMT) method is firstly proposed to produce high-quality and large-sized aluminum alloy billets. A three-dimensional model was established to describe the ICEMT process of direct chill casting (DC casting). The effect of ICEMT on the fluid flow patterns and temperature field in the DC casting of ϕ880 mm AA2219 billets is numerically analyzed. Moreover, the mechanisms of the ICEMT process on grain refinement and macrosegregation were discussed. The calculated results indicate that the electromagnetic field appears to be coupled circinate at the cross section of the melt, the fluid flow becomes unstable accompanied by the bias flow, and the temperature profiles are significantly more uniform. An experimental verification was conducted and the results prove that compared with traditional direct chill casting, the microstructures of the AA2219 large-scale billet under the ICEMT process are uniform and fine. Full article

Although wearable antennas have made great progress in recent years, how to design high-performance antennas suitable for most wireless communication systems has always been the direction of RF workers. In this paper, a new approach for the design and manufacture of a compact, low-profile, broadband, omni-directional and conformal antenna is presented, including the use of a customized flexible dielectric substrate with high permittivity and low loss tangent to realize the compact sensing antenna. Poly-di-methyl-siloxane (PDMS) is doped a certain proportion of aluminum trioxide (Al₂O₃) and Poly-tetra-fluoro-ethylene (PTFE) to investigate the effect of dielectric constant and loss tangent. Through a large number of comparative experiments, data on different doping ratios show that the new doped materials are flexible enough to increase dielectric constant, reduce loss tangent and significantly improve the load resistance capacity. The antenna is configured with a multisection microstrip stepped impedance resonator structure (SIR) to expand the bandwidth. The measured reflection return loss (S11) showed an operating frequency band from 0.99 to 9.41 GHz, with a band ratio of 146%. The antenna covers two important frequency bands, 1.71–2.484 GHz (personal communication system and wireless body area network (WBAN) systems) and 5.15–5.825 GHz (wireless local area network-WLAN)]. It also passed the SAR test for human safety. Therefore, the proposed antenna offers a good chance for full coverage of WLAN and large-scale development of wearable products. It also has potential applications in communication systems, wireless energy acquisition systems and other wireless systems. Full article

In the state of Baden-Wuerttemberg, Southwest-Germany, a large-scale forest liming trial was government-funded in 1983 and a lime treatment was carried out in autumn 1983 until early winter 1984. Repeated liming was applied in 2003. The limed sites and adjacent control plots were surveyed repeatedly: in 2003 before the second lime application and again in 2010 and 2015. Research of this scope presents a rare opportunity to evaluate firstly the long-term development of acidified soils with their potential for natural recovery on established control plots, and secondly the long-term effects of repeated lime application—at a collective of study sites of various growth regions and soil properties. A natural recovery in soil pH was observed since 2003, on average limited to an increase of 0.2–0.4 pH units in the forest floor and 0.1–0.3 pH units in the mineral soil until 2015. The majority of the organic layers still show very strong or extreme acidity with a pH value 3.9 on average and in the mineral soil with pH values between 3.8 and 4.6 on average. The exchangeable cations calcium and magnesium slightly increased also, although the base saturation remained <20% by 2015. The exchangeable acid cation concentrations indicated no significant changes and thus no recovery. The lime treatment greatly accelerated the rise in pH by 1.2–1.3 units and base saturation by 40–70% in the organic layer, as well as 0.3–1.2 pH units and base saturation by 7–50% in mineral soil. These effects were decreasing (yet still significant) with depth in the measured soil profile as well as with time since last treatment. Changes in soil cation exchange capacity after liming were significant in 0–5 cm mineral soil, below that they were negligible as the significant increase in base cations were accompanied by decreasing acid cations aluminum and iron (III) especially in the upper soil profile. Additionally, a decrease of forest floor and an enrichment of organic carbon and nitrogen in the mineral topsoil tended to follow liming at some sites. Overall the liming effects had a high variability among the study sites, and were more pronounced in the more acidic and coarser textured sites. Liming of acidified forest soils significantly adds to natural recovery and therefore helps to establish greater buffering capacities and stabilize forest nutrition for the future. Full article

Increasing customer requirements for quality and productivity in extruding aluminum products has led to the development of different types of extrusion dies. In this study, three different types of dies, including traditional flat die, pocket die and spread die were designed to extrude complex and large-scale solid profiles. The design parameters for these dies were used from actual extrusion experience. The results obtained from steady-state simulation such as velocity, temperature, extrusion force, and die deformation were used to assess the advantages and disadvantages of the dies. Transient simulations were performed to analyze the evolution of transverse weld in the pocket and spread dies. The effects of ram speeds on the related extrusion parameters were also investigated. The research results provide useful guides for designers and engineers in selecting these types of extrusion dies. Full article