Current Issue : April-June Volume : 2022 Issue Number : 2 Articles : 5 Articles
Magnetic metallic nanoparticles (MNPs) of Ni, Ni82Fe18, Ni50Fe50, Ni64Fe36, and Fe were prepared by the technique of the electrical explosion of metal wire. The average size of the MNPs of all types was in the interval of 50 to 100 nm. Magnetic polymeric composites based on polyvinyl butyral with embedded metal MNPs were synthesized and their structural, adhesive, and magnetic properties were comparatively analyzed. The interaction of polyvinyl butyral (supplied as commercial GE cryogenic varnish) with metal MNPs was studied by microcalorimetry. The enthalpy of adhesion was also evaluated. The positive values of the enthalpy of interaction with GE increase in the series Ni82Fe18, Ni64Fe36, Ni50Fe50, and Fe. Interaction of Ni MNPs with GE polymer showed the negative change in the enthalpy. No interfacial adhesion of GE polymer to the surface of Fe and permalloy MNPs in composites was observed. The enthalpy of interaction with GE polymer was close to zero for Ni95Fe5 composite. Structural characterization of the GE/Ni composites with the MNPs with the lowest saturation magnetization confirmed that they tended to be aggregated even for the materials with lowest MNPs concentrations due to magnetic interaction between permalloy MNPs. In the case of GE composites with Ni MNPs, a favorable adhesion of GE polymer to the surface of MNPs was observed....
The application of Artificial Intelligence to the industrial world and its appliances has recently grown in popularity. Indeed, AI techniques are now becoming the de-facto technology for the resolution of complex tasks concerning computer vision, natural language processing and many other areas. In the last years, most of the the research community efforts have focused on increasing the performance of most common AI techniques—e.g., Neural Networks, etc.—at the expenses of their complexity. Indeed, many works in the AI field identify and propose hyper-efficient techniques, targeting high-end devices. However, the application of such AI techniques to devices and appliances which are characterised by limited computational capabilities, remains an open research issue. In the industrial world, this problem heavily targets low-end appliances, which are developed focusing on saving costs and relying on—computationally—constrained components. While some efforts have been made in this area through the proposal of AI-simplification and AI-compression techniques, it is still relevant to study which available AI techniques can be used in modern constrained devices. Therefore, in this paper we propose a load classification task as a case study to analyse which state-of-the-art NN solutions can be embedded successfully into constrained industrial devices. The presented case study is tested on a simple microcontroller, characterised by very poor computational performances—i.e., FLOPS –, to mirror faithfully the design process of low-end appliances. A handful of NN models are tested, showing positive outcomes and possible limitations, and highlighting the complexity of AI embedding....
Embedded systems are nowadays employed in a wide range of application, and their capability to implement calculation-intensive algorithms is growing quickly and constantly. This result is obtained by the exploitation of powerful embedded processors that are often connected to coprocessors optimized for a particular application. This work presents an open-source coprocessor dedicated to the real-time generation of a synthetic signal that mimics the echoes produced by a moving fluid when investigated by ultrasounds. The coprocessor is implemented in a Field Programmable Gate Array (FPGA) device and integrated in an embedded system. The system can replace the complex and inaccurate flow-rigs employed in laboratorial tests of Doppler ultrasound systems and methods. This paper details the coprocessor and its standard interfaces, and shows how it can be integrated in the wider architecture of an embedded system. Experiments showed its capability to emulate a fluid flowing in a pipe when investigated by an echographic Doppler system....
Due to the limitation of energy consumption and power consumption, the embedded platform cannot meet the real-time requirements of the far-infrared image pedestrian detection algorithm. To solve this problem, this paper proposes a new real-time infrared pedestrian detection algorithm (RepVGG-YOLOv4, Rep-YOLO), which uses RepVGG to reconstruct the YOLOv4 backbone network, reduces the amount of model parameters and calculations, and improves the speed of target detection; using space spatial pyramid pooling (SPP) obtains different receptive field information to improve the accuracy of model detection; using the channel pruning compression method reduces redundant parameters, model size, and computational complexity. )e experimental results show that compared with the YOLOv4 target detection algorithm, the Rep-YOLO algorithm reduces the model volume by 90%, the floating-point calculation is reduced by 93.4%, the reasoning speed is increased by 4 times, and the model detection accuracy after compression reaches 93.25%....
This paper reports a type of metamaterial plate enabling in-plane ultra-wide vibration isolation in engineering equipment development. It is composed of periodic hexagonal lattice structures. The acoustic black hole (ABH) structures are embedded in each cell wall of the conventional hexagonal lattice, which results in the reduction of local stiffness in the cell wall and the local mass in the hexagonal corner. The lattice can be simplified as the form of lumped masses vibrating on springs, and two types of eigenstates can be obtained: the rotational eigenstates and the transverse eigenstates. The geometric nonlinearity of the ABH structure leads to unevenly distributed vibration modes, resulting in the ultra-wide bandgap. Experimental results prove the effective attenuation capacity. Compared with the traditional hexagonal lattice, the proposed design provides greater advantages in practical application....
Loading....