Insulator Defect Detection Based on ML-YOLOv5 Algorithm.

To address the challenges of balancing accuracy and speed, as well as the parameters and FLOPs in current insulator defect detection, we propose an enhanced insulator defect detection algorithm, ML-YOLOv5, based on the YOLOv5 network. The backbone module incorporates depthwise separable convolution, and the feature fusion C3 module is replaced with the improved C2f_DG module. Furthermore, we enhance the feature pyramid network (MFPN) and employ knowledge distillation using YOLOv5m as the teacher model. Experimental results demonstrate that this approach achieved a 46.9% reduction in parameter count and a 43.0% reduction in FLOPs, while maintaining an FPS of 63.6. It exhibited good accuracy and detection speed on both the CPLID and IDID datasets, making it suitable for real-time inspection of high-altitude insulator defects.

Distance and depth modulation of Talbot imaging via specified design of the grating structure.

For positioning Talbot encoder and Talbot lithography, etc., properties manipulation of Talbot imaging is highly expected. In this work, an investigation on the distance and depth modulation of Talbot imaging, which employs a specially designed grating structure, is presented. Compared with the current grating structure, the proposed grating structure is characterized by having the phase layers with uneven thicknesses. Such a specific structural design can cause the offset of Talbot image from its nominal position, which in turn generates the spatial distance modulation of self-imaging and imaging depth expansion. Theoretical analysis is performed to explain its operating principle, and simulations and experiments are carried out to demonstrate its effectiveness.

Acoustics-Controlled Microdroplet and Microbubble Fusion and Its Application in the Synthesis of Hydrogel Microspheres.

Droplet fusion technology is a key technology for many droplet-based biochemical medical applications. By integrating a symmetrical flow channel structure, we demonstrate an acoustics-controlled fusion method of microdroplets using surface acoustic waves. Different kinds of microdroplets can be staggered and ordered in the symmetrical flow channel, proving the good arrangement effect of the microfluidic chip. This method can realize not only the effective fusion of microbubbles but also the effective fusion of microdroplets of different sizes without any modification. Further, we investigate the influence of the input frequency and peak-to-peak value of the driving voltage on microdroplets fusion, giving the effective fusion parameter conditions of microdroplets. Finally, this method is successfully used in the preparation of hydrogel microspheres, offering a new platform for the synthesis of hydrogel microspheres.

Light-driven untethered soft actuators based on biomimetic microstructure arrays.

Soft actuators based on smart materials and structures that can perform more diverse tasks skillfully, are being intensively sought. Despite the good progress made in the past few years, locomotion and transportation functionalities of the untethered soft-bodied devices for various natural terrains remain challenging. Inspired by a gecko crawling system, an untethered soft actuator with the abilities of picking up, transporting, and delivering objects controlled by NIR light is proposed. The soft actuator consisting of photo-responsive MWCNTs units and mushroom shaped microstructures, was fabricated by an integrative soft-lithography method with inking and imprinting processes. The integrated MWCNTs unit can convert NIR light irradiation into thermal energy, which can make the body of the soft actuator generate a strong shape deformation intrinsically in a self-contained way, leading to a combined discontinuous and continuous locomotion. Moreover, the integrated mushroom shaped microstructures can also realize grasping and manipulation of the object that was not constrained by the object's shapes and sizes, which was further addressed from experimental and theoretical perspectives. Thus, the combined use of smart materials and structures opens up new research avenues and represents a step forward toward future applications of light-driven untethered soft actuators.

Three-Dimensional Manipulation of Micromodules Using Twin Optothermally Actuated Bubble Robots.

A 3D manipulation technique based on two optothermally generated and actuated surface-bubble robots is proposed. A single laser beam can be divided into two parallel beams and used for the generation and motion control of twin bubbles. The movement and spacing control of the lasers and bubbles can be varied directly and rapidly. Both 2D and 3D operations of micromodules were carried out successfully using twin bubble robots. The cooperative manipulation of twin bubble robots is superior to that of a single robot in terms of stability, speed, and efficiency. The operational technique proposed in this study is expected to play an important role in tissue engineering, drug screening, and other fields.

Amorphous Carbon-Induced Surface Defect Repair for Reinforcing the Mechanical Properties of Carbon Fiber.

Graphene oxide (GO) was prepared using metal-catalyzed crystallization of amorphous carbon on a carbon fiber surface to improve the mechanical properties of the carbon fiber (CF). The deposited GO was used for repairing of surface structure defects on CF, thereby improving the tensile strength and interfacial strength force of CF. The grown morphology of GO and the changes in CF surface microstructure before and after remediation were investigated in detail by scanning tunneling microscopy and Raman spectroscopy. The effects of surface repair on the mechanical properties of the CF and the resulting composites were investigated systematically. The results of scanning tunneling microscopy show that the graphene oxide formed on the surface of carbon fiber present uniform dispersion. Raman spectroscopy curves indicate that CF successfully remediated the defects in the CF surface. The results of mechanical properties testing show that such a remediation method could significantly enhance the tensile strength of CF and increase the interfacial strength versus raw fibers; that is, the tensile strength of CF was enhanced by 42% and the interfacial strength by 33.7%.

Design and development of an optical encoder with sub-micron accuracy using a multiple-tracks analyser grating.

In this paper, an optimized optical encoder based on generalized grating imaging is presented. A multiple-tracks analyser grating is proposed to eliminate the second and third harmonic signals, and a photodiodes array with optimized cell width is used to suppress the fifth harmonic signal. The photodiodes array also guarantees the consistency and stability of the encoder signals benefiting from single-field photoelectric scanning. High-quality encoder signals are expected to be obtained from the above optimization, thereby ensuring high encoder accuracy. In the experiment, measured encoder signals with the approximately ideal Lissajous figure are obtained. FFT analysis of the encoder signals shows that the second, third, and fifth harmonic distortions are smaller than 0.3%, 0.5%, and 0.1%, respectively. The calibration results of the optical encoder show that the positioning error within one signal period is ±0.12 µm, and the positioning error over 150 mm measuring range is within ±0.2 µm.

Fabrication of high edge-definition steel-tape gratings for optical encoders.

High edge definition of a scale grating is the basic prerequisite for high measurement accuracy of optical encoders. This paper presents a novel fabrication method of steel tape gratings using graphene oxide nanoparticles as anti-reflective grating strips. Roll-to-roll nanoimprint lithography is adopted to manufacture the steel tape with hydrophobic and hydrophilic pattern arrays. Self-assembly technology is employed to obtain anti-reflective grating strips by depositing the graphene oxide nanoparticles on hydrophobic regions. A thin SiO2 coating is deposited on the grating to protect the grating strips. Experimental results confirm that the proposed fabrication process enables a higher edge definition in making steel-tape gratings, and the new steel tape gratings offer better performance than conventional gratings.

A novel optical rotary encoder with eccentricity self-detection ability.

Eccentricity error is the main error source of optical rotary encoders. Real-time detection and compensation of the eccentricity error is an effective way of improving the accuracy of rotary optical encoders. In this paper, a novel rotary optical encoder is presented to realize eccentricity self-detection. The proposed encoder adopts a spider-web-patterned scale grating as a measuring standard which is scanned by a dual-head scanning unit. Two scanning heads of the dual-head scanning unit, which are arranged orthogonally, have the function of scanning the periodic pattern of the scale grating along the angular and radial directions, respectively. By this means, synchronous measurement of angular and radial displacements of the scale grating is realized. This paper gives the details of the operating principle of the rotary optical encoder, developing and testing work of a prototype. The eccentricity self-detection result agrees well with the result measured by an optical microscope. The experimental result preliminarily proves the feasibility and effectiveness of the proposed optical encoder.