摘要:An in situ method for online detection of dissolved acetylene (C2H2) in transformer oil is proposed using a microcavity photothermal (PT) sensor. A fiber-optic PT sensing probe integrating oil–gas separation and gas detection was designed based on PT spectroscopy. The sensing chamber has a volume of only 1 μL. The fiber-optic Fabry-Perot (F-P) microcavity functions simultaneously as a gas absorption cell and an optical interferometer. Near-infrared excitation and probe beams are transmitted through a single fiber via a wavelength-division multiplexer. The PT signal induced by gas absorption is detected using the F-P interferometer and retrieved through white-light interferometric phase demodulation. After erbium-doped fiber amplification, the excitation power reached 410 mW, and the C2H2 sensitivity was 0.064 pm/(μL·L-1). Experiments yielded a detection limit of 0.7 μL/L for the volume fraction of dissolved C2H2 in oil. The system offers a compact sensing chamber, immunity to electromagnetic interference, pump-free operation, and convenient field deployment, providing a practical approach to transformer-oil dissolved-gas monitoring.
关键词:photothermal spectroscopy;fiber-optic photothermal sensing;dissolved gas detection;trace gas detection
摘要:Laser-frequency drift and excessive linewidth limit rapid, cost-effective single-point measurements in cavity ring-down spectroscopy (CRDS). This work investigates the relationship between modulation frequency and demodulation-phase selection in frequency-modulation spectroscopy (FMS) and applies it to laser-frequency stabilization for acetylene CRDS. The modulation-frequency dependence of absorption- and dispersion-phase error signals was analyzed theoretically. The analysis identified an approximately 300 MHz crossover: the absorption phase dominated below 300 MHz, whereas the dispersion phase provided a larger error signal at higher modulation frequencies. An FPGA-based frequency-stabilization system was developed, and the measured peak-to-peak error-signal variation agreed closely with the theoretical prediction. After optimizing gas pressure and concentration, a proportional-integral-derivative controller locked the laser to the center of the acetylene absorption line, thereby improving frequency stability and reducing the effective linewidth. The stabilized laser was then evaluated in CRDS measurements. Allan-variance analysis yielded detection limits of 6.23×10-10 cm-1 at 9 MHz in the absorption phase and 2.51×10-10 cm-1 at 350 MHz in the dispersion phase, compared with 1.55×10-9 cm-1 without stabilization. These values correspond to improvements by factors of 2.49 and 6.18, respectively. Although high-frequency dispersion-phase detection achieved the lower detection limit, its advantage should be balanced against increased hardware requirements. The approach supports high-precision, low-cost gas monitoring in environmental and industrial sensing applications.
关键词:laser spectroscopy;cavity ring-down spectroscopy;frequency modulation spectroscopic;laser frequency stabilization;trace gas detection
摘要:Weak absorption and spectral overlap limit trace isotope detection under low-pressure dynamic conditions. To address this issue, a wavelength modulation spectroscopy (WMS) method is proposed for hydrogen and oxygen isotope measurements in water vapor. A distributed feedback (DFB) laser centered at 1.863 μm was employed, and a miniaturized detection system with a 12 m optical path and an 80 cm³ absorption cell was developed. By means of a single scan, simultaneous detection of H216O, H217O, H218O, and HDO was achieved. Calibration experiments were performed using different standard water samples, and relationships were established between the fitted integrated area ratios and isotope reference values. Hydrogen and oxygen isotope ratios, together with the corresponding δ values, were then obtained. The fitted integrated areas exhibited linear relationships with water vapor pressure, with correlation coefficients exceeding 0.99. Under a steady pressure of 340 Pa, the optimal integration times were determined to be 18 s, 23 s, and 20 s for δD, δ18O, and δ17O, respectively. The minimum Allan deviations were 2.015‰, 3.636‰, and 1.317‰. Simulated lunar-soil experiments further verified dynamic isotope monitoring, and isotope variations during adsorption fractionation were successfully captured. The 10%-90% pressure response time was 71 s. A comparison experiment was also conducted using a direct absorption system at 2.731 μm as a reference, with an optical path length of 25 m. Consistent δD variation trends were observed in both systems. The proposed method enables steady-state and dynamic measurements of hydrogen and oxygen isotopes in water vapor under low-pressure conditions, thereby providing technical support for miniaturized in situ isotope detection in dynamic environments.
关键词:low-pressure water vapor;hydrogen and oxygen isotopes;wavelength modulation spectroscopy;isotope retrieval
摘要:To address the limitation that traditional TDLAS systems fail to meet the millisecond-scale transient combustion measurement requirements of engines because of processing delays introduced by complex spectral line fitting and numerical inversion algorithms, a rapid hardware-based spectral processing system using logarithmic amplification is proposed and developed. The study covers the overall system architecture, the design of an automatic background radiation subtraction circuit, the logarithmic amplification circuit, algorithmic correction methods, and the real-time data processing framework. In the analog front end, a proprietary background radiation subtraction circuit and a logarithmic ratio amplifier based on the LOG114 chip are employed to directly output a voltage signal proportional to absorbance. In the digital section, a field-programmable gate array (FPGA) is used for high-speed data acquisition and absorbance integration accumulation, followed by temperature inversion calculations implemented on a microcontroller. Experimental results show that, at a scanning frequency of 100 kHz, stable background radiation subtraction is achieved, and a single temperature calculation requires approximately 2.5 ms. The system is therefore capable of rapidly measuring high-temperature flow fields at millisecond intervals and is expected to capture transient combustion variations in engines.
关键词:TDLAS;combustion field;high frequency response;logarithmic ratio amplification;background radiation subtraction;quick measurement
摘要:Accurate non-contact measurement of the main cable void ratio remains challenging under strong sunlight, high temperature, and high humidity during outdoor cable-compaction construction. To address this issue, a main-cable void-ratio measurement system based on cooperative acquisition by multiple area structured-light cameras was developed. Four area structured-light cameras were arranged at 90° intervals around the main cable to acquire full-circumference point clouds of the cable surface. To improve imaging stability and environmental adaptability under complex outdoor conditions, mechanical shading, a thermoelectric cooling unit, and a window heating film were integrated into the system. For multi-camera calibration, Zhang's calibration method constrained by field-of-view overlap was employed to obtain the global extrinsic parameters of the four cameras, unify the multi-camera coordinate system, and achieve high-precision point-cloud registration. Experiments demonstrated that the system could stably acquire high-quality main-cable point clouds at 50 ℃, high humidity, and an illumination intensity of 105 lx. The average reprojection error of the multi-camera joint calibration was less than 0.7 pixels, millimeter-level spatial registration of the main-cable point cloud was achieved, and the relative error in main-cable cross-sectional area measurement was below 0.3%. These results demonstrate the feasibility and engineering applicability of the proposed system for non-contact precision measurement of main-cable void ratio in outdoor cable-compaction construction.
摘要:Semiconductor lasers emitting at 852 nm are essential light sources for cesium-based atomic systems because their wavelength is resonant with the Cs D2 transition. They support cesium atomic clocks, quantum sensing, cold-atom manipulation, and precision spectroscopy. This review addresses the need for a systematic assessment of 852 nm laser technologies by summarizing recent advances in quantum-well material systems, device architectures, performance optimization, and engineering applications. The development of AlGaAs/GaAs, aluminum-free InGaAsP/GaAs, and InAlGaAs/GaAs active-region platforms is first reviewed. The effects of material design on carrier confinement, facet reliability, thermal stability, and electro-optical conversion efficiency are analyzed. Major device configurations—including vertical-cavity surface-emitting lasers, distributed-feedback lasers, external-cavity diode lasers, vertical external-cavity surface-emitting lasers, and high-power edge-emitting lasers-are then evaluated according to their structural characteristics and performance metrics. Their relative capabilities in linewidth, output power, slope efficiency, single-mode stability, and system-level integration are compared. Commercial 852 nm semiconductor laser products are further assessed from an application-oriented perspective. Current domestic activities are concentrated primarily on product distribution, module packaging, and system integration, whereas core laser chips and gain chips remain largely dependent on overseas suppliers. Future development should prioritize narrower linewidths, higher output power, improved electro-optical efficiency, wider operating-temperature ranges, enhanced reliability, and chip-scale integration. These advances are expected to support next-generation cesium atomic clocks and quantum precision-measurement systems.
关键词:semiconductor lasers;quantum well structure;core performance metrics;cesium atomic clocks
摘要:Double-mirror components are widely used in advanced manufacturing equipment and complex optical instruments. However, complete three-dimensional (3D) surface-form measurement is challenging because the two functional surfaces typically have a large angle between their surface normals. When phase-measuring deflectometry (PMD) is applied separately to the two surfaces, the reconstructed point clouds are expressed in independent coordinate systems. The lack of a common physical datum prevents determination of their relative spatial transformation and degrades point-cloud registration accuracy. A virtual-datum calibration method is proposed to establish a common coordinate system without requiring a physical reference shared by the two subsystems. The method uses virtual reference planes retained during subsystem calibration as datum-transfer carriers. Phase-reconstructed measurements of a plane mirror acquired at multiple poses are combined with turntable-coordinate-system constraints to determine the spatial transformation between the two virtual reference planes. The two surface datasets are thereby transformed into a unified coordinate system for 3D reconstruction. Experimental validation achieved point-cloud registration of the double-mirror surfaces. The right surface exhibited a root-mean-square (RMS) deviation of 0.059 mm relative to a reference plane, confirming the feasibility of complete 3D surface-form reconstruction. The method provides a practical calibration strategy for precision measurement of multi-surface reflective components.
关键词:phase measuring deflectometry;double mirror objects;virtual datum calibration;specular surface metrology
摘要:Three-dimensional (3D) nanoprobe systems are critical components of micro/nano coordinate measuring machines; however, the measurement capability of existing probes remains insufficient for demanding practical applications. To address this limitation, a 3D nanoprobe based on capacitive displacement sensing is developed and experimentally evaluated. The probe comprises capacitive displacement sensors, electrode plates, a compliant mechanism, and a stylus. Three high-precision capacitive displacement sensors are arranged in an equilateral-triangular configuration. A geometric model converts the sensor outputs into the displacement of the probing sphere and determines the probe orientation. Experimental results demonstrated a 3D resolution better than 3 nm. The measurement range reached 260 μm in the horizontal direction and 70 μm in the vertical direction. Contact repeatability was better than 9.4 nm horizontally and 8.7 nm vertically. These results confirm the suitability of the proposed probe for micro/nanoscale dimensional measurement and support its application in precision manufacturing and inspection.
摘要:This study aims to systematically investigate the mapping relationship between wafer-level uniformity in lithography and dry etching processes, and the physical morphology and optical performance of silicon photonic passive devices on an 8-inch silicon-on-insulator (SOI) platform. The experiments were conducted on a 220 nm top-silicon SOI platform. A two-step full-process fabrication flow was implemented using deep ultraviolet (DUV) lithography and inductively coupled plasma reactive ion etching (ICP-RIE). The process included the fabrication of grating couplers with a 70 nm shallow etch and single-mode straight waveguides with a 220 nm full etch. The physical dimensions and optical propagation loss were systematically characterized across five 8-inch wafers using atomic force microscopy (AFM), critical-dimension scanning electron microscopy (CD-SEM), and an automated optical probe station. A certain degree of wafer-to-wafer stability was observed among the five wafers. Statistical analysis based on standard deviation, semi-range, and their relative variations indicated that the average etch depth and average linewidth remained within relatively stable ranges across different wafers. However, within-wafer (WIW) radial process variations were observed. The reduction in microscopic dimensions in the edge region, together with possible sidewall morphology degradation, may alter the optical mode distribution and increase sidewall scattering loss. As a result, the propagation loss of the single-mode waveguides exhibited an increasing trend from the wafer center toward the edge. These results suggest that macro-loading effects and lateral etching behavior during dry etching of large-diameter wafers are important factors influencing within-wafer performance consistency in silicon photonic chips. The findings provide experimental evidence and process optimization guidance for future optical proximity correction (OPC), etching recipe optimization, and wafer-level yield improvement on 8-inch SOI platforms.
摘要:To accurately capture the amplitude and direction of dynamic structural displacement variations, a high-sensitivity fiber Bragg grating (FBG) displacement sensor based on a boss–groove structure is proposed to enhance the strain response of a uniform-strength beam. The sensor provides bidirectional displacement sensing capability. Its measurement principle, structural design, temperature compensation method, and prototype performance evaluation were systematically investigated. Experimental results show that, over a measurement range of ±60 mm, the sensor achieves a sensitivity of 63.446 pm/mm and a comprehensive measurement accuracy of 0.145 9 mm, while exhibiting favorable stability, repeatability, hysteresis performance, temperature adaptability, and creep resistance. Field application results further indicate that three deployed sensors operated stably and effectively captured structural deformation information for support segments in a subway tunnel in a certain city. The proposed sensor is therefore suitable for precise bidirectional displacement monitoring in civil and mechanical engineering applications.
摘要:High-precision pose measurement of circular holes is essential for ensuring coaxiality during hole-shaft assembly of precision components, including micro-thermocouples. For micro thick-walled holes with diameters of 0.03-3.7 mm and a structural constant k>2.5, conventional three-dimensional vision systems are difficult to implement because of their complex hardware and imaging models. Moreover, in monocular backlight imaging, wall occlusion produces blurred light-spot edges at tilted poses, thereby reducing measurement robustness. A dual-mode pose-measurement method based on grayscale integral projection is proposed for monocular backlight imaging of thick-walled circular holes. A geometric backlight-imaging model establishes the relationship between light-spot morphology and hole inclination. For coarse alignment, grayscale integral projection converts the two-dimensional image into a one-dimensional transmitted-flux distribution, suppressing interference from locally blurred edges. A linear inverse model is applied at inclination angles exceeding 70% of the theoretical cutoff angle to rapidly correct large angular deviations. For fine alignment, a closed-loop extremum-search algorithm exploits the maximum transmitted flux at the zero-pose condition to achieve sub-degree pose positioning. Experiments conducted under large-inclination and burr-interference conditions achieved a detection success rate exceeding 91%. The proposed method provides a low-cost and robust solution for circular-hole pose alignment in precision assembly systems.
关键词:monocular vision;pose detection;thick-walled circular holes;grayscale integral projection;precision assembly;dual-modal method
摘要:Single-exposure Western blot (WB) imaging is limited by a restricted dynamic range, saturation of high-abundance protein bands, and loss of low-abundance signals. A dual-channel morphology-guided multi-exposure fusion algorithm (MGAWF) is proposed to address these limitations. A multi-cue weighting model incorporating adaptive exposure, morphological saliency, and spatial attention is designed for WB multi-exposure image sequences. The final weight map is generated through nonlinear gain adjustment and normalization. In the dual-channel pyramid framework, morphology-guided gating enables nonlinear multiscale feature extraction in the image-decomposition channel, whereas a Gaussian pyramid smooths the weight map across scales. The fused high-dynamic-range image is reconstructed by multiscale linear fusion followed by inverse Laplacian-pyramid transformation. Experimental results yielded a standard deviation of 20.62, a spatial frequency of 3.92, and a sum of correlations of differences of 0.80. The method suppressed halo and artifact formation near high-contrast band edges, reduced background noise, and preserved low-abundance band signals. Quantitative optical-density analysis indicated improved response sensitivity and an expanded effective dynamic range. MGAWF provides a quantitative image-reconstruction approach for multi-exposure WB imaging and supports reliable protein-band analysis over a broader signal range.
摘要:Steel surface defect detection is challenged by complex industrial backgrounds, which weaken low-contrast defect representations, as well as by the loss of local structural details during cross-scale feature fusion. To address these issues, an improved detection framework, termed DFD-DETR, is proposed, integrating a dual-frequency-domain enhancement strategy with a dynamic kernel guidance mechanism. Specifically, a Dual-Frequency Domain Enhancement (DFE) module is designed to suppress low-frequency background interference caused by rolling textures while enhancing the high-frequency edge responses of subtle defects. The DFE module consists of a pre-stage high-pass filtering module and a post-stage spectral gating module. In the pre-stage, low-frequency background components, such as rolling textures, are suppressed by a fixed high-pass frequency response, whereas in the post-stage, the fused frequency-domain features are adaptively recalibrated through a learnable spectral gating mechanism, thereby strengthening the edge responses of subtle defects. To further mitigate the loss of fine-grained information during multi-scale feature interaction, a Dynamic Kernel Guidance (DKG) module is introduced. Spatially adaptive dynamic convolution kernels are generated from high-level semantic features, and locally guided aggregation is performed on high-resolution shallow features, thereby enhancing structural representation across multiple defect scales. Extensive experiments on the NEU-DET and GC10-DET datasets demonstrate that the proposed method achieves mAP@0.5 scores of 80.8% and 73.4%, respectively. Compared with the baseline RF-DETR, the proposed framework exhibits superior performance in low-contrast and multi-scale defect detection under complex backgrounds, indicating improved detection capability and generalization performance.