According to the mounting style and application requirements of electro-optic payload (EO) on small UAV platform, a technical scheme of forward-ground wide-area scanning based on roll-pitch structure EO was proposed. The mathematical model of scanning state was constructed, the algorithm for optimizing control parameters to maximize scanning width of EO based on platform parameters was proposed, the rationality of the algorithm was analyzed by example data, and the real-time performance of the algorithm was measured. The technical status of forward-ground wide-area scanning was tested by simulation and flight test. The results show that the control parameters obtained by the proposed algorithm are optimal, it takes less than 40 ms to run on the target embedded platform and can optimize online in real-time. The proposed scanning technology can still achieve forward-ground continuous coverage under certain platform disturbances, and the overlapping rate of the field of view (FOV) is variable during scanning. The closer to the scanning edge area, the smaller the overlapping range of adjacent FOV. The scanning range should select the area sandwiched between the scanning control points at both ends of transverse direction to avoid leakage due to small overlapping rate in scanning edge area.

Star sensor provides high-accuracy attitude information to satellite. With the continuous development of satellite internet platform technology in the national defense and civil fields, the star sensor is required to develop towards the small volume, light weight and high precision. Therefore, the optical system of star sensor should be designed in accordance with the application requirements. The relationship between color temperature and irradiance of sixth-magnitude star in the same integral band was established by using blackbody radiation law, thus confirming the effective aperture. Based on the Gaussian model of star point, it was proposed to realize the small aperture and high sensitivity detection by increasing the energy concentration of the optical system. The working field of view of star sensor was determined on the basis of Monte Carlo method. The technology requirements of stray light suppression was confirmed according to sensitivity. The optical system was designed based on the above mentioned verification results. The energy concentration of 3 pixels×3 pixels is better than 95%. In the range of ?40 ℃ to 60 ℃, the centroid shift of star point is less than 0.40 μm. At the stray light suppression angle of 32°, the extinction ratio of the system is 3.0E-8. In the last part, the rationality of the optical system design was verified through the accuracy calibration of centroid in laboratory, test of accuracy and sensitivity in out-field stargazing, and stray light test. The test results show that the calibration accuracy of the star sensor is 1.45″, the accuracy of out-field stargazing is 4.2″ (3σ), the limiting sensitivity is 6.04 magnitude star, and the mean value of the gray level of stray light background is 36.51.

The performance of near-field goniophotometer (NFG) was determined by the imaging accuracy of the luminance meter, data capacity of the disk and computing power of the computer. With the development of imaging technology and computer technology, the performance of NFG has been improved dramatically. The theoretical principle of NFG was introduced, and the merits of NFG compared with far-field goniophotometer were summarized. The development status of NFG was reviewed, while some typical NFGs at home and abroad were compared. In view of the current situation of NFG, including insufficient accuracy of metadata, slow scanning reconstruction speed and limited traceability, the possible solutions from the aspects of components, software and systems were presented. Combined with the characteristics and advantages of NFG, its applications in lighting, display, transportation, industrial vision and cultural relic protection were prospected.

The spatial combination of four near-infrared laser beams was demonstrated by combining polarized beam and spectral beam, which included solid nanosecond pulse laser, high-power continuous wave (CW) fiber laser, high repetition rate nanosecond pulse fiber laser and sub-millisecond (long pulse) fiber laser. The simulation software was developed to design and optimize the main relevant parameters such as spectral characteristics and polarization characteristics of polarization beam combiner, spectral beam combiner and each laser. Based on the simulation results, a compound laser system composed of the above four laser systems and a set of polarization/spectral beam combination links was developed. The beam combination efficiency of the system is up to 93.9%, which realized the coaxial emission of multiple pulses and CW laser beams in the near-infrared band. The system can be used to study the cooperative effect of the laser-material interaction.

A new optical element was proposed, which could generate a diffraction-free beam with a certain distance from the element. The cross section of an ordinary axial pyramid was an isosceles triangle and the element was designed by defining a parabola on the two waists. The 3D printing technology was used to fabricate optical elements with different structures. The results of simulation calculation using angular spectrum theory and the experiment conducted at 0.1 THz frequency clearly show that there is a certain distance between the generated diffraction-free beam and the element. The beam behavior of the element was analyzed according to the spatial frequency spectrum. The results show that the element generates a new type of diffraction-free beam, and according to the space-frequency shape of the element, the diffraction-free beam generated by the element is named comb-like space-frequency diffraction-free beam.

A sort of common path and integrated machine design on TV camera with dual field of view and laser ranging reception was designed. According to the application requirements, the work on device selection about CMOS and avalanche tube was carried out at first. Then, the work on optical, mechanical, and electrical design was carried out. By means of the shared previous objective lens group, the spectroscope coated beam-splitting film and relay lens group were used to achieve the design of common path optical system. The big field of TV camera view was used for searching targets, the small field of TV camera view was used for tracking targets, and the unit of laser ranging reception was used for receiving laser echo. The body tube and the complete machine shell were designed integrally and processed by monoblock casting way, and the whole structural strength of integrated machine was improved. The startup chip for dual H-bridge motor was designed parallelly to improve the operating rated current, and according to the detection results of limit switch, the precise limit control of transmission components by electrical machinery was achieved. Analysis of the design results shows that the image quality of the big/small field of view of integrated machine is good, the image plane is stable, and the unit of laser ranging reception maintains a good working state.

In order to meet the needs for the performance of the primary-fourth double-sided integrated mirror in a certain on-axis four-mirror infrared optical system, the mirror body needs to have higher stiffness with high lightweight rate to keep the two mirror face shapes stable. The topology optimization method was used to optimize the initial model of double-sided integrated mirror with the maximum stiffness was set as the optimization objective, and the volume fraction of the design domain was constrained. The optimization results show that the stiffness is higher while the primary mirror and the fourth mirror are connected through ribbed plate arranged radially around the optical axis. The optimal size parameters for the primary-fourth and ribbed plate were obtained by free size optimization. Based on the optimization results, the primary-fourth double-sided integrated mirror was designed, and the lightweight rate achieved 82.4% by using the final solution. The mode of the primary-fourth double-sided integrated mirror and the influence of 1 g gravity on the root-mean-square (RMS) of the surface shape were analyzed, a first-order mode frequency achieved 417 Hz, and the maximum change of surface shape under gravity was 1/22 λ （λ= 632.8 nm). The analysis results show that the optimized mirror body has sufficient stiffness and good resistance to the deformation under gravity. The mirror was processed and tested according to the final design, and the measured results show that the RMS of primary-fourth mirror surface shapes are both about λ/7, which meets the needs for the performance of infrared optical systems, proves the effectiveness of this topology optimization design method, and provides some references for the optimization design of double-sided integrated mirrors.

Augmented reality (AR) display systems are developing towards ultra-miniaturization with the use of micro-light emitting diodes (Micro-LED) as miniature image sources in AR display technology. However, the Micro-LED with a large luminous angle, the collimating lens with a small size and high image quality, and the low optical efficiency at the edge of the field of view (FOV) and nonuniform illumination by vignetting of the collimating lens pose challenges for the design of AR optical systems. To solve these problems, a collimating lens was designed with high uniformity of illumination and high optical efficiency of the edge field of view for the diffraction grating waveguide by adding the calculated aspherical field lens. Then, by analyzing the theory in detail, its surface shape parameters were solved, and the lens with relatively uniform illumination was designed. The new collimating lens had a diagonal FOV of 41.2°, an F/# (F#) of 1.87, and a modulation transfer function greater than 0.5 at 125 lp/mm. The simulation results show that the illumination uniformity is relatively improved by 14%, and the optical efficiency of the edge FOV is relatively increased by 15%. The lens has good imaging performance, and can be applied in the ultra-small AR diffraction grating waveguide head-mounted display system.

Aiming at the measurement of photoelastic modulator (PEM) in ultra-high speed full polarization field, a circular PEM with target frequency near 100 kHz was studied, which was composed of an elastic-optic crystal and two piezoelectric actuators with azimuth difference of 45°. The resonant frequency of the elastic-optic crystal and the vibration frequency of the piezoelectric actuators were calculated theoretically. When the two frequencies were consistent, the PEM worked in the resonant state, and the modulation efficiency of PEM reached the highest. When the PEM reached the resonant state and tended to be stable, the amplitude and phase of the driving voltage of the two piezoelectric actuators were adjusted to achieve two special modulation modes of pure standing wave mode and pure traveling wave mode. The pure standing wave mode could realize the delay and azimuth of the modulated fast axis under pure electric control, and the pure traveling wave mode could realize the high-speed rotation of the modulated fast axis under pure electric control at the half-frequency speed of the PEM resonant frequency. Finally, the COMSOL finite element simulation was used to simulate the two modulation modes of PEM, and the vibration mode and fast axis direction of PEM in two modulation states were verified. It shows that the PEM can realize multifunctional elastic-optic modulation, which provides theoretical support for the subsequent field of light polarization.

In view of the problem of low contrast of infrared sea-skimming small target and easy interference from noise, sea clutter and cloud, which leads to tracking failure, a tracking method of infrared sea-skimming small target based on sea-sky line was proposed. Through analysis of infrared imaging system, the Pierre-Simon Laplace filter was used to suppress background noise, and according to position information of sea-sky line in image, the target area was determined and the target was detected. After detecting the suspicious target, the real target was selected by using feature information such as gray level, size and area. Experimental results show that the proposed algorithm has a good tracking effect for the infrared sea-skimming small target.

针对低慢小无人机探测任务中精度不高、在嵌入式平台上部署实时性能差的问题，提出了一种基于改进YOLOv4的小型无人机目标检测算法。通过增加浅层特征图、改进锚框、增强小目标，提高网络对小目标的检测性能，通过稀疏训练和模型修剪，大大缩短了模型运行时间。在1080Ti上平均精度（mAP）达到85.8%，帧率(FPS)达75 frame/s，实现了网络轻量化。该模型部署在Xavier边缘计算平台上，可实现60 frame/s的无人机目标检测速度。实验结果表明：与YOLOv4和YOLOv4 - tiny相比，该算法实现了运行速度和检测精度的平衡，能够有效解决嵌入式平台上的无人机目标检测问题。

An underwater image enhancement algorithm based on multi-scale residual attention network was proposed for the problems of color shift, color fading and information loss of underwater images caused by water scattering and absorption. An improved UNet3+-Avg structure and attention mechanism was introduced by the network, and the multi-scale dense feature extraction module as well as the residual attention recovery module were designed. In addition, a joint loss function combining Charbonnier loss and edge loss enabled the network to learn rich features at multiple scales, improving the image color while retaining a large amount of object edge information. The average peak signal-to-noise ratio (PSNR) of the enhanced images reaches 23.63 dB and the structural similarity ratio (SSIM) reaches 0.93. Experimental results with other underwater image enhancement networks show that the images enhanced by this network achieve significant results in both subjective perception and objective evaluation.

In holographic 3D display and holographic video applications, reducing the amount of information in holograms to meet the requirements of communication bandwidth is an important research task. The wavelet threshold and quantization approaches were used to lossy compress holograms, and the effects of using different wavelet basis thresholds on the reconstructed image quality and compression ratio of holograms were compared and analyzed. The research results show that after lossy compression of the wavelet threshold of off-axis digital hologram, the reconstructed image quality obtained by Haar and db5 wavelet bases is the highest. Its peak signal to noise ratio (PSNR) reaches 26.38 dB, and the compression ratio obtained by using sym2 and db2 wavelet is the highest, reaching 6.68. After lossy compression of the wavelet threshold of phase hologram, the reconstructed image quality obtained by Haar and Daubechies wavelets is the highest, with a PSNR of 32.32 dB. The compression ratio obtained by using the Biorthogonal wavelet is the highest, reaching 4.47. The research results provide a reference for how to effectively use the wavelet transform to achieve hologram compression.

To address the problem that infrared images and visible images have different feature expressions in different scenes, an saliency-based dual discriminator generative adversarial network method was proposed to fuse the infrared and visible feature information. Different from the traditional generative adversarial network, a dual discriminator approach was adopted to discriminate the saliency regions in the source images and the fusion images respectively in this algorithm, and the saliency regions of the two source images were used as the input of the discriminator so that the fusion image retained more salient features. The gradient constraint was introduced into its loss function so that the salient contrast and rich texture information could retain in the fusion image. The experimental results show that the proposed method outperforms other comparison algorithms in four evaluation indexes: entropy (EN), mean gradient (MG), spatial frequency (SF) and edge intensity (EI). This study achieves efficient fusion of infrared images and visible images, which is expected to gain applications in fields such as target recognition.

In the measurement of radial shearing interference, the center-offset between beam expansion and beam shrinkage situation is a frequent event. In the case, the traditional wavefront reconstruction algorithms will introduce large reconstruction errors, and their applications are also challenged. The iterative wavefront reconstruction algorithm for center-offset radial shearing interference was derived. Through theoretical simulation, the influences of combined aberration, single-order Zernike aberrations, center offset and its calculation errors on reconstruction algorithms were analyzed. The results show that the proposed algorithm can accurately reconstruct the wavefront phase in the case of center offset between shearing beams, and its reconstruction accuracy is significantly better than that of traditional algorithm. Under the simulation conditions, the root mean square (RMS) value of wavefront error of the proposed reconstruction algorithm is not higher than 0.05 λ with the increase of center offset. However, the reconstruction error of traditional algorithm reaches 1.6 λ (RMS value) and is approximately linearly related to the center offset.

The selection of suitable hyperparameters in wavefront sensorless adaptive optics systems is the key to achieve the best performance of iterative control algorithms. Existing iterative control algorithms for hyperparameter setting generally use the traversal method, which is easy to understand and implement, but is computationally intensive and time-consuming, and may also miss the global optimum because of finding a local optimum. A Bayesian optimization method was adopted for selecting hyperparameters suitable for iterative control algorithms of adaptive optics systems. The commonly-used stochastic parallel gradient descent algorithm (SPGD), Momentum-SPGD and CoolMomentum-SPGD control algorithms were used as examples to compare and analyze the calibration effects of control algorithms using the traversal method and Bayesian optimization method to select hyperparameters, respectively. The results show that the advantages of using Bayesian optimization method for hyperparameter selection were obvious. For the SPGD control algorithm, the number of sample instances required for the Bayesian optimization method is 10% of that for the traversal method when the same convergence effect is achieved, and for the Momentum-SPGD and CoolMomentum-SPGD control algorithms, the number of sample instances required for the Bayesian optimization method is 7% and 9% of that for the traversal method, respectively. The above findings can provide a theoretical basis for hyperparameter setting in the practical application of iterative control algorithms for adaptive optical systems.

The curved bionic compound eye camera has the characteristics of large field of view and sensitivity to moving objects, and has broad application prospects in the fields of national defense, medicine and civil industry, such as wide-area detection, endoscopic catheter, precision guidance, etc. The research on velocity measurement technology of high-speed moving target based on the curved bionic compound eye camera was reported. Firstly, a velocity measurement model based on the principle of curved bionic compound eye imaging was established, then the position of each sub-eye was found on the original compound eye image by using the Hoff circle detection algorithm, and the overlapping field of view between adjacent sub-eyes was analyzed by image segmentation. Finally, the sub-images of the two adjacent sub-eyes were extracted and matched with the SIFT algorithm, and the number of pixels in the overlapping area was calculated from the pixel coordinates of the extracted feature points, so as to get the moving speed of the car. Further, the relevant velocity measurement experiment was designed to verify. In the experiment, the shooting distance was controlled between 6 m and 10 m. The velocity test was carried out on the vehicle moving in a uniform straight line at a speed of 8.33 m/s. The test results show that the relative error of the velocity test can be controlled within 4%. Compared with the velocity measurement results of single aperture imaging system, the curved bionic compound eye velocity measurement technology has higher accuracy and repeatability.

The traditional optical gas detection method needs to convert the time domain signal to frequency domain signal to obtain the concentration of gas, which will introduce errors in the conversion process and cannot accurately reflect the change of frequency with time, resulting in the gas concentration cannot be accurately calibrated. A detection method based on frequency domain absorption peaks to calibrate gas concentration was proposed. Taking CO2 as an example for detection, the concentration of CO2 in the mixed gas of CO2 and N2 was determined by using the direct absorption method based on the Beer-Lambert law. This method obtained the peak intensity of the absorption spectrum by collecting the light intensity and frequency information of the laser beam and performing the Lorentz fitting by the adjacent averaging method and Levenberg-Marquardt (LM) optimization algorithm. The experimental results show that the linear relationship between laser peak intensity and gas concentration is satisfied, and the Pearson correlation coefficient can reach 0.999. Therefore, the detection method based on the frequency domain absorption peak calibration of gas concentration can effectively simplify the system structure, avoid the effect of frequency drift in the time domain to frequency domain conversion, and make the linearity between laser peak intensity and gas concentration good.

In response to the issue of accuracy in measuring the elevation of spatial coordinate points relative to a reference point, a new combined elevation measurement method based on a physical horizontal reference plane was proposed. By utilising the concepts of laser triangulation and leveraging the geometric features of measuring light pathways, a high-precision elevation measurement system was built. This technique employed a physical horizontal reference. Within a Lambertian scattering light field, an analysis was done to determine the influence of tilt angle modifications on the centroid displacement of laser energy. An related error correction model was built, which permitted tilt angle error compensation for elevation data by the use of a tilt angle sensor. Additionally, an autonomously constructed calibration platform was deployed for verification studies targeted at calibrating the suggested elevation measuring system. The findings demonstrate that the proposed combined elevation measurement method achieves a repeatable measurement error within a fluctuation range of ±20 μm across a 500 mm measurement range. Compared to the measurement methodology of single laser displacement sensor, the combined elevation measurement method greatly reduces the fluctuation of measurement errors. It is capable of supporting precise measurements of spatial coordinate points in any orientation and effectively boosting the performance of laser displacement sensors in the area of elevation measurement, consequently giving more genuine and effective measurement results.

A novel measurement system based on Hanning-window-changing density fringes and two-dimensional vision was proposed for monitoring the tank settlement. The artificial Hanning-window-changing density fringes were attached to the surface of the tested storage tank, and the surface fringes were collected in real time by a far-focus imaging camera at a certain distance, and then the images were processed by a specific image processing algorithm to extract the settlement displacement of the storage tank. The image sequences of tank settlement under different conditions were simulated by SOLIDWORKS 3D modeling, motion simulation and image imaging module, and the main measurement performance influencing factors, measurement feasibility and accuracy were simulated and analyzed. The simulated experimental measurement system of tank settlement was constructed, and its measurement performance was analyzed experimentally. The results show that the proposed system can realize real-time monitoring of the tank settlement displacement, and the measurement accuracy can be better than 10 μm at a distance of 7.5 m.

The hard tissue of fish, especially the skeleton, is the tissue that supports fish body and protects its internal organs. The characteristic detection and analysis of the hard tissue is the data basis for studying fish swimming motion, fish anatomy, fish body modeling and so on. With the development of X-ray technology and the wide application of domestic equipment, the cost of instruments and equipment has been significantly reduced, which makes the application of X-ray in fishery research and automatic production possible. The basic principle of X-ray technology and its application in the detection of fish body tissues were first introduced. The application of X-ray technology in the field of fishery was mainly divided into nondestructive detection of fish tissues and organs, and identification of trace elements in fish body. X-ray technologies, including photography, digital imaging, diffraction and X-ray absorption spectroscopy, were introduced respectively. Then, their applications in the modeling of fish body tissues and organs, fish skeleton detection, fish fossil research, otolith analysis and trace element detection of fish body were reviewed, and problems existing in the application of X-ray in the field of fishery were summarized. Finally, the fishery application of X-ray was prospected.

The commonly-used single vision lens has the problem of variational refractive power when it is worn and actually perceived. Response to this problem, a new design method based on lens-eye joint model for single vision lens was proposed. According to the new design method, the variational visual angles were matched to different areas of lens, and the constant perceived diopter on the corneal surface was set as precondition. The diopter on the lens surface was recomputed, and finally the diopter distribution and the corresponding surface shape data were obtained. Based on the proposed design method, the diopter of ?6.00 D actually felt on the corneal surface was taken as an example to generate surface shape data for lens manufacturing, and the surface shape and diopter of the manufactured lens were measured and evaluated. The test results show that the central diopter of the manufactured lens is ?6.30 D, the diopter values gradually vary from the center to the edge of the manufactured lens, and the maximum diopter error is smaller than ±0.23 D, which verifies the effectiveness of the proposed designed method.

Aiming at the problems of large size and poor channel selectivity of traditional multi-channel narrow-band filters, a multi-channel narrow-band filter composed of distributed Bragg reflector (DBR) and metasurface micro-nano array was designed. The effects of the period of one sided DBR medium and the refractive index of the defect layer on the transmission spectrum were analyzed by using the transfer matrix method, and the structure of the narrow-band filter was optimized. The unit structure of the narrow-band filter was built by Comsol Multiphysics software for simulation, and the influence of the metasurface and the incident angle on the filtering characteristics of the narrow-band filter was studied. The simulation results show that the central wavelength of the narrow-band filter can be tuned by adjusting the edge length of the metasurface, and a transmission band with narrow bandwidth and high transmittance can be obtained in the red light band from 634 nm to 714 nm. The central wavelength of each channel can be adjusted. By adjusting the side length of the metasurface, the narrow-band filtering of multiple spectral channels can be achieved. The changes in the small angle range (0°~10°) of the incident light have little effect on the filtering performance. The results provide a new idea for the design of multi-channel narrow-band filters.

The lidar operates at 905 nm is considered to have essential application values in automotive autonomous driving. The antireflection coatings made by magnetron sputtering commonly used in industry have the problem of insufficient wear resistance, which is difficult to adapt to the harsh outdoor environment. The sol-gel method combined with high-temperature curing process was used to provide a solution of high wear resistance and low cost for antireflection coatings applied to vehicular lidar which operated at 905 nm, and the influences of curing temperature, concentration of hydrochloric acid (HCl), water content in weak acid environment on refractive index of TiO2 were studied. The optical and mechanical properties of the coatings with different underlying curing methods were evaluated from the reflectivity, roughness, hardness and wear resistance. The results show that the coatings have excellent anti-reflection performance at the wavelength of 905 nm, the reflectivity is less than 1% when the incident angle is 15° and less than 5% when the incident angle is 60°, and the minimum roughness is 0.005 μm, which generally meets the optical requirements of the lidar with a horizontal angle of 120°. The coatings show the excellent capability of enduring harsh environments, with pencil hardness of 8H and withstanding 8 000 times of reciprocating friction without perceptible damages.

Beam quality factor M2 is the main parameter to characterize the transverse-mode property of high power laser. In view of the problem that beam quality analyzer can only be used to evaluate the beam quality of small caliber and low power laser, the principle and simulation of attenuation and compression technology of beam quality measurement of high power laser were studied. The simulation model of attenuation and compression module was established, and the thermal aberration of optical component under high power laser was studied by finite element method. The results show that the effect of the thermal aberration on the beam quality factor is less than 5% when the peak and valley (PV) value is less than 82 nm. If depolarization occurs when the beam passes through the attenuation component, the beam quality factor will be smaller. The influence of wavefront distortion caused by compression module on beam quality factor measurement was analyzed based on Zernick polynomial and beam quality factor calculation model. Through the Zemax simulation, the influence on beam quality factor measurement is less than 5% when the field of view between the incident light and the center optical axis of the attenuation and compression module is less than 7 °.

Obtaining the light source with high uniformity and high stability is the key technology for the radiometric calibration of solar absolute radiometer. The existing light source in the laboratory cannot meet the uniformity and stability at the same time. Therefore, it was proposed to obtain a highly uniform and stable surface light source through galvanometer scanning. Firstly, the two-dimensional laser scanning mode was established, and the optical system with fast scanning galvanometer and off-axis mirror as the core was designed. Secondly, according to the galvanometer scanning system, the driver was established, the control software was developed, and three scanning paths were designed. Finally, the appropriate scanning path was selected for experimental verification and its uniformity and stability was tested. The experimental results show that the non-uniformity of the scanning spot of the galvanometer is better than ±1%, the divergence angle is less than ±0.26°, and the stability of the light source is better than 0.02%. The feasibility of the laser galvanometer scanning light source as the irradiance calibration light source of the absolute radiometer is verified, which provides key technical support and experimental basis for achieving high-precision solar irradiance calibration.

Laser particle size analyzer has been widely used in the measurement of particle size distribution in various fields due to its advantages of fast speed and non-contact. However, the relationship between the angular distribution of scattered light and particle size is complex. In order to obtain the consistent relative accuracy under different particle sizes, the measurement range of particle size distribution will be small and cannot meet the requirements of wide distribution particle size measurement. According to the Fresnel principle of Mie scattering approximation, it was proposed to use the catadioptric optical path. The particle scattering signal was divided into two beams by the splitter, and the transmitted as well as reflected scattering signals were collected by two sets of compound lenses and two photodetectors, respectively. The combined signal was inversed to obtain the particle size distribution, thereby improving the range of particle size measurement. Two kinds of standard particles and their mixtures were used in the experiment. The results show that the relative error of the volume median diameter D50 measured by a single standard particle is not greater than 7.9%, and the correct peak distribution can also be obtained for mixed particles.

Particle velocity is an important parameter for analyzing the propagation law of stress waves in a solid medium. Combining the laser Doppler effect and an all-fiber interferometric velocimetry system, a measurement method of stress wave particle velocity in solid medium based on fiber coated probe was proposed. The optical fiber coated probe was embedded in the polymethyl methacrylate (PMMA) at the same radius from the burst center, and the miniature explosive ball with 0.125 g TNT equivalent was used as the explosion source to fill in the center cavity and generate stress wave. Based on the time-frequency analysis method of the short-time Fourier transform, the velocity of the optical fiber end surface could be calculated from the collected signal, and then the medium particle velocity could be deduced. The experimental results show that the velocity of the data measured by different fiber coated probes is 22.648 m/s and 23.505 m/s, respectively. The relative difference between the resulting particle velocity and the data obtained by the traditional circular electromagnetic particle speedometer method is less than 5.00%, which indicates the feasibility of the proposed method.

A method for identifying and measuring the hidden defects of optical cables based on Brillouin optic time domain reflectometer (BOTDR) distributed detection technology and demodulated signals was proposed to identify and measure the hidden defects of optical cables and ensure the safe operation of multi-strand carbon fiber optic cables used in projects such as capacity expansion and large span. In the production stage, the optical fiber was embedded in the composite core of multi-strand carbon fiber cable as a sensor. According to the principle of BOTDR technology and optic time domain reflectometer (OTDR) technology, a distributed sensing system for the detection of the hidden defects of carbon fiber cable was constructed. The hidden defects and position distribution of carbon fiber cable were analyzed in a multi-dimensional way by using the high-precision sensing of the optical fiber on temperature, stress and propagation loss, which realized hidden defect identification of optical cable. The Morlet wavelet was used to demodulate Brillouin scattering signal, extract envelope information and remove signal noise. Levenberg-Marquardt algorithm was used to fit Brillouin scattering spectrum data, accurately estimate the optimal Brillouin frequency shift parameter, and improve the overall defect recognition accuracy. The experimental results show that this method can accurately characterize the hidden defects of optical cable, realize the fast and effective identification of hidden defects of multi-strand carbon fiber cable, and promote the better application of carbon fiber cable in capacity expansion, large span and other projects.