An intravascular optical coherence tomographic (IV-OCT) system is developed, in which the diameter at the tip of the catheter probe is only 1 mm. In order to ensure the central axes of the Grin lens and the right angle prism attached to the micro motor shaft are aligned, a size-matching plastic sheath is designed to wrap the Grin lens. Then the Grin lens with the plastic sheath and the right angle prism are installed into a polytetrafluoroethylene (PTFE) sheath to assemble the probe, which has 1 mm diameter at the tip. A hardware filter is applied to removing the direct current component and the harmonic component in the k-clock signal generated by the swept laser source, and the resolution of the system is improved. The filtered k-clock signal is used as a trigger signal to resample the interference signal in the uniform k-space. Windowing, Fast Fourier Transform (FFT), logarithm, and background removal are applied to the resampled signal to get the A-scan data. At last, the coordinate transformation is applied to the A-scan data to reconstruct the circular display sample image. The axial resolution and the lateral resolution are 11.8 μm and 24 μm, respectively in actual measurement. The system imaging frame rate is 30 frame/s. With the proposed IV-OCT system, images of tubular white tape sample, shallot leaf, lotus root, and in vitro duck blood vessels are obtained.

Aiming at the capacity limit of single-mode fiber, a 4×100 Gbit/s dual polarization quadrature phase-shift keying (DP-QPSK) interleaved long-distance quasi-single-mode bi-directional transmission is achieved when we use a few-mode fiber, and the transmission distance is over 1700 km. We also study the effects of multi-path interference (MPI) and double-Rayleigh scattering (DRS) on the system. The loss caused by MPI can be effectively compensated by the constant modulus algorithm (CMA) with 301 taps, and the performance of the system increases about 1 dB. Meanwhile, a waveform shaper is used to suppress signal degradation resulting from DRS. The results show that the transmission distance can be extended from 1400 km to 1700 km by these methods.

Coherent combining of fiber lasers is an effective approach to pass through the power limitation of single channel fiber laser and achieve high output power, and is also an important foundation for the traditional high power laser system developing towards laser phased array of high power fiber laser system. The system configuration of high power fiber laser coherent combining is introduced, and the main study object and the key technique are pointed out. The research status and development tendency of fiber laser coherent combining are presented. The main technique challenges for fiber laser coherent combining are analyzed.

Constrained by such factors as thermal damages and nonlinear effects, the single-mode output power of one single fiber lasers is limited. Spectral beam combining technology to integrate different laser outputs based on some means is believed to be an inevitable choice to achieve high power laser output. The development process and current status of spectral beam combing technology are introduced and the basic principle, key factors, advantages and disadvantages of several common kinds of spectral beam combing systems are presented. The latest progresses on narrow linewidth high power fiber laser and its spectral beam combining in Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, are also introduced, and the development prospects of spectral beam combining technology of high power fiber lasers are prospected.

Thulium-doped fiber lasers is widely applied in laser medical, eye-safe radar, nonmetal material processing, and electro-optical countermeasure system, which has an irreplaceable role compared to other wavelength fiber lasers. Here, the progress on high power ultra-short pulse thulium-doped fiber lasers of our group is mainly introduced, which includes the controlling of pulse width and spectrum shape of mode-locked thulium-doped fiber oscillator by using fiber Bragg grating. High repetition and high stability picosecond pulsed laser output at waveband of 2 μm is realized. Using the picosecond fiber oscillator as a seed source, we demonstrate a picosecond thulium-doped fiber amplifier with hectowatt all-fiber structure. The maximum average output power of final step power amplifier is 120.4 W, corresponding to the pulse duration of 16 ps. In addition, we also design and build an all-fiber, all-polarization-maintaining structure, and picosecond thulium-doped fiber amplifier system. The linear polarization picosecond pulsed laser output with average power of 240 W is realized. The polarization extinction ratio is greater than 15 dB, and the pulse duration is 45 ps.

The research status of traditional dynamic thermally-induced modal instability in fiber lasers is introduced, and the theoretical predictions of quasi-static modal instability since 2016 are also introduced. Then, the status of non-thermally-induced modal instability, including electrostriction effects and nonlinear effects in modal instability, is introduced. Finally, the power scaling limit of modal instability on high power fiber amplifier is summarized and analyzed.

In recent years, high power Raman fiber lasers have progressed significantly. The output wavelength covers from visible to mid-infrared. The maximum output power can be higher than a thousand watts. For power scaling, the Raman fiber lasers have utilized schemes of laser oscillator, master oscillator power amplifier, and pump and signal integrated amplifier. Motivated by important scientific applications, high power narrow linewidth Raman fiber amplifiers have advanced greatly as well. Moreover, new Raman fiber laser schemes are emerging, for example, cladding pumping, laser diode direct pumping, and Raman fiber laser beam combining etc. There is a great potential for Raman fiber laser technology in the future.

The experimental research progress of 2.0 μm-band Tm-doped continuous wave (CW) single-frequency fiber laser is introduced. With a focus on the key technologies in realizing the single-frequency fiber lasers, the present research status and development directions of the Tm-doped single-frequency lasers with different cavities are summarized. Based on master-oscillator power amplifier (MOPA) of laser seeds, the power amplification of single-frequency lasers is realized and the research progress of high-power Tm-doped single-frequency fiber MOPA lasers at home and abroad is summarized. In addition, some research works in our research group related to the fabrication of highly Tm-doped germanate glass fibers, the construction of 2.0 μm-band single frequency laser oscillators, and the power amplification of single-frequency lasers are presented.

In view of the principal limiting factors of semiconductor pump fiber laser for power raising, the advantages and disadvantages of tandem pump in the aspect of high power output are analyzed. The development of tandem pump scheme is briefly reviewed. The key techniques, research status and the latest progress of tandem pump Yb-doped fiber laser are presented in detail. In view of the tandem pump absorption and nonlinear effect, the next step research scheme and technical route are analyzed. The new application of tandem pump technology is briefly introduced.

High power fiber laser is the current focus of laser research and development. Pumped coupling power will directly determine output power of fiber laser. This paper investigates current research of fiber laser pumping coupling technology. We start with the two basic structures, end pumping coupling and side pumping coupling, and discuss in detail the technical programs and features of various fiber laser pumping couplers. Through comprehensive comparison, existing problems and future research directions are put forward.

Since the emergence of fiber lasers, the output power of fiber laser has been increased from milliwatt level to myriawatt level along with the rapid development of semiconductor material and fiber fabrication technology. However, with the increasing of output power of fiber laser, many phenomenons which are not shown at a low output power have appeared gradually, such as fiber thermal damage, nonlinearity effect and mode instability, which limit the application of fiber laser. Therefore, the demand for the quality of Yb-doped fiber is higher and higher. The fabrication technologies, doping components and structural design of high power Yb-doped fibers are discussed. The status and development tendency of thermal stability, power stability and mode stability of high power Yb-doped fiber are analyzed.

In order to solve the problem of large vertical divergence angle of semiconductor laser and poor beam quality, the photonic crystal is introduced to realize mode extension and separation, which improves the properties of laser diode and realizes high brightness and high beam quality laser output. Photonic crystal working principle is theoretically analyzed and simulated as well, and some photonic crystal semiconductor lasers with different properties are introduced. Based on the photonic crystal semiconductor laser, some structures are integrated to get high power, low vertical divergence angle and high brightness output beam. Experimental results prove that the band gap effects of the photonic crystal improve the quality and brightness of laser beam, and break through the limitation of normal semiconductor, which help the semiconductor laser effectively used in the fiber laser pump and laser processing, and establish the foundations for the direct application of semiconductor laser.

An all polarization-maintaining fiber femtosecond laser amplification system, using Yb-doped large-mode-area double-cladding fiber (LMA-DCF) as gain medium and adopting the chirped pulse amplification technology, was constructed. Benefiting from the all polarization-maintaining fiber construction, the amplification system showed high -integration and long-term stability. Furthermore, the quality and duration of the pulses after compression were optimized by controlling the length of the stretcher fiber, which made it possible to compensate the third-order- dispersion accumulated in the system by introducing nonlinear chirp in the amplifier. Meanwhile, we analyzed the impact on amplified pulses caused by periodic spectral modulation. Through optimizing artisan craftsmanship of polarization-maintaining fiber (2+1)×1 combiner, the issue of periodic spectral modulation was solved. Finally, the high-quality ultrashort laser pulses with an average power of 9 W and a pulse duration of 260 fs at the repetition rate of 111 MHz after compression, corresponding to single pulse energy up to 81 nJ and pulse polarization ratio of 13 dB, were generated.

Two types of hollow-core anti-resonant fiber (HC-ARF) with properties of low-loss, single mode, and near-infrared broadband transmission are fabricated, which are used for the efficient propagation of high power picosecond ultrashort pulse. The propagation of a ultrashort pulse with average power of 74 W, single pulse energy of 185 μJ, and peak power of 10.8 MW through a node-less HC-ARF is realized, and the spectral and temporal characteristics of output lasers have no obvious changes.

Thermal effect of the gain fiber is one of the main factors limiting the power improvement of high power fiber amplifiers. Using traditional temperature measurement methods, we can only obtain the surface temperature of fiber while the core temperature cannot be detected. In this paper, the temperature of the gain fiber core in an all-fiber amplifier is measured by optical frequency domain reflectometry (OFDR). Firstly, the temperature measurement results by OFDR are calibrated and the measuring accuracy of the gain fiber temperature is verified when the amplifier is in operation. Then, the temperature distribution of the gain fiber core in the all-fiber amplifier is measured when the output power is 6 W and the results agree with the current theoretical results. The temperature measurement method proposed can provide a reference for the temperature monitoring in high power fiber lasers in the future.

A tunable, narrow linewidth mid-infrared Cr∶ZnSe laser consisting of X-type cavity pumped by a home-made 1908 nm thulium-doped fiber laser is built. The maximum continuous wave output power is 1.6 W at central wavelength of 2420 nm with respect to incident pump power of 6.8 W, and the corresponding slope efficiency is 26.7%. Littrow configuration is adopted to realize the continuous tuning between 2284 nm to 2716 nm with a tuning range of 432 nm, the full width at half maximum (FWHM) is 0.13 nm. The output power of more than 500 mW is achieved in the range from 2350 nm to 2510 nm by using 22% output ratio of coupling output mirror with respect to incident pump power of 5 W. Littman configuration is adopted to realize the continuous tuning from 2305 nm to 2658 nm with a tuning range of 353 nm, and the FWHM is less than 0.05 nm. The output power of more than 300 mW is achieved in the range from 2350 nm to 2520 nm.

The random fiber laser is established with Yb-doped fiber as gain and single-mode fiber as the random distributed feedback, The maximum output power is 5.1 W and the full width at half maximum (FWHM) is 0.34 nm. The high power laser is realized by employing a main oscillator power amplifier configuration, seeded by the random fiber laser above, whose output power reaches 1102 W with an optical-to-optical efficiency of 78.5%. During the whole amplification process, the laser FWHM is well maintained but the root mean square (RMS) linewidth broadens to 1.24 nm. To suppress the spectral width of the random fiber seed laser, a narrowband fiber Bragg grating is used to filtering the random fiber seed laser, and then to enlarge. The maximum output power reaches 1093 W with FWHM of 0.61 nm and RMS linewidth of 0.92 nm. No amplified spontaneous emission and stimulated Brillouin scattering is observed in the whole amplification process.

A main oscillation fiber amplification (MOFA) system is proposed, and the increasing of pulse signal peak power and the broadening of repetition frequency modulation range of the proposed system are studied experimentally. The electrical pulse modulation and the acousto-optic modulation are employed to modulate optical pulse sequence. Amplified spontaneous emission effect and self-excited oscillation of semiconductor seed source, which are generated in the process of small signal amplifying, are suppressed. The limitation of seed source to the repetition frequency of the proposed system is eliminated, and the repetition frequency can be tuned continuously in the range of 1~200 kHz. In the process of optical pulse modulation with the acousto-optic modulator, the acousto-optic Q-switch can effectively suppress the spectral broadening caused by self-phase modulation. The rod-like photonic crystal fiber with large mode field is used as the gain medium to increase the peak power of optical pulse and the beam quality of output light, and the nonlinear effect caused by high peak power is suppressed. Ultimately, the stable optical pulse output with the pulse width of 1.12 ns and the peak power of 2.01 MW is achieved, the average power is more than 56 W, and the beam quality of the output light is close to the diffraction limit (M2x=1.49, M2y=1.54).

A high repetition rate and wide spectrum picosecond pulsed all-fiber Yb-doped laser is experimentally studied. The laser with all-fiber structure uses the master oscillator power amplifier technology. The broadening of the spectrum is caused by the nonlinear effect. Passively mode-locked technique of oscillator is realized by semiconductor saturable absorber mirror. The laser operates at 1031.3 nm with 3 dB spectral width of 1.51 nm. Its pulse width is 3.1 ps and the repetition frequency is 21.3269 MHz. An acoustic-optic modulator is applied to reduce the repetition rate to 2.1 MHz. The down-converted signal light passes through a three-stage Yb-doped fiber amplifier, and the main power amplification is realized in 30 μm/250 μm large mode field double-clad Yb-doped fiber. In the end, the pulse laser output is obtained with average output power of 20 W, pulse width of 20.8 ps, spectrum range from 640~1700 nm, single pulse energy of 9.5 μJ, and corresponding peak power of 0.46 MW.

A supercontinuum spectrum source with all-fiber structure is proposed to obtain high power supercontinuum spectrum with the pulse width of 12 ps and the center wavelength of 1064 nm. The source is used as the pump source, and the output power is amplified to 189 W in a Yb-doped fiber with a length of 10 mm. A narrow-band filter and a interstage isolator are used to amplify the pulse signal. The amplified pulse signal is injected into a photonic crystal fiber with a length of 0.5 m, and supercontinuum spectrum is generated with the spectrum range of 460-1700 nm and the output power of 102.8 W. For the existence of quantum defect and spectrum transmission loss, the light-light slope efficiency of the supercontinuum spectrum decreases from 90% to 30% when the pump power increases from 1.5 W to 189 W.

Based on the diffraction theory, an analytical model for the beam quality after the beam passing through volume Bragg grating (VBG) is built. Considering the parameters of input beam and VBGs, the optical field distribution of an arbitrary beam passing through VBG after diffraction is analyzed，and the beam waist in second moment definition and the beam quality parameter of M2 can be calculated. The effect of space period of VBGs, input beam size, spectral bandwidth and initial input beam quality M2 on the diffraction beam quality are analyzed. The results show that the diffracted beam quality is influenced by the VBGs and the input sub-beam commonly. The spatial dispersion becomes larger with the decrease of period of VBGs, and the output beam quality after diffraction deteriorates significantly with the increase of input sub-beam size, spectral bandwidth and the initial beam quality. The model gives the analytical formula for the beam quality after diffraction, which can be used to obtain the beam quality after diffraction for different input beams easily and quickly. At the maximum input power of kilowatt level, the contribution of the initial M2 and the spectral line width to the diffracted beam quality is measured, and it is compared with the corresponding theoretical analysis.

A continuous wave laser based on Nd-doped double-clad large-mode-area polarization-maintaining photonic crystal fiber (PCF) is realized at 1.06 μm. A linear-cavity is employed, and one end face of the fiber is polished to 0° providing 4% feedback serve as a cavity mirror and a free output coupler. The slope efficiency obtained is 18.8% with a threshold pump power of 2.4 W. The maximum output power is 4.94 W and the beam quality factor is 1.30, when the pump power is 30 W. The measured central wavelength is 1062.7 nm and the polarized crosstalk factor is -7.9 dB. Experimental results show that the Nd-doped double-clad large-mode-area PCF has well single mode propagation and polarization-maintaining characteristics, and is beneficial to achieve high power laser output with high beam quality.

Based on the theoretical model of divided-pulse amplification, the effect of all parameters in the divided-pulse fiber amplification system with five-step birefringent crystals group cascading on the property of amplified picosecond pulse is numerically investigated. The results indicate that, the difference of nonlinear phase shift accumulation between divided pulses leads to the reduction of the combining efficiency. The optimal combining efficiency is obtained when the time separation between two adjacent divided pulses is larger than 2.5 times of pulse width. Under the condition of nonlinear coefficient and maximum gain of fiber amplifiers being fixed, the selection ranges of gain fiber length and seed pulse peak power are presented in order to obtain a transform-limited amplified pulse with an output peak power of 10 MW. Under the condition of fiber length being fixed, it is suggested to choose a seed pulse with low peak power and a fiber with large gain in order to suppress the impact of rotation angle bias and transmissivity on the property of the amplified pulse.

By taking glycerol, ethanol, DT-3 and glycidyl azide polymer (GAP) all with carbon doped as typical liquid working substances, and under different incident laser energies, the ablation product and the splash of working substance induced by this ablation product are investigated, and the thrust evolution law is obtained. The results indicate that, the high speed splash process at the beginning of splash is the main stage of thrust formation. The subsequent splash process makes the mass flow rate increase, but makes less contribution to impulse coupling, which inversely degrades the overall thrust performance.

With stainless steel as research target, the molten pool shape and weld morphology during the process of single/dual beam laser welding are comparatively studied, and the relationship between both of them is established. The results indicate that the evolution processes of molten pool shape are different for single and dual beam laser welding and the molten pool size and weld morphology of dual beam laser welding are influenced by the spot spacing. As for the process of dual beam laser welding, the interaction of two keyholes is perpendicular to the connection line of the two laser beams which induces an enhanced melt flow, while as for that of single beam laser welding, the melts flow radiantly and evenly from the keyhole. The difference between melt flow patterns is the key to induce the difference in weld morphology under single/dual beam laser welding.

To solve the problem of cladding layer cracking of single-crystal Ni-based superalloy DD4 during the laser direct forming process, DD4 parts are fabricated on the directionally solidified DZ125L substrate by different processes. The kinds and the characteristics of cracks in cladding layers of DD4 parts and the effects of basic process parameters on the crack rate of DD4 parts are investigated. The laser direct forming process which is assisted by induction heating is proposed to eliminate the cladding layer cracks of DD4 parts. The results indicate that the cracks in the cladding layers of DD4 parts mainly include solidification crack and liquidation crack, and most of the cracks are the latter one. Reducing laser power, increasing scanning speed, and choosing appropriate overlapping space and elevation can decrease the crack rate of cladding layers of DD4 parts. The crack rate of the DD4 parts can be reduced to 0.230 mm·mm-2 by optimizing process parameters. The crack rate of DD4 parts decreases with the elevation of the induction heating temperature, and the crack rate is 0.017 mm·mm-2 when the induction heating temperature is 1200 ℃.

A finite element model is developed by Abaqus software to investigate the effect of laser power density on the residual principal stress distribution on the surface around the hole. Laser peening (LP) experiments under different power densities are conducted, and the residual stresses of feature points around holes are measured along three directions by X-ray diffraction stress analyzer, and the corresponding principal stress of each test point is calculated. The results indicate that the residual stress is anisotropic on the surface around the hole after LP. The minimum residual principal stress is an effective indicator to characterize the LP effect, which decreases with the increment of laser power density. In addition, a typical residual compressive stress ring occurs around the hole after LP, where the minimum residual principal stress value is obviously less than that in other peening zones. When the laser power density exceeds a certain threshold value, this ring has a tendency to be away from the hole wall.

When double thin-walled parts with connecting ribs are formed, the bump defects are easy to occur in the connection between ribs and thin walls. The bump defects will be more obvious with the accumulation if the control is improper, and it is easy to lead to accumulation failures. We propose a lateral lap process by equal volume method, which can obtain intersecting surfaces of ribs and walls with flat cladding morphology by controlling the starting and stopping offset in the nozzle scan path to control the lateral overlap. We also propose a power control technology based on the height detection technique to ensure that the morphology of the cladding layer in the accumulation process remains stable until the laser direct forming of the double thin-walled parts with connecting ribs is completed. The gap of the formed double thin-walled parts with connecting ribs is about 2.6 mm, and the size deviation is less than 5%. The microstructure is dense and uniform, without obvious holes, cracks and other defects, and the microhardness is 650-770 HV.

With a 6 kW fiber laser, Fe17Mn5Si10Cr5Ni shape memory alloy coating is obtained by powder presetting and laser cladding on the 304 stainless steel surface. Microstructure and phase composition of the coating is studied by the scanning electron microscope, X-ray diffractometer and reciprocating friction tester. Wear resistance and contact fatigue strength of the coating and substrate are analyzed, and residual stress is measured with the orifice method. Results show that the coating is composed of planar crystal, cellular crystal and columnar crystal from the bottom of molten pool to surface; wear mechanism of the coating is abrasive wear, while that of the substrate is adhesive wear, and wear loss of the coating is nearly a third of that of the substrate; and wear resistance and contact fatigue strength of the coating are better than those of the substrate. Besides, residual stress of the coating is smaller than that of the substrate. The key reason for good mechanical properties of the coating is the stress induced γ→ε phase transformation.

Hastelloy-X specimens are processed and post-processed by the selective laser melting technique, the macro-/micro-structural characteristics of as-deposited, heat-treated and heat-plus-heat-isostatic-pressure-treated specimens are analyzed, and the tensile property at room/high temperature is tested, respectively. The results indicate that the micro-cracks appear at both horizontal and vertical directions of as-deposited specimens, but no precipitates occur. After heat treatments, the transverse microstructures become equiaxed grains, while the longitudinal microstructures are interleaved fine or coarse columnar grains, and within these grains, there exist two kinds of precipitates with quite different morphologies. After the hot-isostatic-pressure treatment, the grains grow up significantly, the distribution of the precipitates are homogenously distributed, and the tensile property at room temperature reaches the forging standard. The transverse tensile property is characterized by high strength and low plasticity compared to the longitudinal one.

Aiming at the uneven heat dissipation induced by the uneven water channel flow of microchannel-heat sink for original curamik, the numerical simulation based on FLUENT software is carried out. Optimization scheme is proposed from the aspect of internal structure and the heat sink material. The influence of the microchannel width, interval and microchannel ridge length on the chip surface temperature rise and pressure drop are obtained, when the heat sink height and the width of the import and export are invariant. Based on the optimized parameters, the pure Ni microchannel heat sink is manufactured by selective laser melting technology, and the chip is packaged and tested. The results show that thermal resistance of microchannel heat sink reaches 0.39 K/W, the pressure drop reaches 140 kPa, which can meet the need of 80 W semiconductor laser cooling requirements.

A novel concave cone surface-enhanced Raman scattering (SERS) fiber probe modified with the gold nanoparticles was demonstrated. We studied the preparation method of the concave cone fiber probes, analyzed the relationship between the morphology of the probe and the corrosion time, modified the gold nanoparticles onto the inner surface of the concave cone by the chemical self-assembly method, and tested the SERS detection performance of the prepared probe. Experimental results show that the concave cone fiber probe has a lower fiber Raman background, about 1/3 of the tapered probe prepared with the same fiber. With 633 nm excitation, Raman spectra of rhodamine 6G (R6G) aqueous solution of 100 nmol/L can be detected by the concave cone fiber probes modified with gold nanoparticles. Due to this concave cone structure, the metal nanoparticles are more difficult to fall off, and the probe has better damage resistance. Based on the above advantages, this type of concave cone fiber probe may have potential application value in the field of SERS remote detection.

Aiming at the surface aberration for large-aperture and ultra-thin crystal used on the final optical system of inertial confinement fusion facility, a new design of precision mounts with low-stress, named clip-like clamped support all four edges, is proposed. Based on the harmonic conversion model and crystal alignment, the third harmonic generation efficiency will fall and the alignment focal spot will widen because of the crystal surface aberration. So the two control aims must be achieved that the surface aberration should be less than 5 μm and the crystal clamping aberration should be less than machining aberration. Based on the mechanics model, the clip-like clamped support all four edges of precise assembly structure is designed. The precision machining control, finite element analysis, and experimental verification are carried out. Experimental results show that the total surface aberration is less than 5 μm by using the method of precision mounts with low-stress, this method meets the demand of crystal surface aberration control.

A new method of time-delay signature suppression in semiconductor lasers with optical feedback by utilizing Brillouin scattering effect in optical fibers is proposed. Experimental results show that, when a chaotic laser with an average power of 200 mW is injected into the single mode fiber with a length of 10 km and after the action of Brillouin backscattering in fibers, the peak value at time delay of 105 ns of the chaotic laser auto-correlation trace is reduced from 0.251 to 0.075 and the peak value at time delay of 105 ns of the chaotic laser mutual information trace is reduced from 0.087 to 0.008. On this basis, the influence of optical injection power on time-delay signature suppression effect is experimentally investigated and the results indicate that, when the average power of chaotic laser is within the range of 200-1500 mW, the time-delay signature can be effectively suppressed; when the number of measurements is within the range of 1-50, the time-delay signature suppression effect of chaotic laser by the fiber Brillouin scattering is relatively stable.

Based on time division multiplexing technology, a distributed weak fiber Bragg grating (FBG) array vibration detection system is proposed. By using a path-match interferometry method, the interference signals between two adjacent weak FBGs are detected, and relevant information of vibration such as phase, frequency and location is obtained, thus we can achieve high sensitivity distributed measurement. The 3×3 phase demodulation technique is employed to effectively eliminate signal fading in the interferometer. The feasibility of the system is verified by a weak FBG array with 660 gratings and 2.5 m distance between adjacent FBGs. Experimental results show that the interference fringe visibility is related to the wavelength difference between adjacent weak FBGs. Meanwhile, the proposed system can accurately distinguish weak signal with different frequencies (ranging from 20 Hz to 1000 Hz) and has better frequency response. The sensor system has higher sensitivity in contrast to standard seismic geophone.

In order to solve the problem of cross sensitivity of temperature and strain that generally exits in fiber-optic sensor, the strain and temperature sensing characteristics of LP01 and LP02 mode interference based on the inter-modal interference principle of the few mode fiber are studied, and phase sensitivity theory of inter-modal interference sensing is analyzed in detail. A temperature-insensitive strain-sensing few mode fiber with core diameter of 15.1 μm, core refractive index of 1.4512 at 1550 nm and cladding refractive index of 1.444 is designed based on the numerical simulation results of the few mode fiber. The experimental system is built to research the strain and temperature sensing characteristics of few mode fiber. The results show that the theoretical calculation can predict the experimental result well. The phase sensitivity of the strain in the few mode fiber is 0.0196 rad/μm in the range of 0~600 μm, which is insensitive to the temperature in the range of 30~330 ℃. It can effectively improve the problem of cross-sensitive of temperature and strain.

For the foundation of low altitude flight safety service, based on the detection database collected at the northeast of China form July to September in 2015 and using a Chinese new all-fiber coherent wind lidar combined with a double theodolite pibal and wind profile radar, the wind measurement precision and reliability of this new wind lidar are evaluated by comparing the correlation of laser radar, the wind profile radar and pibal wind detection under clear-air, cloudy, foggy and hazy, and rainy conditions. The results show that, under clear-air, cloudy, foggy and hazy conditions, the wind lidar is more accurate than the wind profiler lidar, and the standard deviation between horizontal wind and real value error is less than a half of wind profiler lidar. During the rainy day, the detection range of wind lidar is limited, and the maximum observed altitude decreases to only half of the wind profiler which shows a severe influence compared with wind profiler under percipitation. The wind lidar results are more stable than that of the wind profiler. The dataset correlation coefficient, which can reach up to 0.9, between the measured data detected by the wind lidar and the real value has a tiny change. Under clear-air, cloudy, and fog and hazy conditions, the wind accuracy of two lidar are both affected by the variation of wind speed, meanwhile, the change trends are different as the wind speed increases.