Map
WeChat
Email alert
RSS2024 Vol. 48, No. 3
Display Method:
[Abstract]
[PDF 255KB]
2024, 48(3): 295-302.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.001
[FullText HTML]
Abstract:
For the purpose of improving the performance of discrete modulation-continuous variable quantum key distribution (CVQKD) protocols, the amplitude phase-shift keying(APSK) modulation format method was used, proposed to use Gaussian mixed model (GMM) classification algorithm at the receiver side to identify quantum states to enhance the performance of the system, dividing the key transmission system into two stages: state learning and state prediction. In the state learning stage, the classifier based on GMM was trained for known classes of quantum states, learning the amplitude-phase distribution of different classes of quantum states. The state prediction stage then used the minimum Euclidean distance to calculate the posterior probability that the quantum state to be measured belongs to each class, thus determining the class of the quantum state to be measured, and generating the final key by parameter estimation, reverse coordination and secrecy enhancement. Numerical simulation results show that the 128-APSK discrete modulation continuous-variable quantum key distribution protocol can effectively generate secure keys under reverse coordination and collective attack, and the transmission distance can approach 60 km when the secure code rate is 10-6 bit/symbol. This study provides a reference for further improving the system performance of discrete modulation continuous-variable quantum key distribution protocols.
For the purpose of improving the performance of discrete modulation-continuous variable quantum key distribution (CVQKD) protocols, the amplitude phase-shift keying(APSK) modulation format method was used, proposed to use Gaussian mixed model (GMM) classification algorithm at the receiver side to identify quantum states to enhance the performance of the system, dividing the key transmission system into two stages: state learning and state prediction. In the state learning stage, the classifier based on GMM was trained for known classes of quantum states, learning the amplitude-phase distribution of different classes of quantum states. The state prediction stage then used the minimum Euclidean distance to calculate the posterior probability that the quantum state to be measured belongs to each class, thus determining the class of the quantum state to be measured, and generating the final key by parameter estimation, reverse coordination and secrecy enhancement. Numerical simulation results show that the 128-APSK discrete modulation continuous-variable quantum key distribution protocol can effectively generate secure keys under reverse coordination and collective attack, and the transmission distance can approach 60 km when the secure code rate is 10-6 bit/symbol. This study provides a reference for further improving the system performance of discrete modulation continuous-variable quantum key distribution protocols.
2024, 48(3): 303-311.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.002
Abstract:
In order to avoid the problem of difficult direct observation and analysis of skin characteristics and mechanisms in aircraft skin images after laser cleaning, a method combining k-means clustering algorithm based on Lab color space and edge detection based on Canny operator was adopted to jointly analyze macroscopic and microscopic images of the paint layer. Firstly, perform color space conversion on the cleaned image, converting the original RGB color space into Lab color space, and perform image segmentation using k-means clustering algorithm. Then use the Canny operator to perform edge detection and extract edge information from the electron microscopy images at the junction of each paint layer. Then, study the characteristics and mechanisms of the processed images separately. Finally, the image processing results were validated through thermal stress analysis. The results indicate that, when laser energy density is 6.37 J/cm2 and 1.91 J/cm2, the top coat and primer of the aircraft skin can be completely removed by the laser. This study provides a reference for laser automatic paint removal.
In order to avoid the problem of difficult direct observation and analysis of skin characteristics and mechanisms in aircraft skin images after laser cleaning, a method combining k-means clustering algorithm based on Lab color space and edge detection based on Canny operator was adopted to jointly analyze macroscopic and microscopic images of the paint layer. Firstly, perform color space conversion on the cleaned image, converting the original RGB color space into Lab color space, and perform image segmentation using k-means clustering algorithm. Then use the Canny operator to perform edge detection and extract edge information from the electron microscopy images at the junction of each paint layer. Then, study the characteristics and mechanisms of the processed images separately. Finally, the image processing results were validated through thermal stress analysis. The results indicate that, when laser energy density is 6.37 J/cm2 and 1.91 J/cm2, the top coat and primer of the aircraft skin can be completely removed by the laser. This study provides a reference for laser automatic paint removal.
2024, 48(3): 312-317.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.003
Abstract:
In order to screen potential diabetes patients in high-risk groups, a non-invasive screening method for diabetes in high-risk groups based on optical coherence tomography (OCT) is proposed, and the patients could be early treated and reduce complications. This method first divided the 3-D OCT image into many small regions. Due to the different correlations between blood glucose changes and scattering coefficients in different regions, the principal component analysis was used to transform the scattering coefficients of these blood glucose sensitive regions into a comprehensive indicator S. Then S was normalized at different times of 0 min~120 min to obtain the relative quantity Si(i=0, 1, 2, …, 120), with a relative quantity of S120 at 120 min. According to the experience value of normalized blood glucose of healthy subjects, moderate diabetes, and severe diabetes in 0 min~120 min, the judgment threshold 1 and threshold 2 is 0.47 and 0.78, respectively. If 0 < S120 < 0.47, 0.47 < S120 < 0.78, and 0.78 < S120 < 1, the subject is healthy, moderate diabetes and severe diabetes, respectively. This method is high practicability and has great significance for screening diabetes in high-risk groups with optical non-invasive methods.
In order to screen potential diabetes patients in high-risk groups, a non-invasive screening method for diabetes in high-risk groups based on optical coherence tomography (OCT) is proposed, and the patients could be early treated and reduce complications. This method first divided the 3-D OCT image into many small regions. Due to the different correlations between blood glucose changes and scattering coefficients in different regions, the principal component analysis was used to transform the scattering coefficients of these blood glucose sensitive regions into a comprehensive indicator S. Then S was normalized at different times of 0 min~120 min to obtain the relative quantity Si(i=0, 1, 2, …, 120), with a relative quantity of S120 at 120 min. According to the experience value of normalized blood glucose of healthy subjects, moderate diabetes, and severe diabetes in 0 min~120 min, the judgment threshold 1 and threshold 2 is 0.47 and 0.78, respectively. If 0 < S120 < 0.47, 0.47 < S120 < 0.78, and 0.78 < S120 < 1, the subject is healthy, moderate diabetes and severe diabetes, respectively. This method is high practicability and has great significance for screening diabetes in high-risk groups with optical non-invasive methods.
2024, 48(3): 318-326.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.004
Abstract:
To explore the wind field detection effect and low-level wind shear identification ability of light detection and ranging (LiDAR) during a wet downburst, a typical wet downburst storm process on 2021-05-18 was analyzed and investigated by using wind LiDAR and meteorological observation records from Xining Caojiapu International Airport, and using the method of combining multiple data with specific events for theoretical analysis and data validation. The results show that the downburst has complex wind field structures. Before the downburst arrived at the airport, the downdraft outflow exceeded 14 m/s, which is coupled with the ambient wind to form a headwind shear convergence line, and the strongest outflow region is located 1 km to 2 km behind the convergence line. When the downburst arrives over the runway, it creates a divergence wind field on the ground. The wind speed in the center of the divergence area is much lower than that at the edge of the outflow, the vertical airflow had drastic changes, and the downburst lasted for about 10 min. The LiDAR has a good identification of the vertical airflow distribution inside the cumulonimbus, the fine structure of the divergence wind field, and the formation and evolution of the convergence line. The interactive use of different detection patterns and data products of LiDAR is very conducive to monitoring and warning of low-level wind shear at the airport. This study provides a reference for the application of LiDAR in the prediction and research of downburst wind shear.
To explore the wind field detection effect and low-level wind shear identification ability of light detection and ranging (LiDAR) during a wet downburst, a typical wet downburst storm process on 2021-05-18 was analyzed and investigated by using wind LiDAR and meteorological observation records from Xining Caojiapu International Airport, and using the method of combining multiple data with specific events for theoretical analysis and data validation. The results show that the downburst has complex wind field structures. Before the downburst arrived at the airport, the downdraft outflow exceeded 14 m/s, which is coupled with the ambient wind to form a headwind shear convergence line, and the strongest outflow region is located 1 km to 2 km behind the convergence line. When the downburst arrives over the runway, it creates a divergence wind field on the ground. The wind speed in the center of the divergence area is much lower than that at the edge of the outflow, the vertical airflow had drastic changes, and the downburst lasted for about 10 min. The LiDAR has a good identification of the vertical airflow distribution inside the cumulonimbus, the fine structure of the divergence wind field, and the formation and evolution of the convergence line. The interactive use of different detection patterns and data products of LiDAR is very conducive to monitoring and warning of low-level wind shear at the airport. This study provides a reference for the application of LiDAR in the prediction and research of downburst wind shear.
2024, 48(3): 327-333.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.005
Abstract:
In order to reduce the laser dressing error of concave-faced diamond wheels, the error models of laser masking effect and laser bevel dispersion effect were established, the error sources of laser tangential dressing of concave-faced diamond wheels were analyzed, and the deflection laser dressing method of concave-faced forming wheels was proposed. Through theoretical analysis and experimental verification, the influence of dressing parameters on the contour accuracy and circular radius of concave-faced wheels was explored. The results show that in the deflection angle range of 1°~1.5°, the beveled edge profile dressing accuracy is 8 μm; under the suitable deflection angle, the circular arc radius error is less than 10 μm; when dressing the concave arc radius of 0.2 mm and 0.5 mm, the compensation radius is 0.03 mm and 0.06 mm respectively, the concave arc radius error is reduced by 0.02 mm and 0.03 mm respectively. The deflection laser dressing method effectively improves dressing accuracy, reduces the dressing error, and provides a new idea for the dressing of concave diamonds.
In order to reduce the laser dressing error of concave-faced diamond wheels, the error models of laser masking effect and laser bevel dispersion effect were established, the error sources of laser tangential dressing of concave-faced diamond wheels were analyzed, and the deflection laser dressing method of concave-faced forming wheels was proposed. Through theoretical analysis and experimental verification, the influence of dressing parameters on the contour accuracy and circular radius of concave-faced wheels was explored. The results show that in the deflection angle range of 1°~1.5°, the beveled edge profile dressing accuracy is 8 μm; under the suitable deflection angle, the circular arc radius error is less than 10 μm; when dressing the concave arc radius of 0.2 mm and 0.5 mm, the compensation radius is 0.03 mm and 0.06 mm respectively, the concave arc radius error is reduced by 0.02 mm and 0.03 mm respectively. The deflection laser dressing method effectively improves dressing accuracy, reduces the dressing error, and provides a new idea for the dressing of concave diamonds.
2024, 48(3): 334-339.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.006
Abstract:
In order to effectively counteract high-repetition-frequency jamming, a solution based on the analysis of the destructive mechanism of high-repetition-frequency jamming on laser guided radar signal processing was developed. The solution involves the use of randomly modulated laser waveforms and incorporates prior knowledge of the distance and velocity characteristics of the laser target indicator, target, and detector. A high-repetition-frequency jamming signal identification and discrimination method was designed based on adaptive overlapping wave gates, effectively controlling the impact of repetition-frequency jamming on the target acquisition process. It provided an end-to-end solution for combating high-frequency jamming, from waveform design to signal processing to effective target capture. The proposed solution was theoretically analyzed and experimentally verified. The results show that in different interference scenarios, the effective capture time is consistently below 0.2 s. The effective capture time is in the sub-second range, comparable to the typical capture time under no jamming background. This technology can effectively counteract fixed/random frequency jamming ranging from tens of Hertz to 250 kHz, enhancing the capability of guided weapons to counter high-repetition-frequency jamming.
In order to effectively counteract high-repetition-frequency jamming, a solution based on the analysis of the destructive mechanism of high-repetition-frequency jamming on laser guided radar signal processing was developed. The solution involves the use of randomly modulated laser waveforms and incorporates prior knowledge of the distance and velocity characteristics of the laser target indicator, target, and detector. A high-repetition-frequency jamming signal identification and discrimination method was designed based on adaptive overlapping wave gates, effectively controlling the impact of repetition-frequency jamming on the target acquisition process. It provided an end-to-end solution for combating high-frequency jamming, from waveform design to signal processing to effective target capture. The proposed solution was theoretically analyzed and experimentally verified. The results show that in different interference scenarios, the effective capture time is consistently below 0.2 s. The effective capture time is in the sub-second range, comparable to the typical capture time under no jamming background. This technology can effectively counteract fixed/random frequency jamming ranging from tens of Hertz to 250 kHz, enhancing the capability of guided weapons to counter high-repetition-frequency jamming.
2024, 48(3): 340-344.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.007
Abstract:
To evaluate the effect of ion source assistance on different oxide films, TiO2, Ta2O5, ZrO2 and SiO2 were deposited with different radio frequency(RF) sources. The index and thickness were measured by a spectroscopic ellipsometer. The Young's module and the hardness of these films were compared. Experiments show that the value of Young's module and hardness are increased, 2 GPa for TiO2, 4 GPa for Ta2O5, 2 GPa for ZrO2, in accordance with the film index. The nano-indention method is effective in evaluating the ability of ion sources.
To evaluate the effect of ion source assistance on different oxide films, TiO2, Ta2O5, ZrO2 and SiO2 were deposited with different radio frequency(RF) sources. The index and thickness were measured by a spectroscopic ellipsometer. The Young's module and the hardness of these films were compared. Experiments show that the value of Young's module and hardness are increased, 2 GPa for TiO2, 4 GPa for Ta2O5, 2 GPa for ZrO2, in accordance with the film index. The nano-indention method is effective in evaluating the ability of ion sources.
2024, 48(3): 345-351.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.008
Abstract:
In order to realize the high precision detection of the volume fraction of methane, the photonic spin Hall effect phenomenon with high-quality factor is excited by a multilayer structure. In an asymmetric arrangement, a methane-sensitive film was introduced into the structure, and the volume fraction of methane could be detected by the change in the refractive index of the sensitive film. The effects of gas porosity, period number, metal thickness, and sensitive film thickness on the photon spin Hall effect were studied, and the transfer matrix method was used for numerical analysis. The results show that the sensor can detect methane gas with volume fraction of 0~3% (refractive index change is 1.4364~1.4478) with a sensitivity of 29.6°. The highest figure of merit and the lowest detection limit are 395 and 0.00012, respectively. The sensor has a simple structure and strong detection ability, which provides a new idea for the research of optical sensors.
In order to realize the high precision detection of the volume fraction of methane, the photonic spin Hall effect phenomenon with high-quality factor is excited by a multilayer structure. In an asymmetric arrangement, a methane-sensitive film was introduced into the structure, and the volume fraction of methane could be detected by the change in the refractive index of the sensitive film. The effects of gas porosity, period number, metal thickness, and sensitive film thickness on the photon spin Hall effect were studied, and the transfer matrix method was used for numerical analysis. The results show that the sensor can detect methane gas with volume fraction of 0~3% (refractive index change is 1.4364~1.4478) with a sensitivity of 29.6°. The highest figure of merit and the lowest detection limit are 395 and 0.00012, respectively. The sensor has a simple structure and strong detection ability, which provides a new idea for the research of optical sensors.
2024, 48(3): 352-356.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.009
Abstract:
To achieve measurement of voltage direction for optical voltage transformers, theoretical analysis and experimental verification were conducted by the Pockels effect and theory of polarized light, and a measuring method of voltage direction was presented based on the Pockels effect. The relation among different direction voltage, the state of light polarization and light propagating direction were obtained. The results show that, when the value of applied voltage is 100 V, observing toward light, the long and short axis of the transmitted light are located at the spatial angles of 20° and 110°, respectively, and the polarization direction of light passing through a λ/4 wave-plate is located in the second and fourth quadrants of the coordinates, which indicates that the transmitted light is right-lateral elliptically polarized light, the direction of voltage is along the light propagating direction. In addition, the spatial angles are 130° and 40°, and the polarization direction with a λ/4 wave-plate is located in the first and third quadrants, which indicates that the transmitted light is left-lateral elliptically polarized light, the direction of voltage is along the opposite direction of light propagation. The experimental results agree well with the theory, which can guide the design of optical voltage transformers that can measure the magnitude and direction of voltages.
To achieve measurement of voltage direction for optical voltage transformers, theoretical analysis and experimental verification were conducted by the Pockels effect and theory of polarized light, and a measuring method of voltage direction was presented based on the Pockels effect. The relation among different direction voltage, the state of light polarization and light propagating direction were obtained. The results show that, when the value of applied voltage is 100 V, observing toward light, the long and short axis of the transmitted light are located at the spatial angles of 20° and 110°, respectively, and the polarization direction of light passing through a λ/4 wave-plate is located in the second and fourth quadrants of the coordinates, which indicates that the transmitted light is right-lateral elliptically polarized light, the direction of voltage is along the light propagating direction. In addition, the spatial angles are 130° and 40°, and the polarization direction with a λ/4 wave-plate is located in the first and third quadrants, which indicates that the transmitted light is left-lateral elliptically polarized light, the direction of voltage is along the opposite direction of light propagation. The experimental results agree well with the theory, which can guide the design of optical voltage transformers that can measure the magnitude and direction of voltages.
2024, 48(3): 357-364.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.010
Abstract:
The austenitic stainless steel additive manufactured by laser powder bed fusion (LPBF) has a good application prospect because of its good printability and mechanical properties, but there are still some problems that limit its industrial application. Adding nano-reinforcing phases is one of the effective strategies for regulating the properties of LPBF austenitic stainless steel. The review summarized the research progress of nanoparticles-reinforced LPBF austenitic stainless steel. We focused on discussing the effect of nanoparticles on the densification, microstructure, and mechanical properties. The strengthening mechanism was analyzed. Due to the addition of nanoparticles, the porosity of the composite material increased, and the density decreased. Cellular structure grains were finer with low anisotropy. Austenitic stainless steel added reinforcement phase not only significantly improved strength but also maintains good plasticity, mainly attributed to the comprehensive effects of grain refinement, dislocation strengthening, Orowan strengthening, and load transfer strengthening. Finally, the research directions of nanoparticles reinforced austenitic stainless steel by LPBF that need to be further explored in the future were prospected.
The austenitic stainless steel additive manufactured by laser powder bed fusion (LPBF) has a good application prospect because of its good printability and mechanical properties, but there are still some problems that limit its industrial application. Adding nano-reinforcing phases is one of the effective strategies for regulating the properties of LPBF austenitic stainless steel. The review summarized the research progress of nanoparticles-reinforced LPBF austenitic stainless steel. We focused on discussing the effect of nanoparticles on the densification, microstructure, and mechanical properties. The strengthening mechanism was analyzed. Due to the addition of nanoparticles, the porosity of the composite material increased, and the density decreased. Cellular structure grains were finer with low anisotropy. Austenitic stainless steel added reinforcement phase not only significantly improved strength but also maintains good plasticity, mainly attributed to the comprehensive effects of grain refinement, dislocation strengthening, Orowan strengthening, and load transfer strengthening. Finally, the research directions of nanoparticles reinforced austenitic stainless steel by LPBF that need to be further explored in the future were prospected.
2024, 48(3): 365-372.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.011
Abstract:
To achieve rapid and non-destructive detection of strawberry hardness, strawberry hyperspectral data, and hardness information were collected for five consecutive days, and a hardness prediction method based on high spectral multi-index threshold layer-by-layer segmentation was proposed. Firstly, the spectral reflectance differences of different components (pulp, moldy pulp, strawberry seeds, and sepals) were analyzed, and the characteristic bands were identified. Subsequently, new normalized feature indices were constructed based on the characteristic bands, which were selected based on spectral reflectance differences, and the segmentation thresholds were determined. The layer-by-layer segmentation method was used to eliminate the interference of irrelevant parts. Three methods(successive projections algorithm, principal component analysis, and quadratic combination dimensionality reduction) were used to reduce the spectral information redundancy and extract features. The regression models were established for the original spectral data and the reduced feature data by random forest and partial least squares regression, respectively. The best prediction model was determined to fit the hardness of the strawberry pulp. The hardness distribution image was obtained for the intuitive display of the strawberry hardness prediction result. The result shows that the partial least squares model based on quadratic dimensionality reduction yielded the best performance, with correlation coefficients of 0.9101 and 0.9099 for the test set and prediction set, respectively, and with a root-mean-square error of 0.1344 for the test set. This study provides a reference for non-destructive detection and display of strawberry hardness.
To achieve rapid and non-destructive detection of strawberry hardness, strawberry hyperspectral data, and hardness information were collected for five consecutive days, and a hardness prediction method based on high spectral multi-index threshold layer-by-layer segmentation was proposed. Firstly, the spectral reflectance differences of different components (pulp, moldy pulp, strawberry seeds, and sepals) were analyzed, and the characteristic bands were identified. Subsequently, new normalized feature indices were constructed based on the characteristic bands, which were selected based on spectral reflectance differences, and the segmentation thresholds were determined. The layer-by-layer segmentation method was used to eliminate the interference of irrelevant parts. Three methods(successive projections algorithm, principal component analysis, and quadratic combination dimensionality reduction) were used to reduce the spectral information redundancy and extract features. The regression models were established for the original spectral data and the reduced feature data by random forest and partial least squares regression, respectively. The best prediction model was determined to fit the hardness of the strawberry pulp. The hardness distribution image was obtained for the intuitive display of the strawberry hardness prediction result. The result shows that the partial least squares model based on quadratic dimensionality reduction yielded the best performance, with correlation coefficients of 0.9101 and 0.9099 for the test set and prediction set, respectively, and with a root-mean-square error of 0.1344 for the test set. This study provides a reference for non-destructive detection and display of strawberry hardness.
2024, 48(3): 373-378.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.012
Abstract:
In order to realize the rapid quantitative detection of copper and zinc in brass samples, three brass samples were studied by calibration-free laser-induced breakdown spectroscopy(CF-LIBS). Six characteristic spectral lines of copper and four characteristic spectral lines of zinc were selected as the analysis objects. Under the spectrum acquisition modes of a single pulse and an average of 10 pulses, the relative standard deviation (RSD) of characteristic spectral lines of copper and zinc were comparatively analyzed. The results show that under the condition of an average of 10 pulses, the average RSD of copper and zinc is decreased by 8.8% and 12.3%, respectively, compared with the single pulse. The mass fraction of copper and zinc in three brass samples are quantitatively calculated by CF-LIBS, the relative errors between the temperature of copper and zinc and the average plasma temperature are 5.2%, 0.9%, and 7.1%, respectively. The relative errors in copper mass fraction in the three samples are 5.6%, 0.5% and, 1.6%, respectively; The relative errors in zinc mass fraction are 9.2%, 2.0%, and 13.3%, respectively. CF-LIBS is expected to provide method support for the rapid and accurate detection of element mass fraction in brass and provide data and technical support for the further development of portable detection equipment for the element mass fraction of alloy samples.
In order to realize the rapid quantitative detection of copper and zinc in brass samples, three brass samples were studied by calibration-free laser-induced breakdown spectroscopy(CF-LIBS). Six characteristic spectral lines of copper and four characteristic spectral lines of zinc were selected as the analysis objects. Under the spectrum acquisition modes of a single pulse and an average of 10 pulses, the relative standard deviation (RSD) of characteristic spectral lines of copper and zinc were comparatively analyzed. The results show that under the condition of an average of 10 pulses, the average RSD of copper and zinc is decreased by 8.8% and 12.3%, respectively, compared with the single pulse. The mass fraction of copper and zinc in three brass samples are quantitatively calculated by CF-LIBS, the relative errors between the temperature of copper and zinc and the average plasma temperature are 5.2%, 0.9%, and 7.1%, respectively. The relative errors in copper mass fraction in the three samples are 5.6%, 0.5% and, 1.6%, respectively; The relative errors in zinc mass fraction are 9.2%, 2.0%, and 13.3%, respectively. CF-LIBS is expected to provide method support for the rapid and accurate detection of element mass fraction in brass and provide data and technical support for the further development of portable detection equipment for the element mass fraction of alloy samples.
2024, 48(3): 379-386.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.013
Abstract:
In order to avoid that the problems of the existing hyperspectral image denoising optimization model only considers the limited intrinsic structure characteristics, and does not realize the accurate representation of image features, a denoising method based on spatial spectral depth image prior and smoothing is proposed. The model combines tight frame transform with a deep learning model with high expression and strong learning ability. Firstly, on the basis of low-rank matrix decomposition, the potential-spatial spectral features were learned by using specific depth images prior. Secondly, a tight frame of end and abundance matrix was constructed respectively to explore the local smoothing of the empty spectrum and solve the semi-fitting behavior of the depth image prior. Finally, an efficient iterative algorithm was designed to solve the model. The results show that the method based on space spectrum depth image prior has better performance under various complex noise interference, and the peak signal-to-noise ratio(PSNR) is improved by at least 1 dB, and high quality restored images are obtained. The method provides a reference for hyperspectral image denoising.
In order to avoid that the problems of the existing hyperspectral image denoising optimization model only considers the limited intrinsic structure characteristics, and does not realize the accurate representation of image features, a denoising method based on spatial spectral depth image prior and smoothing is proposed. The model combines tight frame transform with a deep learning model with high expression and strong learning ability. Firstly, on the basis of low-rank matrix decomposition, the potential-spatial spectral features were learned by using specific depth images prior. Secondly, a tight frame of end and abundance matrix was constructed respectively to explore the local smoothing of the empty spectrum and solve the semi-fitting behavior of the depth image prior. Finally, an efficient iterative algorithm was designed to solve the model. The results show that the method based on space spectrum depth image prior has better performance under various complex noise interference, and the peak signal-to-noise ratio(PSNR) is improved by at least 1 dB, and high quality restored images are obtained. The method provides a reference for hyperspectral image denoising.
2024, 48(3): 387-394.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.014
Abstract:
To investigate the effect of La2O3 on the organization and properties of laser melting shape memory alloy coatings, a broadband laser melting technique was used to prepare a La2O3-doped Fe17Mn5Si10Cr5Ni shape memory alloy composite coating on the surface of 42CrMo medium carbon steel. The results of the effect of La2O3 on the microstructure, Vickers hardness, wear resistance, corrosion resistance and surface residual stress of the composite coatings were obtained by characterization. The results show that La2O3 significantly reduces the grain size, when the doping mass fraction of 0.009 grain size reaches a minimum value of 3.03 μm. The composite coating has a maximum hardness of 454.7 HV0.2 and the lowest wear and the lowest self-corrosion current of 4.287×10-7 A/cm2 and the highest self-corrosion potential of -0.843 V are obtained, and the coating protection rate reaches 94.83%. La2O3 significantly improves the corrosion resistance of the coating. The residual stresses change from tensile to compressive stresses, with the highest residual compressive stress value of -378 MPa at Fe-Mn-Si/La2O3. This study provides a practical reference for the development and promotion of low-stress high-performance laser cladding coatings.
To investigate the effect of La2O3 on the organization and properties of laser melting shape memory alloy coatings, a broadband laser melting technique was used to prepare a La2O3-doped Fe17Mn5Si10Cr5Ni shape memory alloy composite coating on the surface of 42CrMo medium carbon steel. The results of the effect of La2O3 on the microstructure, Vickers hardness, wear resistance, corrosion resistance and surface residual stress of the composite coatings were obtained by characterization. The results show that La2O3 significantly reduces the grain size, when the doping mass fraction of 0.009 grain size reaches a minimum value of 3.03 μm. The composite coating has a maximum hardness of 454.7 HV0.2 and the lowest wear and the lowest self-corrosion current of 4.287×10-7 A/cm2 and the highest self-corrosion potential of -0.843 V are obtained, and the coating protection rate reaches 94.83%. La2O3 significantly improves the corrosion resistance of the coating. The residual stresses change from tensile to compressive stresses, with the highest residual compressive stress value of -378 MPa at Fe-Mn-Si/La2O3. This study provides a practical reference for the development and promotion of low-stress high-performance laser cladding coatings.
2024, 48(3): 395-404.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.015
Abstract:
To solve the problems of complex online inspection, low accuracy, and slow speed of profile inspection during laser dressing of complex profile forming grinding wheels, a set of image acquisition systems based on laser dressing grinding wheel machine tool was built, and a multi-profile image synthesis detection method was proposed, which defined the concept of profile wave band. Firstly, the guided filter and grayscale transformation were used to preprocess the grinding wheel profile image to enhance the features of the grinding wheel profile, and then the binary image was obtained by global thresholding segmentation and inverse binarization, and then the binary image sequence was synthesized into a profile frequency image with pixel frequency information, and the profile wave band was extracted by thresholding segmentation. Finally, the random sample consensus algorithm combined with the least squares method was used to fit the outermost edge point set of the wave band to complete the detection of straight line and circular arc features. The results show that the detection efficiency of the detection method is similar to that of the single-section visual inspection method, but the error range is reduced by 66.67%. Compared with the grinding remapping detection method, there is a fixed deviation in the inspection results, with an average error of about 7 μm, but the detection efficiency is increased by 91.67%. This method can detect the profile of the forming wheel with high precision and efficiency, which provides a new idea for the profile detection method of the forming grinding wheel.
To solve the problems of complex online inspection, low accuracy, and slow speed of profile inspection during laser dressing of complex profile forming grinding wheels, a set of image acquisition systems based on laser dressing grinding wheel machine tool was built, and a multi-profile image synthesis detection method was proposed, which defined the concept of profile wave band. Firstly, the guided filter and grayscale transformation were used to preprocess the grinding wheel profile image to enhance the features of the grinding wheel profile, and then the binary image was obtained by global thresholding segmentation and inverse binarization, and then the binary image sequence was synthesized into a profile frequency image with pixel frequency information, and the profile wave band was extracted by thresholding segmentation. Finally, the random sample consensus algorithm combined with the least squares method was used to fit the outermost edge point set of the wave band to complete the detection of straight line and circular arc features. The results show that the detection efficiency of the detection method is similar to that of the single-section visual inspection method, but the error range is reduced by 66.67%. Compared with the grinding remapping detection method, there is a fixed deviation in the inspection results, with an average error of about 7 μm, but the detection efficiency is increased by 91.67%. This method can detect the profile of the forming wheel with high precision and efficiency, which provides a new idea for the profile detection method of the forming grinding wheel.
2024, 48(3): 405-410.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.016
Abstract:
In order to obtain more process parameters for laser dressing bronze diamond grinding wheel according to experimental rules, this paper uses back propagation(BP)neural network, particle swarm optimization and genetic algorithm (PSO&GA) to establish a prediction model for laser dressing bronze diamond grinding wheel. Firstly, by analyzing the principle of laser dressing, the grinding wheel profile surface angle, laser deflection angle, incidence angle and spot overlap rate were obtained as the main influencing parameters, and 192 sets of process test data were trimmed with the grinding wheel surface angle error and peak-to-valley (PV)value as the evaluation index. Then, a 4×9×2 three-layer BP neural network prediction model was established, and the predictive model was trained and optimized by the PSO&GA hybrid optimization algorithm. Finally, 16 sets of experimental data were selected to test the BP neural network prediction model, and the prediction results were more accurate, and the training effects of the BP neural network by gradient descent (GD), particle swarm optimization (PSO) and genetic algorithm (GA) were compared. The results show that the angle error prediction bias of the BP neural network trained by the PSO&GA hybrid optimization algorithm is within 0.2°, and the prediction deviation of PV value is within 1.6 μm, and compared with other optimization algorithms, the BP neural network has a faster convergence speed and better convergence accuracy. It provides a good predictive model for laser dressing of bronze diamond grinding wheels.
In order to obtain more process parameters for laser dressing bronze diamond grinding wheel according to experimental rules, this paper uses back propagation(BP)neural network, particle swarm optimization and genetic algorithm (PSO&GA) to establish a prediction model for laser dressing bronze diamond grinding wheel. Firstly, by analyzing the principle of laser dressing, the grinding wheel profile surface angle, laser deflection angle, incidence angle and spot overlap rate were obtained as the main influencing parameters, and 192 sets of process test data were trimmed with the grinding wheel surface angle error and peak-to-valley (PV)value as the evaluation index. Then, a 4×9×2 three-layer BP neural network prediction model was established, and the predictive model was trained and optimized by the PSO&GA hybrid optimization algorithm. Finally, 16 sets of experimental data were selected to test the BP neural network prediction model, and the prediction results were more accurate, and the training effects of the BP neural network by gradient descent (GD), particle swarm optimization (PSO) and genetic algorithm (GA) were compared. The results show that the angle error prediction bias of the BP neural network trained by the PSO&GA hybrid optimization algorithm is within 0.2°, and the prediction deviation of PV value is within 1.6 μm, and compared with other optimization algorithms, the BP neural network has a faster convergence speed and better convergence accuracy. It provides a good predictive model for laser dressing of bronze diamond grinding wheels.
2024, 48(3): 411-415.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.017
Abstract:
In order to enhance the detection sensitivity of laser-induced breakdown spectroscopy (LIBS), the effects of different kinds of halogen ions with different concentrations on the LIBS signal enhancement were studied. Suitable-sized silver nanoparticles were selected based on experimental conditions and absorption spectrum information. The amphiphilic molecule sodium dodecyl sulfate (SDS) provided the necessary viscous force, while halogen ions were introduced into silicon wafers as substrates to induce aggregation of the silver sol nanoparticles. The results show that halide ions induce nanoparticle aggregation and enhance LIBS emission intensities of Mg Ⅰ lines at the wavelengths 382.9 nm, 383.2 nm and 383.8 nm. The limit of detection of Mg has been improved to μg/L. Effects of different halogen ions on nanoparticle enhanced LIBS (NELIBS) signal enhancement: F- > Cl- > Br- > I-. The achievements of this study demonstrate that halide ions induced nanoparticle aggregation enhanced NELIBS signal and can realize the element detection and analysis of Mg in liquid, which meets new requirements of modern analysis technology.
In order to enhance the detection sensitivity of laser-induced breakdown spectroscopy (LIBS), the effects of different kinds of halogen ions with different concentrations on the LIBS signal enhancement were studied. Suitable-sized silver nanoparticles were selected based on experimental conditions and absorption spectrum information. The amphiphilic molecule sodium dodecyl sulfate (SDS) provided the necessary viscous force, while halogen ions were introduced into silicon wafers as substrates to induce aggregation of the silver sol nanoparticles. The results show that halide ions induce nanoparticle aggregation and enhance LIBS emission intensities of Mg Ⅰ lines at the wavelengths 382.9 nm, 383.2 nm and 383.8 nm. The limit of detection of Mg has been improved to μg/L. Effects of different halogen ions on nanoparticle enhanced LIBS (NELIBS) signal enhancement: F- > Cl- > Br- > I-. The achievements of this study demonstrate that halide ions induced nanoparticle aggregation enhanced NELIBS signal and can realize the element detection and analysis of Mg in liquid, which meets new requirements of modern analysis technology.
2024, 48(3): 416-424.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.018
Abstract:
To improve the ability of low-level wind shear identification and early warning, the effects of the single-slope method, double-slope method, and regional-divergence method for low-level wind shear identification in glide path area were analyzed and evaluated based on the data of a Chinese FC-Ⅲ wind light detection and ranging(LiDAR) and pilot reports. Subsequently, the spatial and temporal distribution characteristics of wind shear in Urumqi Airport in winter and spring were analyzed by using radar recognition results. The evaluation shows that the regional divergence method has an optimal performance for identifying the low-level wind shear at the airport, with a success rate of 86.7%. 14:00—16:00 is the occurrence time peak of low-level wind shear at the airport, and the high occurrence periods in winter and spring are 14:00—18:00 and 12:00—20:00, respectively. The wind shear frequency of 35.7% and wind shear intensity of 0.0042/s in spring are higher than those of 17.6% and 0.004/s in winter. The differences in radar detection frequency of wind shears at different parts of the runway and glide paths indicate that the radar can accurately capture the specific location of wind shears. The study provides a reference for the application of wind LiDAR in the field of low-level wind shear recognition.
To improve the ability of low-level wind shear identification and early warning, the effects of the single-slope method, double-slope method, and regional-divergence method for low-level wind shear identification in glide path area were analyzed and evaluated based on the data of a Chinese FC-Ⅲ wind light detection and ranging(LiDAR) and pilot reports. Subsequently, the spatial and temporal distribution characteristics of wind shear in Urumqi Airport in winter and spring were analyzed by using radar recognition results. The evaluation shows that the regional divergence method has an optimal performance for identifying the low-level wind shear at the airport, with a success rate of 86.7%. 14:00—16:00 is the occurrence time peak of low-level wind shear at the airport, and the high occurrence periods in winter and spring are 14:00—18:00 and 12:00—20:00, respectively. The wind shear frequency of 35.7% and wind shear intensity of 0.0042/s in spring are higher than those of 17.6% and 0.004/s in winter. The differences in radar detection frequency of wind shears at different parts of the runway and glide paths indicate that the radar can accurately capture the specific location of wind shears. The study provides a reference for the application of wind LiDAR in the field of low-level wind shear recognition.
2024, 48(3): 425-431.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.019
Abstract:
In order to solve the problem that the inversion of the diffusion attenuation coefficient Kd relies on in-situ measurement data, a fusion algorithm based on Klett and Fernald algorithm was proposed in order to inverse Kd. The fusion algorithm was based on the inversion results of the Klett algorithm and calculated the light detection and ranging (LiDAR) ratio as prior information for the Fernald algorithm. Furthermore, the Fernald method was used to inverse the more accurate diffusion attenuation coefficient Kd. In order to solve the problem of insufficient measured signal data, a mathematical analytic model was used as the simulation data source of the fusion algorithm. Finally, based on the self-developed dual-frequency LiDAR system, the effectiveness of the simulated signal was verified through comparison with trial data. The root mean square error of the Kd value obtained by fusion algorithm inversion was 0.74 m-1, which was better than the traditional Klett method and Fernald method. The experimental results show that the proposed fusion algorithm can achieve the convergence and accuracy requirements for the inversion of the diffuse attenuation coefficient Kd and can quickly calculate the LiDAR ratio without prior information about the in-situ environment. This research is helpful to the bathymetry of shallow water and the measurement of optical parameters of ocean profiles.
In order to solve the problem that the inversion of the diffusion attenuation coefficient Kd relies on in-situ measurement data, a fusion algorithm based on Klett and Fernald algorithm was proposed in order to inverse Kd. The fusion algorithm was based on the inversion results of the Klett algorithm and calculated the light detection and ranging (LiDAR) ratio as prior information for the Fernald algorithm. Furthermore, the Fernald method was used to inverse the more accurate diffusion attenuation coefficient Kd. In order to solve the problem of insufficient measured signal data, a mathematical analytic model was used as the simulation data source of the fusion algorithm. Finally, based on the self-developed dual-frequency LiDAR system, the effectiveness of the simulated signal was verified through comparison with trial data. The root mean square error of the Kd value obtained by fusion algorithm inversion was 0.74 m-1, which was better than the traditional Klett method and Fernald method. The experimental results show that the proposed fusion algorithm can achieve the convergence and accuracy requirements for the inversion of the diffuse attenuation coefficient Kd and can quickly calculate the LiDAR ratio without prior information about the in-situ environment. This research is helpful to the bathymetry of shallow water and the measurement of optical parameters of ocean profiles.
2024, 48(3): 432-437.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.020
Abstract:
The present work mainly discussed the development trend of additive manufacturing (AM) technologies for titanium alloy, as well as the difficulties and research trends of nondestructive testing (NDT) of AM parts, especially for pore defects. The research progress of NDT technology for pore-type defects was introduced, including offline NDT technology, indirect NDT based on feature monitoring and online NDT technology. Then, the latest progress in laser ultrasonic testing technology for porosity in AM metal was reported. At last, the future research directions and suggestions for laser ultrasonic online NDT of metal additive manufacturing parts were shown.
The present work mainly discussed the development trend of additive manufacturing (AM) technologies for titanium alloy, as well as the difficulties and research trends of nondestructive testing (NDT) of AM parts, especially for pore defects. The research progress of NDT technology for pore-type defects was introduced, including offline NDT technology, indirect NDT based on feature monitoring and online NDT technology. Then, the latest progress in laser ultrasonic testing technology for porosity in AM metal was reported. At last, the future research directions and suggestions for laser ultrasonic online NDT of metal additive manufacturing parts were shown.
2024, 48(3): 438-442.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.021
Abstract:
In order to improve the stability of laser diode(LD) end-pumped Z-cavity solid-state lasers, take Nd∶GdVO4 solid-state laser as an example. The influence of parameters like arm length and radius of curvature of the lens of the Z-shaped resonant cavity on the Gaussian beam was studied by numerical simulation, while considering the crystal thermal lensing effect. The relationship curves of the focal length of the crystal thermal lens with the variation of the pump power, the relationship curves of the stability of the resonant cavity with the variation of the focal length of the thermal lens, and the relationship curves of the Gaussian beam waist radius with the variation of the resonant cavity split arm length and the lens radius of curvature were obtained. The profile curves of the Gaussian beam were plotted. The results indicate that a suitable selection of the split-arm parameters of the Z-cavity can obtain a smaller beam waist size at the main plane of the crystal and the rear-end mirror, and the results of this study can provide a theoretical reference and basis for the construction of resonant cavities of Z-type solid-state lasers.
In order to improve the stability of laser diode(LD) end-pumped Z-cavity solid-state lasers, take Nd∶GdVO4 solid-state laser as an example. The influence of parameters like arm length and radius of curvature of the lens of the Z-shaped resonant cavity on the Gaussian beam was studied by numerical simulation, while considering the crystal thermal lensing effect. The relationship curves of the focal length of the crystal thermal lens with the variation of the pump power, the relationship curves of the stability of the resonant cavity with the variation of the focal length of the thermal lens, and the relationship curves of the Gaussian beam waist radius with the variation of the resonant cavity split arm length and the lens radius of curvature were obtained. The profile curves of the Gaussian beam were plotted. The results indicate that a suitable selection of the split-arm parameters of the Z-cavity can obtain a smaller beam waist size at the main plane of the crystal and the rear-end mirror, and the results of this study can provide a theoretical reference and basis for the construction of resonant cavities of Z-type solid-state lasers.
2024, 48(3): 443-448.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.022
Abstract:
In order to solve the problem that the measurement sensitivity of laser speckles decreases when the water temperature is higher than 20 ℃, a laser speckle image recognition and detection method based on depth learning is proposed. The speckle image data sets of 20.1 ℃, 20.2 ℃, and 20.3 ℃ were constructed. A multi-scale convolution neural network was used, combined with appropriate loss function and data enhancement technology, to optimize the characteristics of laser speckle images. Through the training and testing experiments of deep learning models on speckle datasets, high accuracy recognition of underwater temperature information speckle images was achieved, solving the problem of decreased sensitivity in contrast saturation measurement. The experimental results show that compared with AlexNet, VGG, and ResNet models, the accuracy of the GoogleNet model in underwater temperature recognition of speckle images reaches 99%. This study provides theoretical support for the in-depth understanding of temperature field distribution and its impact and provides valuable reference for related application fields.
In order to solve the problem that the measurement sensitivity of laser speckles decreases when the water temperature is higher than 20 ℃, a laser speckle image recognition and detection method based on depth learning is proposed. The speckle image data sets of 20.1 ℃, 20.2 ℃, and 20.3 ℃ were constructed. A multi-scale convolution neural network was used, combined with appropriate loss function and data enhancement technology, to optimize the characteristics of laser speckle images. Through the training and testing experiments of deep learning models on speckle datasets, high accuracy recognition of underwater temperature information speckle images was achieved, solving the problem of decreased sensitivity in contrast saturation measurement. The experimental results show that compared with AlexNet, VGG, and ResNet models, the accuracy of the GoogleNet model in underwater temperature recognition of speckle images reaches 99%. This study provides theoretical support for the in-depth understanding of temperature field distribution and its impact and provides valuable reference for related application fields.
2024, 48(3): 449-454.
doi: 10.7510/jgjs.issn.1001-3806.2024.03.023
Abstract:
In order to reduce the hardness and improve the plasticity of the connection position of Hot formed steel, the Gaussian laser beam was used to quickly scan the surface of the sample, and the high-temperature tempering principle was used to reduce the strength and hardness of the material, thus successfully softening the material in local areas. The results show that the softening zone structure is tempered sorbite, and the slower the scanning speed, the better the softening effect. At temperature of 800 ℃ and scanning line speed of 2 mm/s, the softening zone hardness is 230 HV, which is 47.92% of the base metal hardness. The tensile strength of the softening zone decreases by 767 MPa, which is 50.53% of the base metal. Its elongation after fracture is 17.36%, which is 188.08% of the base metal. The softened tensile specimen exhibits ductile fracture with a cup-shaped cross-section. After softening, spot welding will not change the hardness of the weld nugget. It can widen the heat-affected zone, avoid the sudden drop and rise of its hardness, improve the stress concentration in the heat-affected zone, prevent the interface fracture of the welding spot, and increase the maximum displacement of the weld spot fracture by 60% after softening. Maximum energy absorption work increased by 10.14%. This study demonstrates the feasibility of laser softening hot-formed steel and has a certain guiding significance for the local softening of hot-stamped automotive parts.
In order to reduce the hardness and improve the plasticity of the connection position of Hot formed steel, the Gaussian laser beam was used to quickly scan the surface of the sample, and the high-temperature tempering principle was used to reduce the strength and hardness of the material, thus successfully softening the material in local areas. The results show that the softening zone structure is tempered sorbite, and the slower the scanning speed, the better the softening effect. At temperature of 800 ℃ and scanning line speed of 2 mm/s, the softening zone hardness is 230 HV, which is 47.92% of the base metal hardness. The tensile strength of the softening zone decreases by 767 MPa, which is 50.53% of the base metal. Its elongation after fracture is 17.36%, which is 188.08% of the base metal. The softened tensile specimen exhibits ductile fracture with a cup-shaped cross-section. After softening, spot welding will not change the hardness of the weld nugget. It can widen the heat-affected zone, avoid the sudden drop and rise of its hardness, improve the stress concentration in the heat-affected zone, prevent the interface fracture of the welding spot, and increase the maximum displacement of the weld spot fracture by 60% after softening. Maximum energy absorption work increased by 10.14%. This study demonstrates the feasibility of laser softening hot-formed steel and has a certain guiding significance for the local softening of hot-stamped automotive parts.
