Most image saliency detection models are dependent on prior knowledge and demand high computational cost. However, spectral residual（SR） and phase spectrum of the Fourier transform（PFT） models are simple and fast ...Most image saliency detection models are dependent on prior knowledge and demand high computational cost. However, spectral residual（SR） and phase spectrum of the Fourier transform（PFT） models are simple and fast saliency detection approaches based on two-dimensional Fourier transform without the prior knowledge. For seismic data, the geological structure of the underground rock formation changes more obviously in the time direction. Therefore, one-dimensional Fourier transform is more suitable for seismic saliency detection. Fractional Fourier transform（FrFT） is an improved algorithm for Fourier transform, therefore we propose the seismic SR and PFT models in one-dimensional FrF T domain to obtain more detailed saliency maps. These two models use the amplitude and phase information in FrFT domain to construct the corresponding saliency maps in spatial domain. By means of these two models, several saliency maps at different fractional orders can be obtained for seismic attribute analysis. These saliency maps can characterize the detailed features and highlight the object areas, which is more conducive to determine the location of reservoirs. The performance of the proposed method is assessed on both simulated and real seismic data. The results indicate that our method is effective and convenient for seismic attribute extraction with good noise immunity.展开更多
Seismic waveform clustering is a useful technique for lithologic identification and reservoir characterization.The current seismic waveform clustering algorithms are predominantly based on a fixed time window,which is...Seismic waveform clustering is a useful technique for lithologic identification and reservoir characterization.The current seismic waveform clustering algorithms are predominantly based on a fixed time window,which is applicable for layers of stable thickness.When a layer exhibits variable thickness in the seismic response,a fixed time window cannot provide comprehensive geologic information for the target interval.Therefore,we propose a novel approach for a waveform clustering workfl ow based on a variable time window to enable broader applications.The dynamic time warping(DTW)distance is fi rst introduced to effectively measure the similarities between seismic waveforms with various lengths.We develop a DTW distance-based clustering algorithm to extract centroids,and we then determine the class of all seismic traces according to the DTW distances from centroids.To greatly reduce the computational complexity in seismic data application,we propose a superpixel-based seismic data thinning approach.We further propose an integrated workfl ow that can be applied to practical seismic data by incorporating the DTW distance-based clustering and seismic data thinning algorithms.We evaluated the performance by applying the proposed workfl ow to synthetic seismograms and seismic survey data.Compared with the the traditional waveform clustering method,the synthetic seismogram results demonstrate the enhanced capability of the proposed workfl ow to detect boundaries of diff erent lithologies or lithologic associations with variable thickness.Results from a practical application show that the planar map of seismic waveform clustering obtained by the proposed workfl ow correlates well with the geological characteristics of wells in terms of reservoir thickness.展开更多
The critical breakdown path(CBP)has a significant impact on the breakdown voltage curve and the ignition time of heaterless hollow cathodes(HHCs).To determine the pattern of the variation in the CBP position and its i...The critical breakdown path(CBP)has a significant impact on the breakdown voltage curve and the ignition time of heaterless hollow cathodes(HHCs).To determine the pattern of the variation in the CBP position and its impact on ignition performance,a numerical model named the CBP evaluation(CBPE)was established in this paper to calculate the CBP of a HHC.The CBPE model can be used to screen various potential breakdown paths to identify those that are most likely to satisfy the Townsend breakdown conditions,which are denoted as CBPs.To verify the calculation accuracy of the CBPE model,4.5 A-level HHC ignition tests were conducted on HHCs with three different structures.By comparing the test results and the calculated results of the breakdown voltage,the calculation errors of the CBPE under three HHC conditions ranged from 1.6%to 5.8%,and the trends of the calculated results were consistent with those of the test results.The ignition test also showed the characteristics of the breakdown voltage curve and the ignition time for the three HHCs.Based on the CBPE model,an in-depth analysis was conducted on the mechanism of the patterns revealed by the tests.The main conclusions are presented as follows:(1)the CBP always shifts from the long path to the short path in the HHCs with an increasing gas flow rate;and(2)the ignition time of the HHCs depends on the position of the CBP because different CBP positions can cause different mechanisms of heat transfer from the plasma to the emitter.This study can guide the optimization of the CBP position and the corresponding ignition times of HHCs.展开更多
Inversion of Young’s modulus,Poisson’s ratio and density from pre-stack seismic data has been proved to be feasible and effective.However,the existing methods do not take full advantage of the prior information.With...Inversion of Young’s modulus,Poisson’s ratio and density from pre-stack seismic data has been proved to be feasible and effective.However,the existing methods do not take full advantage of the prior information.Without considering the lateral continuity of the inversion results,these methods need to invert the reflectivity first.In this paper,we propose multi-gather simultaneous inversion for pre-stack seismic data.Meanwhile,the total variation（TV）regularization,L1 norm regularization and initial model constraint are used.In order to solve the objective function contains L1norm,TV norm and L2 norm,we develop an algorithm based on split Bregman iteration.The main advantages of our method are as follows：（1）The elastic parameters are calculated directly from objective function rather than from their reflectivity,therefore the stability and accuracy of the inversion process can be ensured.（2）The inversion results are more in accordance with the prior geological information.（3）The lateral continuity of the inversion results are improved.The proposed method is illustrated by theoretical model data and experimented with a 2-D field data.展开更多
Supervised machine learning algorithms have been widely used in seismic exploration processing,but the lack of labeled examples complicates its application.Therefore,we propose a seismic labeled data expansion method ...Supervised machine learning algorithms have been widely used in seismic exploration processing,but the lack of labeled examples complicates its application.Therefore,we propose a seismic labeled data expansion method based on deep variational Autoencoders(VAE),which are made of neural networks and contains two partsEncoder and Decoder.Lack of training samples leads to overfitting of the network.We training the VAE with whole seismic data,which is a data-driven process and greatly alleviates the risk of overfitting.The Encoder captures the ability to map the seismic waveform Y to latent deep features z,and the Decoder captures the ability to reconstruct high-dimensional waveform Yb from latent deep features z.Later,we put the labeled seismic data into Encoders and get the latent deep features.We can easily use gaussian mixture model to fit the deep feature distribution of each class labeled data.We resample a mass of expansion deep features z* according to the Gaussian mixture model,and put the expansion deep features into the decoder to generate expansion seismic data.The experiments in synthetic and real data show that our method alleviates the problem of lacking labeled seismic data for supervised seismic facies analysis.展开更多
Geological surface modeling is typically based on seismic data, well data, and models of regional geology. However, structural interpretation of these data is error-prone, especially in the absence of structural morph...Geological surface modeling is typically based on seismic data, well data, and models of regional geology. However, structural interpretation of these data is error-prone, especially in the absence of structural morphology information, Existing geological surface models suffer from high levels of uncertainty, which exposes oil and gas exploration and development to additional risk. In this paper, we achieve a reconstruction of the uncertainties associated with a geological surface using chance-constrained programming based on multisource data. We also quantifi ed the uncertainty of the modeling data and added a disturbance term to the objective function. Finally, we verifi ed the applicability of the method using both synthetic and real fault data. We found that the reconstructed geological models met geological rules and reduced the reconstruction uncertainty.展开更多
The utilization of urban underground space in a smart city requires an accurate understanding of the underground structure.As an effective technique,Rayleigh wave exploration can accurately obtain information on the s...The utilization of urban underground space in a smart city requires an accurate understanding of the underground structure.As an effective technique,Rayleigh wave exploration can accurately obtain information on the subsurface.In particular,Rayleigh wave dispersion curves can be used to determine the near-surface shear-wave velocity structure.This is a typical multiparameter,high-dimensional nonlinear inverse problem because the velocities and thickness of each layer must be inverted simultaneously.Nonlinear methods such as simulated annealing(SA)are commonly used to solve this inverse problem.However,SA controls the iterative process though temperature rather than the error,and the search direction is random;hence,SA always falls into a local optimum when the temperature setting is inaccurate.Specifically,for the inversion of Rayleigh wave dispersion curves,the inversion accuracy will decrease with an increasing number of layers due to the greater number of inversion parameters and large dimension.To solve the above problems,we convert the multiparameter,highdimensional inverse problem into multiple low-dimensional optimizations to improve the algorithm accuracy by incorporating the principle of block coordinate descent(BCD)into SA.Then,we convert the temperature control conditions in the original SA method into error control conditions.At the same time,we introduce the differential evolution(DE)method to ensure that the iterative error steadily decreases by correcting the iterative error direction in each iteration.Finally,the inversion stability is improved,and the proposed inversion method,the block coordinate descent differential evolution simulated annealing(BCDESA)algorithm,is implemented.The performance of BCDESA is validated by using both synthetic data and field data from western China.The results show that the BCDESA algorithm has stronger global optimization capabilities than SA,and the inversion results have higher stability and accuracy.In addition,synthetic data analysis also shows that BCDESA can avoid the problems of the conventional SA method,which assumes the S-wave velocity structure in advance.The robustness and adaptability of the algorithm are improved,and more accurate shear-wave velocity and thickness information can be extracted from Rayleigh wave dispersion curves.展开更多
Calculating the flow accumulation matrix is an essential step for many hydrological and topographical analyses.This study gives an overview of the existing algorithms for flow accumulation calculations for singleflow ...Calculating the flow accumulation matrix is an essential step for many hydrological and topographical analyses.This study gives an overview of the existing algorithms for flow accumulation calculations for singleflow direction matrices.A fast and simple algorithm for calculating flow accumulation matrices is proposed in this study.The algorithm identifies three types of cells in a flow direction matrix: source cells,intersection cells,and interior cells.It traverses all source cells and traces the downstream interior cells of each source cell until an intersection cell is encountered.An intersection cell is treated as an interior cell when its last drainage path is traced and the tracing continues with its downstream cells.Experiments are conducted on thirty datasets with a resolution of 3 m.Compared with the existing algorithms for flow accumulation calculation,the proposed algorithm is easy to implement,runs much faster than existing algorithms,and generally requires less memory space.展开更多
In this paper,a cladding-pumped erbium-ytterbium co-doped random fiber laser(EYRFL)operating at 1550 nm with high power laser diode(LD)is proposed and experimentally demonstrated for the first time.The laser cavity in...In this paper,a cladding-pumped erbium-ytterbium co-doped random fiber laser(EYRFL)operating at 1550 nm with high power laser diode(LD)is proposed and experimentally demonstrated for the first time.The laser cavity includes a 5-m-long erbium-ytterbium co-doped fiber that serves as the gain medium,as well as a 2-km-long single-mode fiber(SMF)to provide random distributed feedback.As a result,stable 2.14 W of 1550 nm random lasing at 9.80 W of 976 nm LD pump power and a linear output with the slope efficiency as 22.7%are generated.This simple and novel random fiber laser could provide a promising way to develop high power 1.5μm light sources.展开更多
Depressions in raster digital elevation models(DEM)present a challenge for extracting hydrological networks.They are commonly filled before subsequent algorithms are further applied.Among existing algorithms for filli...Depressions in raster digital elevation models(DEM)present a challenge for extracting hydrological networks.They are commonly filled before subsequent algorithms are further applied.Among existing algorithms for filling depressions,the Priority-Flood algorithm runs the fastest.In this study,we propose an improved variant over the fastest existing sequential variant of the Priority-Flood algorithm for filling depressions in floating-point DEMs.The proposed variant introduces a series of improvements and greatly reduces the number of cells that need to be processed by the priority queue(PQ),the key data structure used in the algorithm.The proposed variant is evaluated based on statistics from 30 experiments.On average,our proposed variant reduces the number of cells processed by the PQ by around 70%.The speed-up ratios of our proposed variant over the existing fastest variant of the Priority-Flood algorithm range from 31%to 52%,with an average of 45%.The proposed variant can be used to fill depressions in large DEMs in much less time and in the parallel implementation of the Priority-Flood algorithm to further reduce the running time for processing huge DEMs that cannot be dealt with easily on single computers.展开更多
基金supported by the National Natural Science Foundation of China (Nos.61571096,61775030,41274127,41301460,and 40874066)
文摘Most image saliency detection models are dependent on prior knowledge and demand high computational cost. However, spectral residual（SR） and phase spectrum of the Fourier transform（PFT） models are simple and fast saliency detection approaches based on two-dimensional Fourier transform without the prior knowledge. For seismic data, the geological structure of the underground rock formation changes more obviously in the time direction. Therefore, one-dimensional Fourier transform is more suitable for seismic saliency detection. Fractional Fourier transform（FrFT） is an improved algorithm for Fourier transform, therefore we propose the seismic SR and PFT models in one-dimensional FrF T domain to obtain more detailed saliency maps. These two models use the amplitude and phase information in FrFT domain to construct the corresponding saliency maps in spatial domain. By means of these two models, several saliency maps at different fractional orders can be obtained for seismic attribute analysis. These saliency maps can characterize the detailed features and highlight the object areas, which is more conducive to determine the location of reservoirs. The performance of the proposed method is assessed on both simulated and real seismic data. The results indicate that our method is effective and convenient for seismic attribute extraction with good noise immunity.
基金supported by the National Science and Technology Major Project (No. 2017ZX05001-003)。
文摘Seismic waveform clustering is a useful technique for lithologic identification and reservoir characterization.The current seismic waveform clustering algorithms are predominantly based on a fixed time window,which is applicable for layers of stable thickness.When a layer exhibits variable thickness in the seismic response,a fixed time window cannot provide comprehensive geologic information for the target interval.Therefore,we propose a novel approach for a waveform clustering workfl ow based on a variable time window to enable broader applications.The dynamic time warping(DTW)distance is fi rst introduced to effectively measure the similarities between seismic waveforms with various lengths.We develop a DTW distance-based clustering algorithm to extract centroids,and we then determine the class of all seismic traces according to the DTW distances from centroids.To greatly reduce the computational complexity in seismic data application,we propose a superpixel-based seismic data thinning approach.We further propose an integrated workfl ow that can be applied to practical seismic data by incorporating the DTW distance-based clustering and seismic data thinning algorithms.We evaluated the performance by applying the proposed workfl ow to synthetic seismograms and seismic survey data.Compared with the the traditional waveform clustering method,the synthetic seismogram results demonstrate the enhanced capability of the proposed workfl ow to detect boundaries of diff erent lithologies or lithologic associations with variable thickness.Results from a practical application show that the planar map of seismic waveform clustering obtained by the proposed workfl ow correlates well with the geological characteristics of wells in terms of reservoir thickness.
文摘The critical breakdown path(CBP)has a significant impact on the breakdown voltage curve and the ignition time of heaterless hollow cathodes(HHCs).To determine the pattern of the variation in the CBP position and its impact on ignition performance,a numerical model named the CBP evaluation(CBPE)was established in this paper to calculate the CBP of a HHC.The CBPE model can be used to screen various potential breakdown paths to identify those that are most likely to satisfy the Townsend breakdown conditions,which are denoted as CBPs.To verify the calculation accuracy of the CBPE model,4.5 A-level HHC ignition tests were conducted on HHCs with three different structures.By comparing the test results and the calculated results of the breakdown voltage,the calculation errors of the CBPE under three HHC conditions ranged from 1.6%to 5.8%,and the trends of the calculated results were consistent with those of the test results.The ignition test also showed the characteristics of the breakdown voltage curve and the ignition time for the three HHCs.Based on the CBPE model,an in-depth analysis was conducted on the mechanism of the patterns revealed by the tests.The main conclusions are presented as follows:(1)the CBP always shifts from the long path to the short path in the HHCs with an increasing gas flow rate;and(2)the ignition time of the HHCs depends on the position of the CBP because different CBP positions can cause different mechanisms of heat transfer from the plasma to the emitter.This study can guide the optimization of the CBP position and the corresponding ignition times of HHCs.
基金supported by the National Natural Science Foundation of China (Nos.61775030,61571096,41301460,61362018,and 41274127)the key projects of Hunan Provincial Department of Education (No.16A174)
文摘Inversion of Young’s modulus,Poisson’s ratio and density from pre-stack seismic data has been proved to be feasible and effective.However,the existing methods do not take full advantage of the prior information.Without considering the lateral continuity of the inversion results,these methods need to invert the reflectivity first.In this paper,we propose multi-gather simultaneous inversion for pre-stack seismic data.Meanwhile,the total variation（TV）regularization,L1 norm regularization and initial model constraint are used.In order to solve the objective function contains L1norm,TV norm and L2 norm,we develop an algorithm based on split Bregman iteration.The main advantages of our method are as follows：（1）The elastic parameters are calculated directly from objective function rather than from their reflectivity,therefore the stability and accuracy of the inversion process can be ensured.（2）The inversion results are more in accordance with the prior geological information.（3）The lateral continuity of the inversion results are improved.The proposed method is illustrated by theoretical model data and experimented with a 2-D field data.
基金Supported by National Natural Science Foundation of China(41804126,41604107).
文摘Supervised machine learning algorithms have been widely used in seismic exploration processing,but the lack of labeled examples complicates its application.Therefore,we propose a seismic labeled data expansion method based on deep variational Autoencoders(VAE),which are made of neural networks and contains two partsEncoder and Decoder.Lack of training samples leads to overfitting of the network.We training the VAE with whole seismic data,which is a data-driven process and greatly alleviates the risk of overfitting.The Encoder captures the ability to map the seismic waveform Y to latent deep features z,and the Decoder captures the ability to reconstruct high-dimensional waveform Yb from latent deep features z.Later,we put the labeled seismic data into Encoders and get the latent deep features.We can easily use gaussian mixture model to fit the deep feature distribution of each class labeled data.We resample a mass of expansion deep features z* according to the Gaussian mixture model,and put the expansion deep features into the decoder to generate expansion seismic data.The experiments in synthetic and real data show that our method alleviates the problem of lacking labeled seismic data for supervised seismic facies analysis.
基金by National Science and Technology Major Project(Grant No.2017ZX05018004004)the National Natural Science Foundation of China (No.U1562218 & 41604107).
文摘Geological surface modeling is typically based on seismic data, well data, and models of regional geology. However, structural interpretation of these data is error-prone, especially in the absence of structural morphology information, Existing geological surface models suffer from high levels of uncertainty, which exposes oil and gas exploration and development to additional risk. In this paper, we achieve a reconstruction of the uncertainties associated with a geological surface using chance-constrained programming based on multisource data. We also quantifi ed the uncertainty of the modeling data and added a disturbance term to the objective function. Finally, we verifi ed the applicability of the method using both synthetic and real fault data. We found that the reconstructed geological models met geological rules and reduced the reconstruction uncertainty.
基金Supported by National Natural Science Foundation of China(NOs.41974150,42174158,42174151,41804126)a supporting program for outstanding talent of the University of Electronic Science and Technology of China(No.2019-QR-01)+1 种基金Project of Basic Scientific Research Operating Expenses of Central Universities(ZYGX2019J071ZYGX 2020J013).
文摘The utilization of urban underground space in a smart city requires an accurate understanding of the underground structure.As an effective technique,Rayleigh wave exploration can accurately obtain information on the subsurface.In particular,Rayleigh wave dispersion curves can be used to determine the near-surface shear-wave velocity structure.This is a typical multiparameter,high-dimensional nonlinear inverse problem because the velocities and thickness of each layer must be inverted simultaneously.Nonlinear methods such as simulated annealing(SA)are commonly used to solve this inverse problem.However,SA controls the iterative process though temperature rather than the error,and the search direction is random;hence,SA always falls into a local optimum when the temperature setting is inaccurate.Specifically,for the inversion of Rayleigh wave dispersion curves,the inversion accuracy will decrease with an increasing number of layers due to the greater number of inversion parameters and large dimension.To solve the above problems,we convert the multiparameter,highdimensional inverse problem into multiple low-dimensional optimizations to improve the algorithm accuracy by incorporating the principle of block coordinate descent(BCD)into SA.Then,we convert the temperature control conditions in the original SA method into error control conditions.At the same time,we introduce the differential evolution(DE)method to ensure that the iterative error steadily decreases by correcting the iterative error direction in each iteration.Finally,the inversion stability is improved,and the proposed inversion method,the block coordinate descent differential evolution simulated annealing(BCDESA)algorithm,is implemented.The performance of BCDESA is validated by using both synthetic data and field data from western China.The results show that the BCDESA algorithm has stronger global optimization capabilities than SA,and the inversion results have higher stability and accuracy.In addition,synthetic data analysis also shows that BCDESA can avoid the problems of the conventional SA method,which assumes the S-wave velocity structure in advance.The robustness and adaptability of the algorithm are improved,and more accurate shear-wave velocity and thickness information can be extracted from Rayleigh wave dispersion curves.
基金the National Natural Science Foundation of China (Grant No.41671427)the Fundamental Research Funds for the Central Universities (ZYGX2016J148).
文摘Calculating the flow accumulation matrix is an essential step for many hydrological and topographical analyses.This study gives an overview of the existing algorithms for flow accumulation calculations for singleflow direction matrices.A fast and simple algorithm for calculating flow accumulation matrices is proposed in this study.The algorithm identifies three types of cells in a flow direction matrix: source cells,intersection cells,and interior cells.It traverses all source cells and traces the downstream interior cells of each source cell until an intersection cell is encountered.An intersection cell is treated as an interior cell when its last drainage path is traced and the tracing continues with its downstream cells.Experiments are conducted on thirty datasets with a resolution of 3 m.Compared with the existing algorithms for flow accumulation calculation,the proposed algorithm is easy to implement,runs much faster than existing algorithms,and generally requires less memory space.
基金supported by the National Natural Science Foundation of China(Grant Nos.61635005,61205048,and 61290312)the PCSIRT project(Grant No.IRT1218)+1 种基金the 111 project(Grant No.B14039)the Sichuan Youth Science and Technology Foundation(Grant No.2016JQ0034).
文摘In this paper,a cladding-pumped erbium-ytterbium co-doped random fiber laser(EYRFL)operating at 1550 nm with high power laser diode(LD)is proposed and experimentally demonstrated for the first time.The laser cavity includes a 5-m-long erbium-ytterbium co-doped fiber that serves as the gain medium,as well as a 2-km-long single-mode fiber(SMF)to provide random distributed feedback.As a result,stable 2.14 W of 1550 nm random lasing at 9.80 W of 976 nm LD pump power and a linear output with the slope efficiency as 22.7%are generated.This simple and novel random fiber laser could provide a promising way to develop high power 1.5μm light sources.
基金the National Natural Science Foundation of China[grant number 41671427]the Open Fund of the State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau[grant number A314021402-1710]the Fundamental Research Funds for the Central Universities[grant number ZYGX2016J148]。
文摘Depressions in raster digital elevation models(DEM)present a challenge for extracting hydrological networks.They are commonly filled before subsequent algorithms are further applied.Among existing algorithms for filling depressions,the Priority-Flood algorithm runs the fastest.In this study,we propose an improved variant over the fastest existing sequential variant of the Priority-Flood algorithm for filling depressions in floating-point DEMs.The proposed variant introduces a series of improvements and greatly reduces the number of cells that need to be processed by the priority queue(PQ),the key data structure used in the algorithm.The proposed variant is evaluated based on statistics from 30 experiments.On average,our proposed variant reduces the number of cells processed by the PQ by around 70%.The speed-up ratios of our proposed variant over the existing fastest variant of the Priority-Flood algorithm range from 31%to 52%,with an average of 45%.The proposed variant can be used to fill depressions in large DEMs in much less time and in the parallel implementation of the Priority-Flood algorithm to further reduce the running time for processing huge DEMs that cannot be dealt with easily on single computers.
