Machine picking in cotton is an emerging practice in India,to solve the problems of labour shortages and production costs increasing.Cotton production has been declining in recent years;however,the high density planti...Machine picking in cotton is an emerging practice in India,to solve the problems of labour shortages and production costs increasing.Cotton production has been declining in recent years;however,the high density planting system(HDPS)offers a viable method to enhance productivity by increasing plant populations per unit area,optimizing resource utilization,and facilitating machine picking.Cotton is an indeterminate plant that produce excessive vegeta-tive growth in favorable soil fertility and moisture conditions,which posing challenges for efficient machine picking.To address this issue,the application of plant growth retardants(PGRs)is essential for controlling canopy architecture.PGRs reduce internode elongation,promote regulated branching,and increase plant compactness,making cotton plants better suited for machine picking.PGRs application also optimizes photosynthates distribution between veg-etative and reproductive growth,resulting in higher yields and improved fibre quality.The integration of HDPS and PGRs applications results in an optimal plant architecture for improving machine picking efficiency.However,the success of this integration is determined by some factors,including cotton variety,environmental conditions,and geographical variations.These approaches not only address yield stagnation and labour shortages but also help to establish more effective and sustainable cotton farming practices,resulting in higher cotton productivity.展开更多
Real-time identification of rock strength and cuttability based on monitoring while cutting during excavation is essential for key procedures such as the precise adjustment of excavation parameters and the in-situ mod...Real-time identification of rock strength and cuttability based on monitoring while cutting during excavation is essential for key procedures such as the precise adjustment of excavation parameters and the in-situ modification of hard rocks.This study proposes an in-telligent approach for predicting rock strength and cuttability.A database comprising 132 data sets is established,containing cutting para-meters(such as cutting depth and pick angle),cutting responses(such as specific energy and instantaneous cutting rate),and rock mech-anical parameters collected from conical pick-cutting experiments.These parameters serve as input features for predicting the uniaxial compressive strength and tensile strength of rocks using regression fitting and machine learning methodologies.In addition,rock cuttabil-ity is classified using a combination of the analytic hierarchy process and fuzzy comprehensive evaluation method,and subsequently iden-tified through machine learning approaches.Various models are compared to determine the optimal predictive and classification models.The results indicate that the optimal model for uniaxial compressive strength and tensile strength prediction is the genetic algorithm-optimized backpropagation neural network model,and the optimal model for rock cuttability classification is the radial basis neural network model.展开更多
Water-weakening presents a promising strategy for the in-situ improvement of rock cuttability.This study unveils the influences of water saturation on the mechanical response and fragmentation characteristics of rock ...Water-weakening presents a promising strategy for the in-situ improvement of rock cuttability.This study unveils the influences of water saturation on the mechanical response and fragmentation characteristics of rock samples.A series of rock-cutting tests using conical pick indentation was conducted on three types of sandstone samples under both dry and water-saturated conditions.The relationships between cutting force and indentation depth,as well as typical cuttability indices are determined and compared for dry and water-saturated samples.The experimental results reveal that the presence of water facilitates shearing failure in rock samples,as well as alleviates the fluctuations in the cutting force-indentation depth curve Furthermore,the peak cutting force(F_(p)),cutting work(W_(p)),and specific energy(SE)undergo apparent decrease after water saturation,whereas the trend in the indentation depth at rock failure(D_(f))varies across different rock types.Additionally,the water-induced percentage reductions in F_(p)and SE correlate positively with the quartz and swelling clay content within the rocks,suggesting that the cuttability improvement due to water saturation is attributed to the combined effects of stress corrosion and frictional reduction.These findings carry significant implications for improving rock cuttability in mechanized excavation of hard rock formations.展开更多
With the continuous expansion of deep underground engineering and the growing demand for safety monitoring,microseismic monitoring has become a core method for early warning of rock mass fracture and engineering stabi...With the continuous expansion of deep underground engineering and the growing demand for safety monitoring,microseismic monitoring has become a core method for early warning of rock mass fracture and engineering stability assessment.To address problems in existing methods,such as low data processing efficiency and poor phase recognition accuracy under low signal-to-noise ratio(SNR)conditions in complex geological environments,this study proposes an intelligent phase picking model based on ResUNet.The model integrates the residual learning mechanism of ResNet with the multi-scale feature extraction capability of UNet,effectively mitigating the vanishing gradient problem in deep networks.It also achieves cross-layer fusion of shallow detail features and deep semantic features through skip connections in the encoder-decoder structure.Compared with traditional short-time average/long-time average(STA/LTA)algorithms and advanced neural network models such as PhaseNet and EQTransformer,ResUNet shows superior performance in picking P-and S-wave phases.The model was trained on 400000 labeled microseismic signals from the Stanford earthquake dataset(STEAD)and was successfully applied to the Shizhuyuan polymetallic mine in Hunan Province,China.The results demonstrate that ResUNet achieves high picking accuracy and robustness in complex geological conditions,offering reliable technical support for early warning of disasters such as rockburst in deep underground engineering.展开更多
The availability of a tremendous amount of seismic data demands seismological researchers to analyze seismic phases efficiently.Recently,deep learning algorithms exhibit a powerful capability of detecting and picking ...The availability of a tremendous amount of seismic data demands seismological researchers to analyze seismic phases efficiently.Recently,deep learning algorithms exhibit a powerful capability of detecting and picking on P-and S-wave phases.However,it remains a challenge to effeciently process enormous teleseismic phases,which are crucial to probe Earth’s interior structures and their dynamics.In this study,we propose a scheme to detect and pick teleseismic phases,such as seismic phase that reflects off the core-mantle boundary(i.e.,PcP)and that reflects off the inner-core boundary(i.e.,PKiKP),from a seismic dataset in Japan.The scheme consists of three steps:1)latent phase traces are truncated from the whole seismogram with theoretical arrival times;2)latent phases are recognized and evaluated by convolutional neural network(CNN)models;3)arrivals of good or fair phase are picked with another CNN models.The testing detection result on 7386 seismograms shows that the scheme recognizes 92.15%and 94.13%of PcP and PKiKP phases.The testing picking result has a mean absolute error of 0.0742 s and 0.0636 s for the PcP and PKiKP phases,respectively.These seismograms were processed in just 5 min for phase detection and picking,demonstrating the efficiency of the proposed scheme in automatic teleseismic phase analysis.展开更多
Picking velocities from semblances manually is laborious and necessitates experience. Although various methods for automatic velocity picking have been developed, there remains a challenge in efficiently incorporating...Picking velocities from semblances manually is laborious and necessitates experience. Although various methods for automatic velocity picking have been developed, there remains a challenge in efficiently incorporating information from nearby gathers to ensure picked velocity aligns with seismic horizons while also improving picking accuracy. The conventional method of velocity picking from a semblance volume is computationally demanding, highlighting a need for a more efficient strategy. In this study, we introduce a novel method for automatic velocity picking based on multi-object tracking. This dynamic tracking process across different semblance panels can integrate information from nearby gathers effectively while maintaining computational efficiency. First, we employ accelerated density clustering on the velocity spectrum to discern cluster centers without the requirement for prior knowledge regarding the number of clusters. These cluster centers embody the maximum likelihood velocities of the main subsurface structures. Second, our proposed method tracks key points within the semblance volume. Kalman filter is adopted to adjust the tracking process, followed by interpolation on these tracked points to construct the final velocity model. Our synthetic data example demonstrates that our proposed algorithm can effectively rectify the picking errors of the clustering algorithm. We further compare the performances of the clustering method(CM), the proposed tracking method(TM), and the variational method(VM) on a field dataset from the Gulf of Mexico. The results attest that our method offers superior accuracy than CM, achieves comparable accuracy with VM, and benefits from a reduced computational cost.展开更多
The picking efficiency of seismic first breaks(FBs)has been greatly accelerated by deep learning(DL)technology.However,the picking accuracy and efficiency of DL methods still face huge challenges in low signal-to-nois...The picking efficiency of seismic first breaks(FBs)has been greatly accelerated by deep learning(DL)technology.However,the picking accuracy and efficiency of DL methods still face huge challenges in low signal-to-noise ratio(SNR)situations.To address this issue,we propose a regression approach to pick FBs based on bidirectional long short-term memory(Bi LSTM)neural network by learning the implicit Eikonal equation of 3D inhomogeneous media with rugged topography in the target region.We employ a regressive model that represents the relationships among the elevation of shots,offset and the elevation of receivers with their seismic traveltime to predict the unknown FBs,from common-shot gathers with sparsely distributed traces.Different from image segmentation methods which automatically extract image features and classify FBs from seismic data,the proposed method can learn the inner relationship between field geometry and FBs.In addition,the predicted results by the regressive model are continuous values of FBs rather than the discrete ones of the binary distribution.The picking results of synthetic data shows that the proposed method has low dependence on label data,and can obtain reliable and similar predicted results using two types of label data with large differences.The picking results of9380 shots for 3D seismic data generated by vibroseis indicate that the proposed method can still accurately predict FBs in low SNR data.The subsequent stacked profiles further illustrate the reliability and effectiveness of the proposed method.The results of model data and field seismic data demonstrate that the proposed regression method is a robust first-break picker with high potential for field application.展开更多
The precision and reliability of first-arrival picking are crucial for determining the accuracy of geological structure inversion using active source ocean bottom seismometer(OBS)refraction data.Traditional methods fo...The precision and reliability of first-arrival picking are crucial for determining the accuracy of geological structure inversion using active source ocean bottom seismometer(OBS)refraction data.Traditional methods for first-arrival picking based on sample points are characterized by theoretical errors,especially in low-sampling-frequency OBS data because the travel time of seismic waves is not an integer multiple of the sampling interval.In this paper,a first-arrival picking method that utilizes the spatial waveform variation characteristics of active source OBS data is presented.First,the distribution law of theoretical error is examined;adjacent traces exhibit variation characteristics in their waveforms.Second,a label cross-correlation superposition method for extracting highfrequency signals is presented to enhance the first-arrival picking precision.Results from synthetic and field data verify that the proposed approach is robust,successfully overcomes the limitations of low sampling frequency,and achieves precise outcomes that are comparable with those of high-sampling-frequency data.展开更多
Conical picks are important tools for rock mechanical excavation.Mean cutting force(MCF)of conical pick determines the suitability of the target rock for mechanical excavation.Accurate evaluation of MCF is important f...Conical picks are important tools for rock mechanical excavation.Mean cutting force(MCF)of conical pick determines the suitability of the target rock for mechanical excavation.Accurate evaluation of MCF is important for pick design and rock cutting.This study proposed hybrid methods composed of boosting trees and Bayesian optimization(BO)for accurate evaluation of MCF.220 datasets including uniaxial compression strength,tensile strength,tip angle(θ),attack angle,and cutting depth,were collected.Four boosting trees were developed based on the database to predict MCF.BO optimized the hyper-parameters of these boosting trees.Model evaluation suggested that the proposed hybrid models outperformed many commonly utilized machine learning models.The hybrid model composed of BO and categorical boosting(BO-CatBoost)was the best.Its outstanding performance was attributed to its advantages in dealing with categorical features(θincluded 6 types of angles and could be considered as categorical features).A graphical user interface was developed to facilitate the application of BO-CatBoost for the estimation of MCF.Moreover,the influences of the input parameters on the model and their relationship with MCF were analyzed.Whenθincreased from 80°to 90°,it had a significant contribution to the increase of MCF.展开更多
文摘Machine picking in cotton is an emerging practice in India,to solve the problems of labour shortages and production costs increasing.Cotton production has been declining in recent years;however,the high density planting system(HDPS)offers a viable method to enhance productivity by increasing plant populations per unit area,optimizing resource utilization,and facilitating machine picking.Cotton is an indeterminate plant that produce excessive vegeta-tive growth in favorable soil fertility and moisture conditions,which posing challenges for efficient machine picking.To address this issue,the application of plant growth retardants(PGRs)is essential for controlling canopy architecture.PGRs reduce internode elongation,promote regulated branching,and increase plant compactness,making cotton plants better suited for machine picking.PGRs application also optimizes photosynthates distribution between veg-etative and reproductive growth,resulting in higher yields and improved fibre quality.The integration of HDPS and PGRs applications results in an optimal plant architecture for improving machine picking efficiency.However,the success of this integration is determined by some factors,including cotton variety,environmental conditions,and geographical variations.These approaches not only address yield stagnation and labour shortages but also help to establish more effective and sustainable cotton farming practices,resulting in higher cotton productivity.
基金supported by the National Natural Science Foundation of China(Nos.52174099 and 52474168)the Science and Technology Innovation Program of Hunan Province,China(No.2023RC3050)+1 种基金the Natural Science Foundation of Hunan,China(No.2024JJ4064)the Open Fund of the State Key Laboratory of Safety Technology of Metal Mines(No.kfkt2023-01).
文摘Real-time identification of rock strength and cuttability based on monitoring while cutting during excavation is essential for key procedures such as the precise adjustment of excavation parameters and the in-situ modification of hard rocks.This study proposes an in-telligent approach for predicting rock strength and cuttability.A database comprising 132 data sets is established,containing cutting para-meters(such as cutting depth and pick angle),cutting responses(such as specific energy and instantaneous cutting rate),and rock mech-anical parameters collected from conical pick-cutting experiments.These parameters serve as input features for predicting the uniaxial compressive strength and tensile strength of rocks using regression fitting and machine learning methodologies.In addition,rock cuttabil-ity is classified using a combination of the analytic hierarchy process and fuzzy comprehensive evaluation method,and subsequently iden-tified through machine learning approaches.Various models are compared to determine the optimal predictive and classification models.The results indicate that the optimal model for uniaxial compressive strength and tensile strength prediction is the genetic algorithm-optimized backpropagation neural network model,and the optimal model for rock cuttability classification is the radial basis neural network model.
基金supported by financial grants from the National Natural Science Foundation of China(Grant Nos.52334003 and 52104111)the National Key R&D Program of China(Grant No.2022YFC2905600)。
文摘Water-weakening presents a promising strategy for the in-situ improvement of rock cuttability.This study unveils the influences of water saturation on the mechanical response and fragmentation characteristics of rock samples.A series of rock-cutting tests using conical pick indentation was conducted on three types of sandstone samples under both dry and water-saturated conditions.The relationships between cutting force and indentation depth,as well as typical cuttability indices are determined and compared for dry and water-saturated samples.The experimental results reveal that the presence of water facilitates shearing failure in rock samples,as well as alleviates the fluctuations in the cutting force-indentation depth curve Furthermore,the peak cutting force(F_(p)),cutting work(W_(p)),and specific energy(SE)undergo apparent decrease after water saturation,whereas the trend in the indentation depth at rock failure(D_(f))varies across different rock types.Additionally,the water-induced percentage reductions in F_(p)and SE correlate positively with the quartz and swelling clay content within the rocks,suggesting that the cuttability improvement due to water saturation is attributed to the combined effects of stress corrosion and frictional reduction.These findings carry significant implications for improving rock cuttability in mechanized excavation of hard rock formations.
基金Project(2022YFC2905100)supported by the National Key Research and Development Program of ChinaProject(52174098)supported by the National Natural Science Foundation of China。
文摘With the continuous expansion of deep underground engineering and the growing demand for safety monitoring,microseismic monitoring has become a core method for early warning of rock mass fracture and engineering stability assessment.To address problems in existing methods,such as low data processing efficiency and poor phase recognition accuracy under low signal-to-noise ratio(SNR)conditions in complex geological environments,this study proposes an intelligent phase picking model based on ResUNet.The model integrates the residual learning mechanism of ResNet with the multi-scale feature extraction capability of UNet,effectively mitigating the vanishing gradient problem in deep networks.It also achieves cross-layer fusion of shallow detail features and deep semantic features through skip connections in the encoder-decoder structure.Compared with traditional short-time average/long-time average(STA/LTA)algorithms and advanced neural network models such as PhaseNet and EQTransformer,ResUNet shows superior performance in picking P-and S-wave phases.The model was trained on 400000 labeled microseismic signals from the Stanford earthquake dataset(STEAD)and was successfully applied to the Shizhuyuan polymetallic mine in Hunan Province,China.The results demonstrate that ResUNet achieves high picking accuracy and robustness in complex geological conditions,offering reliable technical support for early warning of disasters such as rockburst in deep underground engineering.
文摘The availability of a tremendous amount of seismic data demands seismological researchers to analyze seismic phases efficiently.Recently,deep learning algorithms exhibit a powerful capability of detecting and picking on P-and S-wave phases.However,it remains a challenge to effeciently process enormous teleseismic phases,which are crucial to probe Earth’s interior structures and their dynamics.In this study,we propose a scheme to detect and pick teleseismic phases,such as seismic phase that reflects off the core-mantle boundary(i.e.,PcP)and that reflects off the inner-core boundary(i.e.,PKiKP),from a seismic dataset in Japan.The scheme consists of three steps:1)latent phase traces are truncated from the whole seismogram with theoretical arrival times;2)latent phases are recognized and evaluated by convolutional neural network(CNN)models;3)arrivals of good or fair phase are picked with another CNN models.The testing detection result on 7386 seismograms shows that the scheme recognizes 92.15%and 94.13%of PcP and PKiKP phases.The testing picking result has a mean absolute error of 0.0742 s and 0.0636 s for the PcP and PKiKP phases,respectively.These seismograms were processed in just 5 min for phase detection and picking,demonstrating the efficiency of the proposed scheme in automatic teleseismic phase analysis.
基金supported in part by the National Key Research and Development Program of China under Grant 2018YFA0702501in part by NSFC under Grant 41974126,41674116 and 42004101。
文摘Picking velocities from semblances manually is laborious and necessitates experience. Although various methods for automatic velocity picking have been developed, there remains a challenge in efficiently incorporating information from nearby gathers to ensure picked velocity aligns with seismic horizons while also improving picking accuracy. The conventional method of velocity picking from a semblance volume is computationally demanding, highlighting a need for a more efficient strategy. In this study, we introduce a novel method for automatic velocity picking based on multi-object tracking. This dynamic tracking process across different semblance panels can integrate information from nearby gathers effectively while maintaining computational efficiency. First, we employ accelerated density clustering on the velocity spectrum to discern cluster centers without the requirement for prior knowledge regarding the number of clusters. These cluster centers embody the maximum likelihood velocities of the main subsurface structures. Second, our proposed method tracks key points within the semblance volume. Kalman filter is adopted to adjust the tracking process, followed by interpolation on these tracked points to construct the final velocity model. Our synthetic data example demonstrates that our proposed algorithm can effectively rectify the picking errors of the clustering algorithm. We further compare the performances of the clustering method(CM), the proposed tracking method(TM), and the variational method(VM) on a field dataset from the Gulf of Mexico. The results attest that our method offers superior accuracy than CM, achieves comparable accuracy with VM, and benefits from a reduced computational cost.
基金financially supported by the National Key R&D Program of China(2018YFA0702504)the National Natural Science Foundation of China(42174152)+1 种基金the Strategic Cooperation Technology Projects of China National Petroleum Corporation(CNPC)and China University of Petroleum-Beijing(CUPB)(ZLZX2020-03)the R&D Department of China National Petroleum Corporation(2022DQ0604-01)。
文摘The picking efficiency of seismic first breaks(FBs)has been greatly accelerated by deep learning(DL)technology.However,the picking accuracy and efficiency of DL methods still face huge challenges in low signal-to-noise ratio(SNR)situations.To address this issue,we propose a regression approach to pick FBs based on bidirectional long short-term memory(Bi LSTM)neural network by learning the implicit Eikonal equation of 3D inhomogeneous media with rugged topography in the target region.We employ a regressive model that represents the relationships among the elevation of shots,offset and the elevation of receivers with their seismic traveltime to predict the unknown FBs,from common-shot gathers with sparsely distributed traces.Different from image segmentation methods which automatically extract image features and classify FBs from seismic data,the proposed method can learn the inner relationship between field geometry and FBs.In addition,the predicted results by the regressive model are continuous values of FBs rather than the discrete ones of the binary distribution.The picking results of synthetic data shows that the proposed method has low dependence on label data,and can obtain reliable and similar predicted results using two types of label data with large differences.The picking results of9380 shots for 3D seismic data generated by vibroseis indicate that the proposed method can still accurately predict FBs in low SNR data.The subsequent stacked profiles further illustrate the reliability and effectiveness of the proposed method.The results of model data and field seismic data demonstrate that the proposed regression method is a robust first-break picker with high potential for field application.
基金supported by the Major Research Plan on West-Pacific Earth System Multispheric Interactions (Nos.91858215,91958206)the National Natural Science Foundation of China (NSFC)Shiptime Sharing Project (No.41949581)the Key Research and Development Program of Shandong Province (No.2019GHY112019)。
文摘The precision and reliability of first-arrival picking are crucial for determining the accuracy of geological structure inversion using active source ocean bottom seismometer(OBS)refraction data.Traditional methods for first-arrival picking based on sample points are characterized by theoretical errors,especially in low-sampling-frequency OBS data because the travel time of seismic waves is not an integer multiple of the sampling interval.In this paper,a first-arrival picking method that utilizes the spatial waveform variation characteristics of active source OBS data is presented.First,the distribution law of theoretical error is examined;adjacent traces exhibit variation characteristics in their waveforms.Second,a label cross-correlation superposition method for extracting highfrequency signals is presented to enhance the first-arrival picking precision.Results from synthetic and field data verify that the proposed approach is robust,successfully overcomes the limitations of low sampling frequency,and achieves precise outcomes that are comparable with those of high-sampling-frequency data.
基金Project(52374153)supported by the National Natural Science Foundation of ChinaProject(2023zzts0726)supported by the Fundamental Research Funds for the Central Universities of Central South University,China。
文摘Conical picks are important tools for rock mechanical excavation.Mean cutting force(MCF)of conical pick determines the suitability of the target rock for mechanical excavation.Accurate evaluation of MCF is important for pick design and rock cutting.This study proposed hybrid methods composed of boosting trees and Bayesian optimization(BO)for accurate evaluation of MCF.220 datasets including uniaxial compression strength,tensile strength,tip angle(θ),attack angle,and cutting depth,were collected.Four boosting trees were developed based on the database to predict MCF.BO optimized the hyper-parameters of these boosting trees.Model evaluation suggested that the proposed hybrid models outperformed many commonly utilized machine learning models.The hybrid model composed of BO and categorical boosting(BO-CatBoost)was the best.Its outstanding performance was attributed to its advantages in dealing with categorical features(θincluded 6 types of angles and could be considered as categorical features).A graphical user interface was developed to facilitate the application of BO-CatBoost for the estimation of MCF.Moreover,the influences of the input parameters on the model and their relationship with MCF were analyzed.Whenθincreased from 80°to 90°,it had a significant contribution to the increase of MCF.
