Qian Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Qian Li has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Lanlan Xie
College of Environment and Civil Engineering, Chengdu University of Technology, Chengdu 610059, China
5 shared publications Jiang Jie
2 shared publicationsJie Jiang
2 shared publicationsJunping Li
2 shared publicationsLanlan Xie
2 shared publicationsChangchun Zhou
1 shared publication
Latest Publications
- Open Access
Regenerative Surface Memory and Model-Form Sensitivity in a Controlled Drillstring Vibration Hierarchy
Drillstring vibration response maps depend on regenerative cutting, friction, structural contact and the indices used to classify the response. This study examines that dependence through controlled comparisons within an M1–M4 numerical hierarchy. In M1, 39 of 96 regenerative cases receive the historical low-force-proxy label, compared with 14 that fall under a sampled no-engagement definition; one of the latter trajectories activates a numerical guard. The non-monotonic counts motivate closer examination of the discrete surface update. In matched 4 s M2 comparisons, velocity weakening increases the stick–slip index in all 32 designs, with a median difference of 0.01841 relative to the regularized Coulomb. The structural/contact calculations and reduced-fluid output mappings reveal the further sensitivity of the screening labels. Of 17 historical forward disagreements, all persist at nominal thresholds across the tested timesteps, while one persists across all 12 timestep–threshold settings. These comparisons distinguish sensitivity to continuous metrics from sensitivity to response labels and identify the numerical and classification conditions under which the results can be interpreted.
Open AccessRegenerative Surface Memory and Model-Form Sensitivity in a Controlled Drillstring Vibration Hierarchy
Drillstring vibration response maps depend on regenerative cutting, friction, structural contact and the indices used to classify the response. This study examines that dependence through controlled comparisons within an M1–M4 numerical hierarchy. In M1, 39 of 96 regenerative cases receive the historical low-force-proxy label, compared with 14 that fall under a sampled no-engagement definition; one of the latter trajectories activates a numerical guard. The non-monotonic counts motivate closer examination of the discrete surface update. In matched 4 s M2 comparisons, velocity weakening increases the stick–slip index in all 32 designs, with a median difference of 0.01841 relative to the regularized Coulomb. The structural/contact calculations and reduced-fluid output mappings reveal the further sensitivity of the screening labels. Of 17 historical forward disagreements, all persist at nominal thresholds across the tested timesteps, while one persists across all 12 timestep–threshold settings. These comparisons distinguish sensitivity to continuous metrics from sensitivity to response labels and identify the numerical and classification conditions under which the results can be interpreted.
- Open Access
Optimization of Lithology Identification and Formation Parameter Prediction Algorithms Based on Machine Learning
(Objective) To improve the accuracy and efficiency of lithology identification and formation parameter prediction during drilling, the study used drilling parameter data and implemented SVM, XGBoost, random forest, LightGBM and an ensemble of five algorithms for comparative experiments under baseline, noise interference (10% and 20% noise levels) and data imbalance scenarios. (Results) In lithology identification, the stacking ensemble achieved accuracies of 86.80% and 83.10% under 10% and 20% noise scenarios respectively. The accuracy of the random forest algorithm was 91.15% for the baseline scenario, and for the imbalanced scenarios (10% and 5%), the accuracies were 88.9% and 88.1%, respectively. In formation parameter prediction, random forest achieved mean absolute errors (MAEs) of 3.45, 0.0019, 0.0008 and 0.0004 for seismic velocity, pore pressure, fracture pressure and overburden pressure in the baseline scenario and performed best under noise and imbalanced data conditions. (Conclusions) An adaptive hybrid model was ultimately established: stacking ensemble is used for lithology prediction in noisy environments, while random forest is used for lithology prediction in non-noisy environments and for formation parameter prediction across all environments.
Open AccessOptimization of Lithology Identification and Formation Parameter Prediction Algorithms Based on Machine Learning
(Objective) To improve the accuracy and efficiency of lithology identification and formation parameter prediction during drilling, the study used drilling parameter data and implemented SVM, XGBoost, random forest, LightGBM and an ensemble of five algorithms for comparative experiments under baseline, noise interference (10% and 20% noise levels) and data imbalance scenarios. (Results) In lithology identification, the stacking ensemble achieved accuracies of 86.80% and 83.10% under 10% and 20% noise scenarios respectively. The accuracy of the random forest algorithm was 91.15% for the baseline scenario, and for the imbalanced scenarios (10% and 5%), the accuracies were 88.9% and 88.1%, respectively. In formation parameter prediction, random forest achieved mean absolute errors (MAEs) of 3.45, 0.0019, 0.0008 and 0.0004 for seismic velocity, pore pressure, fracture pressure and overburden pressure in the baseline scenario and performed best under noise and imbalanced data conditions. (Conclusions) An adaptive hybrid model was ultimately established: stacking ensemble is used for lithology prediction in noisy environments, while random forest is used for lithology prediction in non-noisy environments and for formation parameter prediction across all environments.
- Open Access
- 28 April 2026
- Published by Frontiers Media SA in Frontiers in Astronomy and Space Sciences
- Vol. 13, 1788224
Evaluation of lunar soil sampling disturbance based on penetration resistance method
Introduction: To investigate the impact of surface sampling operations on the physical and mechanical properties of lunar soil—defined as the granular, sub-centimeter-sized fraction of the lunar regolith—and to mitigate design errors in sampling missions that arise from neglecting such disturbance, we conducted sampling disturbance experiments using the CUG-1A lunar soil simulant. Methods: The degree of lunar soil disturbance was evaluated based on the change in penetration resistance of a standard probe inserted into the soil before and after sampling. Results: The experimental results indicate that the bucket width, sampling depth, sampling speed, and (for shovelling) the entry angle all significantly affect the disturbance degree. Discussion: The shovelling action, a linear penetration-and-lift operation, predominantly induces a compaction effect on the lunar soil simulant. In contrast, the digging action, an arcuate scooping operation, causes compaction during the initial phase but generates a distinct loosening effect at the motion endpoint. The parameter combination that minimizes disturbance for the shovelling experiment is a bucket width of 4 cm, sampling depth of 2 cm, speed of 50 mm/s, and entry angle of 45°. For the digging experiment, the optimal combination is a bucket width of 4 cm, sampling depth of 3.5 cm, and speed of 50 mm/s.
- 28 April 2026
- Published by Frontiers Media SA in Frontiers in Astronomy and Space Sciences
- Vol. 13, 1788224
Open AccessEvaluation of lunar soil sampling disturbance based on penetration resistance method
Introduction: To investigate the impact of surface sampling operations on the physical and mechanical properties of lunar soil—defined as the granular, sub-centimeter-sized fraction of the lunar regolith—and to mitigate design errors in sampling missions that arise from neglecting such disturbance, we conducted sampling disturbance experiments using the CUG-1A lunar soil simulant. Methods: The degree of lunar soil disturbance was evaluated based on the change in penetration resistance of a standard probe inserted into the soil before and after sampling. Results: The experimental results indicate that the bucket width, sampling depth, sampling speed, and (for shovelling) the entry angle all significantly affect the disturbance degree. Discussion: The shovelling action, a linear penetration-and-lift operation, predominantly induces a compaction effect on the lunar soil simulant. In contrast, the digging action, an arcuate scooping operation, causes compaction during the initial phase but generates a distinct loosening effect at the motion endpoint. The parameter combination that minimizes disturbance for the shovelling experiment is a bucket width of 4 cm, sampling depth of 2 cm, speed of 50 mm/s, and entry angle of 45°. For the digging experiment, the optimal combination is a bucket width of 4 cm, sampling depth of 3.5 cm, and speed of 50 mm/s.
- 1 January 2026
- Published by Editorial Office of Earth Science in Earth Science-Journal of China University of Geosciences
- Vol. 51 (8) , 3102-3117
基于最优相关性的钻速建模算法
《地球科学》(ISSN 1000-2383;CN42-1874/P) (原刊名《地球科学——中国地质大学学报》ISSN 1000-2383;CN42-1233/P),创刊于1957年,是由教育部主管、中国地质大学主办的综合性地学学术理论期刊,月刊,2015年改名为《地球科学》,她以反映中国地质大学教学与科研最新的、高水平的基础地质、应用地质、资源与环境地质及地学工程技术研究成果为主要任务。
- 1 January 2026
- Published by Editorial Office of Earth Science in Earth Science-Journal of China University of Geosciences
- Vol. 51 (8) , 3102-3117
基于最优相关性的钻速建模算法
《地球科学》(ISSN 1000-2383;CN42-1874/P) (原刊名《地球科学——中国地质大学学报》ISSN 1000-2383;CN42-1233/P),创刊于1957年,是由教育部主管、中国地质大学主办的综合性地学学术理论期刊,月刊,2015年改名为《地球科学》,她以反映中国地质大学教学与科研最新的、高水平的基础地质、应用地质、资源与环境地质及地学工程技术研究成果为主要任务。
- Open Access
Construction of Sampling Disturbance Model of Lunar Surface
This study establishes a dynamic evolution model of the physical and mechanical properties of lunar simulant as a function of sampling-induced disturbance on the lunar surface, aiming to eliminate design errors in sampling missions caused by neglecting the disturbance of lunar soil. A standard probe was inserted into the lunar soil simulant both before and after disturbance, and the variation in penetration resistance at the exact location was proposed as an indicator of the regolith’s disturbance state. Compression tests and disturbance tests were conducted on CUG-1A lunar soil simulant, with the experimental results subjected to regression analysis and neural network prediction. Based on the compression tests, a regression equation was derived relating the slope of the probe penetration resistance to the internal friction angle and density of the lunar soil simulant, showing a strong correlation between predicted and actual values. The disturbance tests provided penetration resistance curves under various disturbance conditions. By integrating these two components, a correspondence was established between the disturbance conditions and the internal friction angle and density of the lunar soil simulant. The predictive performance of three typical neural network algorithms—LM, BR, and SCG—with varying numbers of neurons was compared. The LM algorithm with 10 neurons was selected for its superior performance. Ultimately, a sampling disturbance model was developed to predict the internal friction angle and density of the lunar soil simulant based on disturbance conditions, demonstrating an extremely high correlation between predicted and actual values.
Open AccessConstruction of Sampling Disturbance Model of Lunar Surface
This study establishes a dynamic evolution model of the physical and mechanical properties of lunar simulant as a function of sampling-induced disturbance on the lunar surface, aiming to eliminate design errors in sampling missions caused by neglecting the disturbance of lunar soil. A standard probe was inserted into the lunar soil simulant both before and after disturbance, and the variation in penetration resistance at the exact location was proposed as an indicator of the regolith’s disturbance state. Compression tests and disturbance tests were conducted on CUG-1A lunar soil simulant, with the experimental results subjected to regression analysis and neural network prediction. Based on the compression tests, a regression equation was derived relating the slope of the probe penetration resistance to the internal friction angle and density of the lunar soil simulant, showing a strong correlation between predicted and actual values. The disturbance tests provided penetration resistance curves under various disturbance conditions. By integrating these two components, a correspondence was established between the disturbance conditions and the internal friction angle and density of the lunar soil simulant. The predictive performance of three typical neural network algorithms—LM, BR, and SCG—with varying numbers of neurons was compared. The LM algorithm with 10 neurons was selected for its superior performance. Ultimately, a sampling disturbance model was developed to predict the internal friction angle and density of the lunar soil simulant based on disturbance conditions, demonstrating an extremely high correlation between predicted and actual values.
- Open Access
- 25 June 2025
- Published by Springer Nature in Journal of Petroleum Exploration and Production Technology
- Vol. 15 (7) , 1-17
Prediction and dynamic optimization of drilling performance based on the combination of mineral composition and operational factors
- 25 June 2025
- Published by Springer Nature in Journal of Petroleum Exploration and Production Technology
- Vol. 15 (7) , 1-17
Open AccessPrediction and dynamic optimization of drilling performance based on the combination of mineral composition and operational factors
- Open Access
Simplified Particle Models and Properties Analysis Designed for DEM Lunar Soil Simulants
The discrete element method (DEM) is one of the most popular methods for simulating lunar soil simulants due to the lack of real lunar soil. To reduce the computational consumption and difficulty because of complex particle models, simplified particle models, in which a single particle consists of two, four, or six elements, are discussed in this paper. Three steps, including random generation, particle replacement, and sedimentation, can generate the proposed simulant. The relationship between the mechanical properties of the simulant and microscopic parameters defined in DEM was analyzed by the orthogonal array testing (OATS) technique. Then, the prediction functions, which can calculate mechanical properties from inputting the microscopic parameters without carrying out the DEM, are also established by a back-propagation artificial neural network (BP-ANN). The widely used physical simulants JSC-1 from the USA and FJS-1 from Japan are simulated in DEM from the prediction function with high accuracy.
Open AccessSimplified Particle Models and Properties Analysis Designed for DEM Lunar Soil Simulants
The discrete element method (DEM) is one of the most popular methods for simulating lunar soil simulants due to the lack of real lunar soil. To reduce the computational consumption and difficulty because of complex particle models, simplified particle models, in which a single particle consists of two, four, or six elements, are discussed in this paper. Three steps, including random generation, particle replacement, and sedimentation, can generate the proposed simulant. The relationship between the mechanical properties of the simulant and microscopic parameters defined in DEM was analyzed by the orthogonal array testing (OATS) technique. Then, the prediction functions, which can calculate mechanical properties from inputting the microscopic parameters without carrying out the DEM, are also established by a back-propagation artificial neural network (BP-ANN). The widely used physical simulants JSC-1 from the USA and FJS-1 from Japan are simulated in DEM from the prediction function with high accuracy.
- Open Access
Application of Response Surface Methodology in Lunar Deep Rock Layer Drilling Experiments in Low-Temperature and Anhydrous Environments
The Chang’e project has completed a sampling mission of the shallow lunar soil layer; however, the exploration of the deep lunar rock layer remains unaddressed. To further investigate the feasibility of deep lunar rock drilling and identify the factors affecting the rate of penetration (ROP) and power in low-temperature, H2O less environment, a model was developed. This study utilized the Box–Behnken method to design a response surface experiment, where the number of polycrystalline diamond compact (PDC) cutters, the backward inclination angle, the chip removal conditions, and the temperature were considered as the key influencing factors. A response surface model for ROP and power was established. The results indicated that the number of PDC cutters, the backward inclination angle, the chip removal conditions, and the temperature significantly affected both ROP and power, with the interaction between the temperature and the backward inclination angle having a particularly strong impact on the ROP. The regression model demonstrated high predictive accuracy for both ROP and power, with goodness of fit (R2) values of 0.95 and 0.96, respectively. The optimal combination of the backward inclination angle, number of PDC cutters, temperature, and chip removal conditions, derived from the response surface experiment, was 25°, four, −15 °C, and 1, respectively, which resulted in high drilling efficiency and low power consumption. This study offers new insights for the design of deep lunar drilling experiments, as well as support for the future optimization of drilling tools.
Open AccessApplication of Response Surface Methodology in Lunar Deep Rock Layer Drilling Experiments in Low-Temperature and Anhydrous Environments
The Chang’e project has completed a sampling mission of the shallow lunar soil layer; however, the exploration of the deep lunar rock layer remains unaddressed. To further investigate the feasibility of deep lunar rock drilling and identify the factors affecting the rate of penetration (ROP) and power in low-temperature, H2O less environment, a model was developed. This study utilized the Box–Behnken method to design a response surface experiment, where the number of polycrystalline diamond compact (PDC) cutters, the backward inclination angle, the chip removal conditions, and the temperature were considered as the key influencing factors. A response surface model for ROP and power was established. The results indicated that the number of PDC cutters, the backward inclination angle, the chip removal conditions, and the temperature significantly affected both ROP and power, with the interaction between the temperature and the backward inclination angle having a particularly strong impact on the ROP. The regression model demonstrated high predictive accuracy for both ROP and power, with goodness of fit (R2) values of 0.95 and 0.96, respectively. The optimal combination of the backward inclination angle, number of PDC cutters, temperature, and chip removal conditions, derived from the response surface experiment, was 25°, four, −15 °C, and 1, respectively, which resulted in high drilling efficiency and low power consumption. This study offers new insights for the design of deep lunar drilling experiments, as well as support for the future optimization of drilling tools.
- 19 August 2024
- Published by Elsevier in Geoenergy Science and Engineering
- Vol. 242
Research on adaptive feature optimization and drilling rate prediction based on real-time data
- 19 August 2024
- Published by Elsevier in Geoenergy Science and Engineering
- Vol. 242
Research on adaptive feature optimization and drilling rate prediction based on real-time data