The accurate prediction of the strength of rocks after high-temperature treatment is important for the safety maintenance of rock in deep underground engineering.Five machine learning(ML)techniques were adopted in thi...The accurate prediction of the strength of rocks after high-temperature treatment is important for the safety maintenance of rock in deep underground engineering.Five machine learning(ML)techniques were adopted in this study,i.e.back propagation neural network(BPNN),AdaBoost-based classification and regression tree(AdaBoost-CART),support vector machine(SVM),K-nearest neighbor(KNN),and radial basis function neural network(RBFNN).A total of 351 data points with seven input parameters(i.e.diameter and height of specimen,density,temperature,confining pressure,crack damage stress and elastic modulus)and one output parameter(triaxial compressive strength)were utilized.The root mean square error(RMSE),mean absolute error(MAE)and correlation coefficient(R)were used to evaluate the prediction performance of the five ML models.The results demonstrated that the BPNN shows a better prediction performance than the other models with RMSE,MAE and R values on the testing dataset of 15.4 MPa,11.03 MPa and 0.9921,respectively.The results indicated that the ML techniques are effective for accurately predicting the triaxial compressive strength of rocks after different high-temperature treatments.展开更多
Experimental data taken from free-field soil in 1-g shaking table tests are valuable for seismic studies on soil-structure interaction.But the available data from medium-to large-scale shaking table tests were not abu...Experimental data taken from free-field soil in 1-g shaking table tests are valuable for seismic studies on soil-structure interaction.But the available data from medium-to large-scale shaking table tests were not abundant enough to cover a large variety of types and conditions of the soil.In the study,1-g shaking table tests of a 3-m-height sand column were conducted to provide seismic experimental data about sand.The sand was directly collected in-situ,with the largest grain diameter being 2 cm and containing a water content of 6.3%.Properties of the sand were estimated under the influence of white noise plus pulse and earthquake motions,including the settlement,the dynamic properties of the sand column,and the three soil layers′shear modulus degradation relationships.The estimated properties were then indirectly verified by means of finite element analysis.Results show that the estimated parameters were effective and could be used in numerical modeling to reproduce approximate seismic responses of the sand column.展开更多
This paper studies the microstructure variation induced by super-absorbent polymer(SAP)to understand the mechanism of macroscopic strength improvement of stabilized soil.The fabric changes of cement elime stabilized s...This paper studies the microstructure variation induced by super-absorbent polymer(SAP)to understand the mechanism of macroscopic strength improvement of stabilized soil.The fabric changes of cement elime stabilized soil were analyzed with respect to the variation of SAP content,water content,lime content and curing time,using mercury intrusion porosimetry(MIP)tests.It can be observed that the delimitation pore diameter between inter-and intra-aggregate pores was 0.2 mm for the studied soil,determined through the intrusion/extrusion cycles.Experimental results showed that fabric in both inter-and intra-aggregate pores varied significantly with SAP content,lime content,water content and curing time.Two main changes in fabric due to SAP are identified as:(1)an increase in intra-aggregate pores(<0.2 mm)due to the closer soilecementelime cluster space at higher SAP content;and(2)a decrease in inter-aggregate pores represented by a reduction in small-pores(0.2e2 mm)due to the lower pore volume of soil mixture after water absorption by SAP,and a slight increase in large-pores(>2 mm)due to the shrinkage of SAP particle during the freezeedry process of MIP test.Accordingly,the strength gain due to SAP for cementelime stabilized soil was mainly due to a denser fabric with less interaggregate pores.The cementitious products gradually developed over time,leading to an increase in intra-aggregate pores with an increasing proportion of micro-pores(0.006e0.2 mm).Meanwhile,the inter-aggregate pores were filled by cementitious products,resulting in a decrease in total void ratio.Hence,the strength development over time is attributable to the enhancement of cementation bonding and the refinement of fabric due to the increasing cementitious compounds.展开更多
基金We acknowledge the funding support from the National Natural Science Foundation of China(Grant No.51778575)Postdoctoral Science Foundation of China(Grant No.2021M692481)Fundamental Research Funds for the Central Universities of China(Grant No.2042021kf0055).The authors would like to thank the anonymous reviewers and editors for their constructive suggestions which greatly improve the quality of this paper.The authors are also grateful for the permission from Elsevier.
文摘The accurate prediction of the strength of rocks after high-temperature treatment is important for the safety maintenance of rock in deep underground engineering.Five machine learning(ML)techniques were adopted in this study,i.e.back propagation neural network(BPNN),AdaBoost-based classification and regression tree(AdaBoost-CART),support vector machine(SVM),K-nearest neighbor(KNN),and radial basis function neural network(RBFNN).A total of 351 data points with seven input parameters(i.e.diameter and height of specimen,density,temperature,confining pressure,crack damage stress and elastic modulus)and one output parameter(triaxial compressive strength)were utilized.The root mean square error(RMSE),mean absolute error(MAE)and correlation coefficient(R)were used to evaluate the prediction performance of the five ML models.The results demonstrated that the BPNN shows a better prediction performance than the other models with RMSE,MAE and R values on the testing dataset of 15.4 MPa,11.03 MPa and 0.9921,respectively.The results indicated that the ML techniques are effective for accurately predicting the triaxial compressive strength of rocks after different high-temperature treatments.
基金Supported by:National Natural Science Foundation of China under Grant Nos.52008233 and U1839201the National Key Research and Development Program of China under Grant No.2018YFC1504305the Innovative Research Groups of the National Natural Science Foundation of China under Grant No.51421005。
文摘Experimental data taken from free-field soil in 1-g shaking table tests are valuable for seismic studies on soil-structure interaction.But the available data from medium-to large-scale shaking table tests were not abundant enough to cover a large variety of types and conditions of the soil.In the study,1-g shaking table tests of a 3-m-height sand column were conducted to provide seismic experimental data about sand.The sand was directly collected in-situ,with the largest grain diameter being 2 cm and containing a water content of 6.3%.Properties of the sand were estimated under the influence of white noise plus pulse and earthquake motions,including the settlement,the dynamic properties of the sand column,and the three soil layers′shear modulus degradation relationships.The estimated properties were then indirectly verified by means of finite element analysis.Results show that the estimated parameters were effective and could be used in numerical modeling to reproduce approximate seismic responses of the sand column.
基金Project(2018YFC1802204)supported by the National Key R&D Program of ChinaProject(51634010)supported by the Key Project of National Natural Science Foundation of ChinaProject(2018SK2026)supported by the Key R&D Program of Hunan Province,China。
基金the China Postdoctoral Science Foundation(Grant Nos.2016M600396 and 2017T100355)the Fundamental Research Funds for the Central Universities of China(Grant No.B200204001)Jiangsu Natural Resources Science and Technology Fund(Grant No.KJXM2019025)are also acknowledged.
文摘This paper studies the microstructure variation induced by super-absorbent polymer(SAP)to understand the mechanism of macroscopic strength improvement of stabilized soil.The fabric changes of cement elime stabilized soil were analyzed with respect to the variation of SAP content,water content,lime content and curing time,using mercury intrusion porosimetry(MIP)tests.It can be observed that the delimitation pore diameter between inter-and intra-aggregate pores was 0.2 mm for the studied soil,determined through the intrusion/extrusion cycles.Experimental results showed that fabric in both inter-and intra-aggregate pores varied significantly with SAP content,lime content,water content and curing time.Two main changes in fabric due to SAP are identified as:(1)an increase in intra-aggregate pores(<0.2 mm)due to the closer soilecementelime cluster space at higher SAP content;and(2)a decrease in inter-aggregate pores represented by a reduction in small-pores(0.2e2 mm)due to the lower pore volume of soil mixture after water absorption by SAP,and a slight increase in large-pores(>2 mm)due to the shrinkage of SAP particle during the freezeedry process of MIP test.Accordingly,the strength gain due to SAP for cementelime stabilized soil was mainly due to a denser fabric with less interaggregate pores.The cementitious products gradually developed over time,leading to an increase in intra-aggregate pores with an increasing proportion of micro-pores(0.006e0.2 mm).Meanwhile,the inter-aggregate pores were filled by cementitious products,resulting in a decrease in total void ratio.Hence,the strength development over time is attributable to the enhancement of cementation bonding and the refinement of fabric due to the increasing cementitious compounds.