Real-time prediction of the rock mass class in front of the tunnel face is essential for the adaptive adjustment of tunnel boring machines(TBMs).During the TBM tunnelling process,a large number of operation data are g...Real-time prediction of the rock mass class in front of the tunnel face is essential for the adaptive adjustment of tunnel boring machines(TBMs).During the TBM tunnelling process,a large number of operation data are generated,reflecting the interaction between the TBM system and surrounding rock,and these data can be used to evaluate the rock mass quality.This study proposed a stacking ensemble classifier for the real-time prediction of the rock mass classification using TBM operation data.Based on the Songhua River water conveyance project,a total of 7538 TBM tunnelling cycles and the corresponding rock mass classes are obtained after data preprocessing.Then,through the tree-based feature selection method,10 key TBM operation parameters are selected,and the mean values of the 10 selected features in the stable phase after removing outliers are calculated as the inputs of classifiers.The preprocessed data are randomly divided into the training set(90%)and test set(10%)using simple random sampling.Besides stacking ensemble classifier,seven individual classifiers are established as the comparison.These classifiers include support vector machine(SVM),k-nearest neighbors(KNN),random forest(RF),gradient boosting decision tree(GBDT),decision tree(DT),logistic regression(LR)and multilayer perceptron(MLP),where the hyper-parameters of each classifier are optimised using the grid search method.The prediction results show that the stacking ensemble classifier has a better performance than individual classifiers,and it shows a more powerful learning and generalisation ability for small and imbalanced samples.Additionally,a relative balance training set is obtained by the synthetic minority oversampling technique(SMOTE),and the influence of sample imbalance on the prediction performance is discussed.展开更多
A new optimization method for the optimization of stacking of composite glass fiber laminates is developed. The fiber orientation and angle of the layers of the cylindrical shells are sought considering the buckling l...A new optimization method for the optimization of stacking of composite glass fiber laminates is developed. The fiber orientation and angle of the layers of the cylindrical shells are sought considering the buckling load. The proposed optimization algorithm applies both finite element analysis and the mode-pursuing sampling （MPS）method. The algorithms suggest the optimal stacking sequence for achieving the maximal buckling load. The procedure is implemented by integrating ANSYS and MATLAB. The stacking sequence designing for the symmetric angle-ply three-layered and five-layered composite cylinder shells is presented to illustrate the optimization process, respectively. Compared with the genetic algorithms, the proposed optimization method is much faster and efficient for composite staking sequence plan.展开更多
Due to their important biological role as markers for different pathologies, sialic acid （SA） analyses are important for clinical research. In this work, a miniaturized capillary electrophoresis with amperometrie de...Due to their important biological role as markers for different pathologies, sialic acid （SA） analyses are important for clinical research. In this work, a miniaturized capillary electrophoresis with amperometrie detection （mini-CE-AD） was developed for the determination of N-aeetylneuraminic acid （NANA）, which is the most widespread form of SAs. NANA was first oxidized by periodic acid in an acidic solution, and then the oxidation product β-formyl pyruvic acid was derivatized with electroactive 2-thiobarbituric acid （TBA） to form an electroactive NANA-TBA adduct, which could be readily determined by mini-CE-AD. The limit of detection （LOD） of NANA-TBA could achieve 0.50 μg/mL （1.6 μmol·L-1, S/N=3） based on an online enrichment approach of moving chemical reaction boundary. The proposed method was successfully applied to the analysis of NANA in human saliva, and the recoveries were in the range of 91.8%-109% with RSDs of 1.8%-3.9%. Due to its simple design and construction, low cost and portability, the mini-CE-AD device will possess more practicability in more field work as an alternative to conventional and microchip CE approaches.展开更多
基金funded by the National Natural Science Foundation of China(Grant No.41941019)the State Key Laboratory of Hydroscience and Engineering(Grant No.2019-KY-03)。
文摘Real-time prediction of the rock mass class in front of the tunnel face is essential for the adaptive adjustment of tunnel boring machines(TBMs).During the TBM tunnelling process,a large number of operation data are generated,reflecting the interaction between the TBM system and surrounding rock,and these data can be used to evaluate the rock mass quality.This study proposed a stacking ensemble classifier for the real-time prediction of the rock mass classification using TBM operation data.Based on the Songhua River water conveyance project,a total of 7538 TBM tunnelling cycles and the corresponding rock mass classes are obtained after data preprocessing.Then,through the tree-based feature selection method,10 key TBM operation parameters are selected,and the mean values of the 10 selected features in the stable phase after removing outliers are calculated as the inputs of classifiers.The preprocessed data are randomly divided into the training set(90%)and test set(10%)using simple random sampling.Besides stacking ensemble classifier,seven individual classifiers are established as the comparison.These classifiers include support vector machine(SVM),k-nearest neighbors(KNN),random forest(RF),gradient boosting decision tree(GBDT),decision tree(DT),logistic regression(LR)and multilayer perceptron(MLP),where the hyper-parameters of each classifier are optimised using the grid search method.The prediction results show that the stacking ensemble classifier has a better performance than individual classifiers,and it shows a more powerful learning and generalisation ability for small and imbalanced samples.Additionally,a relative balance training set is obtained by the synthetic minority oversampling technique(SMOTE),and the influence of sample imbalance on the prediction performance is discussed.
基金Innovation Team Development Program of Ministry of Education of China (No. IRT0763)National Natural Science Foundation of China (No. 50205028).
文摘A new optimization method for the optimization of stacking of composite glass fiber laminates is developed. The fiber orientation and angle of the layers of the cylindrical shells are sought considering the buckling load. The proposed optimization algorithm applies both finite element analysis and the mode-pursuing sampling （MPS）method. The algorithms suggest the optimal stacking sequence for achieving the maximal buckling load. The procedure is implemented by integrating ANSYS and MATLAB. The stacking sequence designing for the symmetric angle-ply three-layered and five-layered composite cylinder shells is presented to illustrate the optimization process, respectively. Compared with the genetic algorithms, the proposed optimization method is much faster and efficient for composite staking sequence plan.
基金This work was supported by the National Natural Science Foundation of China （No. 21205042）, the Daxia Foundation of East China Normal University （No. 2015DX-284） and the Students Innovative Experimental Project of Shanghai Municipality （No. 201610269070）. The authors have declared no conflicts of interest.
文摘Due to their important biological role as markers for different pathologies, sialic acid （SA） analyses are important for clinical research. In this work, a miniaturized capillary electrophoresis with amperometrie detection （mini-CE-AD） was developed for the determination of N-aeetylneuraminic acid （NANA）, which is the most widespread form of SAs. NANA was first oxidized by periodic acid in an acidic solution, and then the oxidation product β-formyl pyruvic acid was derivatized with electroactive 2-thiobarbituric acid （TBA） to form an electroactive NANA-TBA adduct, which could be readily determined by mini-CE-AD. The limit of detection （LOD） of NANA-TBA could achieve 0.50 μg/mL （1.6 μmol·L-1, S/N=3） based on an online enrichment approach of moving chemical reaction boundary. The proposed method was successfully applied to the analysis of NANA in human saliva, and the recoveries were in the range of 91.8%-109% with RSDs of 1.8%-3.9%. Due to its simple design and construction, low cost and portability, the mini-CE-AD device will possess more practicability in more field work as an alternative to conventional and microchip CE approaches.
