Prediction of Disc Cutter Life During Shield Tunneling with AI via the Incorporation of a Genetic Algorithm into a GMDH-Type Neural Network

Disc cutter consumption is a critical problem that influences work performance during shield tunneling processes and directly affects the cutter change decision.This study proposes a new model to estimate the disc cutter life(Hf)by integrating a group method of data handling(GMDH)-type neural network(NN)with a genetic algorithm(GA).The efficiency and effectiveness of the GMDH network structure are optimized by the GA,which enables each neuron to search for its optimum connections set from the previous layer.With the proposed model,monitoring data including the shield performance database,disc cutter consumption,geological conditions,and operational parameters can be analyzed.To verify the performance of the proposed model,a case study in China is presented and a database is adopted to illustrate the excellence of the hybrid model.The results indicate that the hybrid model predicts disc cutter life with high accuracy.The sensitivity analysis reveals that the penetration rate(PR)has a significant influence on disc cutter life.The results of this study can be beneficial in both the planning and construction stages of shield tunneling.

Intelligent modelling of clay compressibility using hybrid meta-heuristic and machine learning algorithms

Compression index Ccis an essential parameter in geotechnical design for which the effectiveness of correlation is still a challenge.This paper suggests a novel modelling approach using machine learning(ML)technique.The performance of five commonly used machine learning(ML)algorithms,i.e.back-propagation neural network(BPNN),extreme learning machine(ELM),support vector machine(SVM),random forest(RF)and evolutionary polynomial regression(EPR)in predicting Cc is comprehensively investigated.A database with a total number of 311 datasets including three input variables,i.e.initial void ratio e0,liquid limit water content wL,plasticity index Ip,and one output variable Cc is first established.Genetic algorithm(GA)is used to optimize the hyper-parameters in five ML algorithms,and the average prediction error for the 10-fold cross-validation(CV)sets is set as thefitness function in the GA for enhancing the robustness of ML models.The results indicate that ML models outperform empirical prediction formulations with lower prediction error.RF yields the lowest error followed by BPNN,ELM,EPR and SVM.If the ranges of input variables in the database are large enough,BPNN and RF models are recommended to predict Cc.Furthermore,if the distribution of input variables is continuous,RF model is the best one.Otherwise,EPR model is recommended if the ranges of input variables are small.The predicted correlations between input and output variables using five ML models show great agreement with the physical explanation.

Micromechanical analysis of the behavior of stiffclay

Cementations formed in geological timescale are observed in various stiff clays.A micromechanical stress strain model is developed for modeling the effect of cementation on the deformation behavior of stiff clay.The proposed approach considers explicitly cementations at intercluster contacts,which is different from conventional model.The concept of inter-cluster bonding is introduced to account for an additional cohesion in shear sliding and a higher yield stress in normal compression.A damage law for inter-cluster bonding is proposed at cluster contacts for the debonding process during mechanical loading.The model is used to simulate numerous stress-path tests on Vallericca stiff clay.The applicability of the present model is evaluated through comparisons between the predicted and the measured results.In order to explain the stress-induced anisotropy arising from externally applied load,the evolution of local stresses and local strains at inter-cluster planes are discussed.

Durable response to pulsatile icotinib for central nervous system metastases from EGFR-mutated non-small cell lung cancer: A case report

BACKGROUND Central nervous system(CNS) metastases are a catastrophic complication of nonsmall cell lung cancer(NSCLC), including brain and leptomeningeal carcinomatosis, and are always accompanied by a poor prognosis. Despite the continuous development of existing treatments, the therapy of CNS metastases remains challenging.CASE SUMMARY We report a patient who was definitively diagnosed with brain and leptomeningeal metastases from NSCLC with a targeted mutation in epidermal growth factor receptor(EGFR). A standard dosage of icotinib(125 mg three times daily) was implemented but ineffective. CNS lesions developed despite stable systemic control, so pulsatile icotinib(1125 mg every 3 d) was administered. This new strategy for administration has lasted 25 mo so far, and resulted in complete remission of neurological symptoms, almost vanished lesions, and longer survival with no notable side effects.CONCLUSION This is the first successful example of pulsatile icotinib for treating isolated CNS progression from EGFR mutation-positive NSCLC, providing a new alternative for the local treatment of CNS metastases.

Enhanced elastic beam model with BADS integrated for settlement assessment of immersed tunnels

Excessive settlement may induce structural damage and water leakage in immersed tunnels,seriously threatening the tunnels’safety.However,making accurate assessment of the settlement in immersed tunnels is difficult due to the incomplete knowledge of the geotechnical parameters and the inadequacy of the model itself.This paper proposes an effective method to accurately assess the settlement in immersed tunnels.An enhanced beam on elastic foundation model(E-BEFM)is developed for the settlement assessment,with the Bayesian adaptive direct search algorithm adopted to estimate unknown model parameters based on previous observations.The proposed method is applied to a field case of the Hong Kong–Zhuhai–Macao immersed tunnel.The original BEFM is used for comparison to highlight the better assessment performance of E-BEFM,particularly for joints’differential settlement.Results show that the proposed method can provide accurate predictions of the total settlement,angular distortion(a representation of tubes’relatively differential settlement),and joints’differential settlement,which consequently supports the associated maintenance decision-making and potential risk prevention for immersed tunnels in service.

MODELLING STRAIN-RATE-DEPENDENCY OF NATURAL SOFT CLAYS COMBINED WITH ANISOTROPY AND DESTRUCTURATION

The paper aims to investigate modelling the strain-rate-dependency of natural soft clays combined with anisotropy and destructuration using an elasto-viscoplastic model. The model is based on Perzyna＇s overstress theory and the elastoplastic model S-CLAYIS. Tests at constant strain^rate and creep tests under both one-dimensional and triaxial conditions on several clays are simulated. Simulations highlight the loading scenarios in which it is necessary to account for anisotropy and/or destructuration in order to get accurate predictions. Comparisons between the predicted and measured results demonstrate that the proposed model can successfully reproduce the time-dependent behaviour of natural soft clays under different loading conditions.

Analysis of cement-treated clay behavior by micromechanical approach

Experimental results show the significant influence of cement content on the mechanical properties of cement-treated clays.Cementation is produced by mixing a certain amount of cement with the saturated clay.The purpose of this paper is to model the cementation effect on the mechanical behavior of cement-treated clay.A micromechanical stress-strain model is developed considering explicitly the cementation at inter-cluster contacts.The inter-cluster bonding and debonding during mechanical loading are introduced in two ways:an additional cohesion in the shear sliding and a higher yield stress in normal compression.The model is used to simulate isotropic compression and undrained triaxial tests under various confining stresses on cement-treated Ariake clay and Singapore clay with various cement contents.The applicability of the present model is evaluated through comparisons between numerical and experimental results.The evolution of local stresses and local strains in inter-cluster planes are discussed in order to explain the induced anisotropy due to debonding at contact level under the applied loads.The numerical simulations demonstrate that the proposed micromechanical approach is well adapted for taking into account the main physical properties of cement-treated clay,including damage and induced anisotropy under mechanical loading.

Large deformation analysis in geohazards and geotechnics

For geohazards and geotechnics,numerous problems involve large deformation,such as the installation of foundations(Jin YF et al.,2018a),landslides(Jin YF et al.,2020b),debris flow(Dai et al.,2017),collapse of underground structures(Zhang et al.,2019),and the formation of sinkholes(Baran-diaran Villegas,2018).Benefitting from the sustained development of computing power,numerical simulations have become useful analytical methods in geomechanics and related fields.

人工智能在岩土工程中的实践

岩土材料的复杂和不确定性致使传统理论在模拟和预测岩土工程问题经常显得无能为力。近年来,随着人工智能和大数据技术的快速发展,人工智能技术在岩土工程领域有了广泛应用,例如岩土参数的优化智能识别和预测、边坡变形的长期预测、基坑开挖过程变形的实时监测和预测以及盾构隧道的变形和盾构机刀盘参数的预测和更新等。为此,本专辑收集了在该研究领域具有代表性的研究成果,介绍了人工智能技术在岩土工程领域的进展和未来发展潜力,希望能帮助读者快速了解人工智能技术在岩土工程中的应用,以及推动岩土工程的智能化发展,为实现岩土工程智能化提供科学依据和技术支撑。

Influence of natural deposition plane orientation on oedometric consolidation behavior of three typical clays from southeast coast of China

The parameters obtained from oedometric consolidation tests are commonly used in the development of constitutive modeling and for engineering practice. This paper focuses on the influence of the natural deposition plane orientation on oedometric consolidation behavior of three natural clays from the southeast coast of China. Oedometer tests were conducted on intact specimens prepared by sampling at a series of angles relative to the natural deposition plane. For each specimen, yield stress,compressibility indexes, secondary compression, and permeability coefficients were determined. The influence of the sampling angle on these properties was investigated, revealing that yield stress, compression index, swelling index, creep index, ratio of secondary compression coefficient to compression index(Cae/Cc) and permeability coefficient were all dependent to some extent on the sampling angle. These findings indicate the role of the anisotropy due to the natural deposition on the oedometric consolidation behavior.

Physical model testing in geotechnical engineering

1Introduction Several characteristics of natural soils complicate the relationship between their mechanical behaviour and geotechnical construction and maintenance in the field.These characteristics include the presence of three phases(solid particle,water,and air),particle constitutions of various minerals(such as quartz,kaolinite,and montmorillonite),and an exceptionally wide range of particle size fromμm-scale(clay particles smallerthan2μm)to100-mm scale(suchas somegravelsandpebbles)。