Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Deformation and stress characteristics of existing twin tunnels induced by close-distance EPBS under-crossing

Effect of ground surface surcharge on deformational performance of tunnel in spatially variable soil

Fragility analysis of gray iron pipelines subjected to tunneling induced ground settlement

Various wastes have gained attention nowaday as they are widely used in stabilization of expansive soils. The construction–demolition waste has not been widely used in soil stabilization, and its use is still under research. A lot of construction–demolition waste is being produced from old dismantled buildings whose disposal is a major concern due to limited availability of land space. This paper presents a laboratory investigation using consolidated drained triaxial test in order to study the stress–strain and volumetric behavior, shear strength parameters, strength ratio at various strain levels and stiffness of un-stabilized as well as crushed construction–demolition waste (CCDW) stabilized soil. The test results revealed that the addition of CCDW in un-stabilized soil has improved its stress–strain and volumetric behavior remarkably, at all confining pressures. The volumetric strain has been found to be decreased with the addition of CCDW. The inclusion of CCDW in un-stabilized soil has decreased the cohesion value to a limited extent and increased the angle of shearing resistance significantly. An increase in strength ratio as well as the stiffness of un-stabilized soil has also been improved when it is stabilized with CCDW. In the end, a practical numerical example has been solved in order to make the readers understand the significance of the improvement in un-stabilized soil after incorporating CCDW into it. This mixture of CCDW-stabilized expansive soil can be used in construction of low-volume traffic roads and embankment stability.

Effect of addition of construction–demolition waste on strength characteristics of high plastic clays

Responses of pipeline to side-by-side twin tunnelling at different depths: 3D centrifuge tests and numerical modelling

The shear strength parameters for geotechnical designs are obtained mainly from consolidated drained (CD) or consolidated undrained (CU) triaxial tests. However, during construction, the excess pore-air pressure generally dissipates instantaneously while the excess pore-water pressure dissipates with time. This condition needs to be simulated in a constant water content (CW) triaxial test. The study on Yunnan red clay is carried out to investigate the soil-water characteristics and the shear strength characteristics under the constant water content condition. Osmotic technique is used to obtain the soil-water characteristic curve. A series of CW triaxial tests are conducted on statically compacted specimens. The experimental results show that the soil-water characteristic curve has a low air entry value of 7 kPa due to large pores in non-uniform pore size distribution, and a high residual value exceeding 10 MPa. In addition, the initial degree of saturation and net confining stress play an important role in affecting the shear characteristics under the constant water content condition. Finally, a new semi-empirical shear strength model in terms of degree of saturation is proposed and then applied to Yunnan red clay. Simulation result shows that the model is capable of capturing some key features of soils. The model can be used in whole engineering practice range, covering both unsaturated and saturated soils.

Soil-water characteristics and shear strength in constant water content triaxial tests on Yunnan red clay

Study on the stability of shield tunnel face in clay and clay-gravel stratum through large-scale physical model tests with transparent soil

The tunnel-pipeline-soil interaction is attracting more and more wide range of scholars and was treated as a plane strain problem, with little attention given to the responses of existing pipeline to twin-tunneling process with different construction sequences. Moreover, there are rare reports available on the influences of tunnel construction sequences on an existing pipeline. In this study, a three-dimensional centrifuge model test and numerical simulations using an advanced hypoplasticity constitutive model were conducted to investigate the influences of twin stacked tunneling with different construction sequences on an existing pipeline. The results shows that the maximum values of greenfield ground settlement, pipeline settlement and additional pipeline bending strain due to twin stacked tunneling when the upper tunnel was excavated first were 25%, 18% and 11% larger, respectively, than those when the lower tunnel was excavated first. The maximum values of greenfield ground settlement, pipeline settlement and additional pipeline bending strain due to twin stacked tunneling when the lower tunnel was excavated first were slightly smaller than those when the twin tunnels were excavated simultaneously. After the twin-tunneling process, the sagging regions were located within ± 1.5DT (tunnel diameter) in these three conditions. The maximum pipeline bending strain within the hogging regions was approximately half of that within the sagging regions. Because of shielding effects from the upper tunnel, the region where vertical stress of the soil reduced was transferred to both sides of the tunnel due to the second tunneling process when the upper tunnel was excavated first. As a result, the extended sagging regions due to the second tunneling process when the upper tunnel was excavated first were two times than that when the lower tunnel was excavated first.

Shaokun Ma

Papers

Responses of pipeline to side-by-side twin tunnelling at different depths: 3D centrifuge tests and numerical modelling

Soil-water characteristics and shear strength in constant water content triaxial tests on Yunnan red clay

Study on the stability of shield tunnel face in clay and clay-gravel stratum through large-scale physical model tests with transparent soil

Settlement and Load Transfer Mechanism of Pipeline Due to Twin Stacked Tunneling with Different Construction Sequences

Effects of construction sequence of double basement excavations on an existing floating pile