Study of AI Based Methods for Characterization of Geotechnical Site Investigation Data (original) (raw)
Related papers
CRC Press eBooks, 2022
CPT soil profile interpretation represents a fundamental aspect for subsoil stratigraphic recon struction of complex geological contexts. In some situations, the soil profile may not exhibit evident boundary changes, making the interpretation more difficult. This crucial aspect plays a key role in the layers boundaries discontinuities identification and the construction of bi-dimensional and three-dimensional geotechnical models. In this paper, CPT and boreholes are used to calibrate and validate a massive and automated site characterization by combining statistical tools and artificial intelligence algorithms (AI). The procedure is applied in the complex stratigraphic context of Terre del Reno (Italy). The proposed data-driven analysis allows to combine the geological and geotechnical knowledge of the subsoil in an efficient and automatic way based on site-specific data, obtaining reliable and indispensable results for the construction of a robust and coherent geotechnical model of the subsoil.
Integrating Geo-Statistics and Geographic Information Systems in Geotechnical Soil Profiling
ERJ. Engineering Research Journal, 2015
Since it s evolution, Geotechnical engineering is considered one of the sciences that deeply based on uncertainties. These uncertainties are attributed to two facts. First fact, the formulas used in geotechnical engineering are empirical formulas and could be deeply affected by the nonhomogeneity of soil formation. The second fact is that always the soil is being characterized through a group of boreholes that are considered representative for the whole site, which could result in the missing of some anomalies in soil composition. Hence, geo-technical engineers are always looking for a technique, that can be used in dealing with these uncertainties. Several techniques were adopted starting from the sixties of the previous century like knowledge base expert systems, finite element analysis and finally geo-statistics. But the most sidelined of these techniques was geo-statistics due to several reasons. Hence, this paper tries to introduce geostatistics focusing on its applications in geotechnical engineering specially in soil profiling. It discusses the reason behind the fact that still the actual implementation of Geo-statistics in Geotechnical engineering is without real presence in the traditional work of any geotechnical engineer. Moreover, it introduces a look ahead towards actual implementation of Geo-statistics in soil profiling through the integration between one of the most mature geo-statistical software (Sgems) and the GIS software which is ARCGIS of ESRI
ERJ. Engineering Research Journal, 2022
Currently, there is a large boom in the construction of new developments, infrastructure, and transportation projects in the Middle East. Geotechnical engineers are responsible to characterize the subsurface ground conditions, obtain design parameters and identify problematic subsurface ground conditions. Geospatial Information Systems combined with statistical algorithms can provide an efficient way for identifying soil parameters/hazards. GIS associates data with their location (coordinates). The spatial data analysis platform ArcGIS is used to determine soil parameters at unsampled locations. An automated workflow is developed to apply different interpolation algorithms providing engineers with an easy-to-use tool to determine the most accurate algorithm for use in a specific project. This technique is applied to assess the liquefaction potential in a project site in Dubai, United Arab Emirates. Four spatial data analysis algorithms: Inverse Distance Weighted (IDW), Natural Neigh...
Bayesian Data Mining for a Generic Geotechnical Database
2018
This paper proposes a Bayesian data mining approach that searches a generic database for data points with soil characteristics similar to a set of site-specific data. A similarity index between the generic and site-specific data points is proposed based on the Bayesian analysis. The effectiveness of the proposed approach is illustrated by considering a generic clay database and a specific site in Sweden. The generic data points identified as “similar” can be combined with the limited site-specific data to construct a transformation model more relevant to a specific site.
Geophysical sensors combined with machine learning algorithms have been used to understand the pedosphere system, landscape processes and to model soil attributes. In this research, we used parent material, terrain attributes and data from geophysical sensors in different combinations, to test and compare different and novel machine learning algorithms to model soil attributes. Also, we analyzed the importance of pedoenvironmental variables in predictive models. For that, we collected soil physico-chemical and geophysical data (gamma-ray emission from uranium, thorium and potassium, magnetic susceptibility and apparent electric conductivity) by three sensors, gamma-ray spectrometer-RS 230, susceptibilimeter KT10-Terraplus and Conductivimeter-EM38 Geonics) at 75 points and, we performed soil analysis afterwards. The results showed varying models with the best performance (R 2 > 0.2) for clay, sand, Fe2O3, TiO2, SiO2 and Cation Exchange Capacity prediction. Modeling with selection of covariates at three phases (variance close to zero, removal by correction and removal by importance), demonstrated to be adequate to increase the parsimony. The prediction of soil attributes by machine learning algorithms demonstrated adequate values for field collected data, without any sample preparation, for most of the tested predictors (R 2 ranging from 0.20 to 0.50). Also, the use of four regression algorithms proved important, since at least one of the predictors used one of the tested algorithms. The performances of the best algorithms for each predictor were higher than the use of a mean value for the entire area comparing the values of Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The best combination of sensors that reached the best model performance to predict soil attributes were gamma-ray spectrometer and susceptibilimeter. The most important variables were parent material, digital elevation model, standardized height and magnetic susceptibility for most predictions. We concluded that soil attributes can be
Bayesian identification of soil stratigraphy based on soil behaviour type index
Canadian Geotechnical Journal, 2018
The cone penetration test (CPT) has been widely used to determine the soil stratigraphy (including the number N and thicknesses HN of soil layers) during geotechnical site investigation because it is rapid, repeatable, and economical. For this purpose, several deterministic and probabilistic approaches have been developed in the literature, but these approaches generally only give the “best” estimates (e.g., the most probable values) of N and HN based on CPT data according to prescribed soil stratification criteria, providing no information on the identification uncertainty (degrees-of-belief) in these “best” estimates. This paper develops a Bayesian framework for probabilistic soil stratification based on the profile of soil behaviour type index Ic calculated from CPT data. The proposed Bayesian framework not only provides the most probable values of N and HN, but also quantifies their associated identification uncertainty based on the Ic profile and prior knowledge. Equations are ...
Managing Risk in Geotechnical Engineering – From Data to Digitalization
Proceedings of the 7th International Symposium on Geotechnical Safety and Risk (ISGSR 2019), 2019
If you scan a page from a soil report, this is called digitization. If you deploy digital technologies, both software such as building information modeling and machine learning and hardware such as autonomous drones and additive manufacturing, to support new and more collaborative forms of project delivery, this is called digitalization. Data lies at the heart of this transformation that is targeted at re-valuing infrastructure from a "brick and mortar" asset to a service for the interests of the end-users. There is a need to view the value of data completely differently from how they are routinely used in current practice. In particular, there is a need to treat data as assets in themselves, over and above their conventional roles as inputs to a physical model or as monitoring data to trigger interventions. This paper explores the availability and nature of geotechnical data and presents two recent advances made in this direction for a specific but important task of estimating soil/rock properties (compressive sampling and Bayesian machine learning). Data-driven decision making does not imply taking the engineer out of the entire life cycle management chain. It is intended to support rather than to replace human judgment.
Probabilistic delineation of soil layers using Soil Behavior Type Index
CRC Press eBooks, 2022
CPTu-based soil profiling has become a key component in the geotechnical design process. However, this is an interpretative process, affected by the inherent variability of soil properties, measurement noise and subjective heuristics. These are difficult to communicate to other interpreters or, even for the same interpreter, to transfer across profiles. A semi-automated tool for CPTu data interpretation is presented as an aid in this interpretation process. A probabilistic-based algorithm is employed to elicit the implicit heuristics in CPTu-based soil profiling and facilitate transference. Univariate normal distributions fit Soil Behavior Type Index data. Soil class boundaries, taken from a conventionally accepted chart, are sequentially activated with user-specified refinement. Thin layers under cone resolution are merged using well-established criteria. An appli cation to CPTu records on finely interlayered deltaic deposits is illustrated, in which output delineations resulting from different analyst choices are compared among themselves and with one based on core description.
Bulletin of the Geological Society of Malaysia, 2002
Where good quality data have been obtained from careful supervision of subsurface exploration program, it is essential that the exact geological conditions be carefully analyzed. Without this it is impossible to check the design assumptions or to apply the results to a similar situation elsewhere. This paper presents the use of geological modelling for civil engineering projects. The modelling is useful for any layman involved in engineering to understand the geological conditions, thus hinder all the surprises during the construction stage. It also useful for understanding geological and deposition process for prediction of the history of the ground. However, modelling must be made precisely. Insufficient data must not be treated as sufficient.