Automatic classification of fine-grained soils using CPT measurements and Artificial Neural Networks

Abstract

Soil classification is a means of grouping soils into categories according to a shared set of properties or characteristics that will exhibit similar engineering behaviour under loading. Correctly classifying site conditions is an important, costly, and time-consuming process which needs to be carried out at every building site prior to the commencement of construction or the design of foundation systems. This paper presents a means of automating classification for fine-grained soils, using a feed-forward ANN (Artificial Neural Networks) and CPT (Cone Penetration Test) measurements. Thus representing a significant saving of both time and money streamlining the construction process. 216 pairs of laboratory results and CPT tests were gathered from five locations across Northern Croatia and were used to train, test, and validate the ANN models. The resultant Neural Networks were saved and were subjected to a further external verification using CPT data from the Veliki vrh landslide. A test site, which the model had not previously been exposed to. The neural network approach proved extremely adept at predicting both ESCS (European Soil Classification System) and USCS (Unified Soil Classification System) soil classifications, correctly classifying almost 90% of soils. While the soils that were incorrectly classified were only partially misclassified. The model was compared to a previously published model, which was compiled using accepted industry standard soil parameter correlations and was shown to be a substantial improvement, in terms of correlation coefficient, absolute average error, and the accuracy of soil classification according to both USCS and ESCS guidelines. The study confirms the functional link between CPT results, the percentage of fine particles FC, the liquid limit w L and the plasticity index I P. As the training database grows in size, the approach should make soil classification cheaper, faster and less labour intensive.

Machine learning and soil sciences: a review aided by machine learning tools

“…In soil sciences, ML algorithms are usually trained using the traditional train-validation split or cross-validation (Keskin et al, 2019;Liang et al, 2019), or even no validation (Feng et al, 2019), except for some studies based on DL or with an engineering background (e.g. Reale et al, 2018), including some of our publications on the use of DL for DSM (Padarian et al, 2019c) or soil spectroscopy (Padarian et al, 2019b, a), which use a train-validation-test split. Considering the increasing size of datasets, we think soil scientists should transition towards the implementation of some DL practices such as dataset split and hyper-parameter optimisation (Bergstra and Bengio, 2012;Snoek et al, 2012), not only for NN, but also for any algorithm that has hyper-parameters.…”

Machine learning and soil sciences: A review aided by machine learning tools

“…In soil sciences, ML algorithms are usually trained using the traditional train/validation split or cross-validation (Keskin et al, 2019;Liang et al, 2019), or even no validation (Feng et al, 2019), except for some studies based on DL or with engineering background (e.g. Reale et al (2018)), including some of our publications on the use of DL for DSM (Padarian et al, 2019c) or soil spectroscopy (Padarian et al, 2019b, a), which use a train/validation/test split. Considering the increasing size of datasets, we think soil scientist should transition towards the implementation of some DL practices such as dataset split 5 and hyper-parameter optimisation (Bergstra and Bengio, 2012;Snoek et al, 2012), not only for NNs but for any algorithm that has hyper-parameters.…”

Horizontal and Vertical Geotechnical Variations of Soils According to USCS Classification for the City of An-Najaf, Iraq Using GIS

“…The Unified Soil Classification System is first proposed by Casagrande in 1942 and developed in 1952 by the Army Corps of Engineers (Das and Sobhan 2013). It is widely used in many building codes and books (Reale et al 2018;Robertson 2016). The soil in this classification system is divided into two master divisions: coarse soil (gravel and sand) and fine soil (clay and silt).…”

“…Fine-grained soils are furthermore classified into clay or silt using a hydrometer test. Finally, soils are subclassified according to their consistency (Reale et al 2018).…”