Jet grouting: using artificial neural networks to predict soilcrete column diameter - part II. (1st January 2015)

• Record Type: Journal Article
• Authors: Nikbakhtan, B.; Apel, D.; Ahangari, K.
• Is Part Of: International journal of mining and mineral engineering. Volume 6:Number 1(2015)
• Page Start: 57
• Page End: 71
• Publication Date: 2015-01-01
• Subjects: jet grouting -- ANNs -- artificial neural networks -- soilcrete column diameter -- soil conductivity -- grout density -- optimisation; Mining engineering -- Periodicals; Mines and mineral resources -- Periodicals; 622.05
• DOI: 10.1504/IJMME.2015.067951
• Languages: English
• ISSNs: 1754-890X

Abstract

Previously authors attempted to predict soilcrete diameter using Sum of Squared–Deviations mathematical model (Part I). It has been shown that the mathematical model cannot relate all impacting parameters with the diameter where the relation between soil conductivity and grout density did not match with literature and field observations. Therefore, in this paper an approach based on artificial neural network (ANN) with a wider range of data are used to calculate the diameter. ANN is a useful predictive method because it utilises both extensive computerised database and existing knowledge of what influences the diameter. This paper attempts to evaluate potential as well as limitations of ANN for predicting the diameter and to develop optimal neural network models to reduce the need for trial jet grouting as much as possible. One of the most significant results of this study is the optimisation of the costs and time needed for mining and civil projects.