Models to predict tunnel boring machines performance are reviewed and discussed.
•
Dimensional homogeneity is fulfilled using dimensional analysis.
•
Multi-gene genetic programming are used to develop equations for modeling.
•
The developed model gives 21.7% better results than the best existing model.
Abstract
An accurate prediction of tunnel boring machine (TBM) performance is one of the complex and crucial issues encountered frequently in tunnel construction, which is the aim of the present study. An improved methodology using dimensional analysis (DA) and multi-gene genetic programming (MGGP) is proposed to obtain a practical and accurate model which can predict TBM performance. Three dimensionless parameters are introduced by applying DA to predict TBM performance more efficiently. These parameters can represent TBM and rock features. The MGGP, as a powerful technique for developing a practical correlation model, was adopted to develop highly accurate models using GPTIPS (Genetic Programming Toolbox for the Identification of Physical Systems). A well-known database of a hard rock mechanized tunneling project of the Queens water conveyance tunnel was used to evaluate the performance of the proposed methodology. The performances of the developed models were examined and compared with other reported models using three statistical criteria. Regarding the sum of squared deviations (SSD), the developed model yielded 21.7% better results than the best existing model. Moreover, it was found that the presented dimensionless parameters have physical meaning and are much better parameters to develop a model for TBM performance prediction.
This section describes a technique called dimensional analysis and a data correlation method called multi-gene genetic programming.
Data collection and analysis
The present study applies the combination of the DA and MGGP approaches to consider multiple influencing characteristics of the ground in the predictive modeling of hard rock TBM performance. To this end, a well-known database collected and presented by Yagiz [19] from a hard rock mechanized tunneling project of the Queens Water Conveyance Tunnel # 3 (QWCT#3) is used. The QWCT#3, constructed between 1997 and 2000, was intended to improve fresh water distribution throughout New York, USA. The
Performing the DA
To predict the ROP (m/h) as the dependent variable, three intact rock properties [i.e., UCS (MPa), BTS (MPa), and BI (kN/mm)], two rock mass properties [i.e., (degree) and DPW (m)], and two TBM specifications [M(m) and (sec −1)] are considered as independent variables. To perform dimensional analysis of the TBM performance prediction problem, we apply the exponent method with the following steps:
In order to evaluate the performance of the developed models using MGGP compared to existing approaches, three criteria, i.e., the determination coefficient (R2), mean absolute percentage error (MAPE), and the sum of squared deviations (SSD), between the predicted values and measured data points are used as follows: where ROP, , and ROP are respectively the observed (measured) values, the mean of the observed
Results of the developed models using MGGP and practical verification
The fitness function of the MGGP is to minimize the sum of squared deviations (SSD) between the predicted values and observed data points. The SSD is known as an effective objective function to minimize the errors between field data and predicted results [e.g., [53], [54]]. The SSD can be defined as: where SSD denotes the dimensionless value of SSD, ROP is the observed value of ROP in the field conditions, ROP is the corresponding value predicted by the
Discussion
One of the most important steps in developing a model is the study into the effects of input parameters on outputs [57], [58]. Accordingly, this section discusses the effects of independent variables on the prediction of ROP. The effectiveness of the proposed dimensionless variables is also analyzed.
Fig. 12 presents the variation of the UCS and BTS along the studied tunnel. Clearly, no significant relationship can be observed between the UCS and BTS. This relationship is shown more clearly in
Summary and conclusions
The present study proposes an improved methodology based on dimensional analysis and a data correlation method for the first time to obtain a practical and accurate model for TBM performance prediction. In order to represent TBM and rock properties, three dimensionless parameters, including ROP, UCS, and BTS were introduced using dimensional analysis. The multi-gene genetic programming was adopted to optimize both parameters and structure of the prediction models. A reliable and
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
