Ibrahim Ocak

Calculation of surface settlements caused by EPBM tunneling using artificial neural network, SVM, and Gaussian processes

Increasing demand on infrastructures increases attention to shallow soft ground tunneling methods in urbanized areas. Especially in metro tunnel excavations, due to their large diameters, it is important to control the surface settlements observed before and after excavation, which may cause damage to surface structures. In order to solve this problem, earth pressure balance machines (EPBM) and slurry balance machines have been widely used throughout the world. There are numerous empirical, analytical, and numerical analysis methods that can be used to predict surface settlements. But substantially fewer approaches have been developed for artificial neural network-based prediction methods especially in EPBM tunneling. In this study, 18 different parameters have been collected by municipal authorities from field studies pertaining to EPBM operation factors, tunnel geometric properties, and ground properties. The data source has a preprocess phase for the selection of the most effective parameters for surface settlement prediction. This paper focuses on surface settlement prediction using three different methods: artificial neural network (ANN), support vector machines (SVM), and Gaussian processes (GP). The success of the study has decreased the error rate to 13, 12.8, and 9, respectively, which is relatively better than contemporary research.

Estimation of Elastic Modulus of Intact Rocks by Artificial Neural Network

The modulus of elasticity of intact rock (Ei) is an important rock property that is used as an input parameter in the design stage of engineering projects such as dams, slopes, foundations, tunnel constructions and mining excavations. However, it is sometimes difficult to determine the modulus of elasticity in laboratory tests because high-quality cores are required. For this reason, various methods for predicting Ei have been popular research topics in recently published literature. In this study, the relationships between the uniaxial compressive strength, unit weight (γ) and Ei for different types of rocks were analyzed, employing an artificial neural network and 195 data obtained from laboratory tests carried out on cores obtained from drilling holes within the area of three metro lines in Istanbul, Turkey. Software was developed in Java language using Weka class libraries for the study. To determine the prediction capacity of the proposed technique, the root-mean-square error and the root relative squared error indices were calculated as 0.191 and 92.587, respectively. Both coefficients indicate that the prediction capacity of the study is high for practical use.

A new approach for estimating the transverse surface settlement curve for twin tunnels in shallow and soft soils

Increasing demand on infrastructures has led to increased attention to shallow soft ground tunneling methods in urbanized areas. Especially in metro tunnel excavations, it is important to control the surface settlements which are observed before and after excavation, which may cause damage to surface structures. Unlike motorway, sewage and other infrastructure tunnels, metro tunnels generally have to be excavated as twin tunnels and must have a larger diameter. Metro tunnels also have shallow depth. Due to their shallow depth, metro tunnels generally have been constructed in weak rocks or weak soils in cities. The construction of twin tunnels will generate ground movements which have the potential to cause damage to existing surface and subsurface structures. To solve this settlement problem, experts have used the Earth pressure balance machine (EPBM) and the slurry balance machine. In such excavations, especially in twin tunnels, the main challenges for constructers are estimating the maximum surface settlement, controlling the interaction of transverse surface settlement and shaping the settlement curve. Incorrect estimation of these parameters can lead to significant problems above the tunnels and in nearby structures. This paper focuses on surface settlement measurements, on the interaction of twin tunnel transverse surface settlement and on the relationship between shield parameters and transverse surface settlement for parallel tunnels using EPBM shields in clay and sand soils in shallow depth. Also, a new equation is proposed for estimating the transverse settlement curve of twin tunnels. The results from this proposed equation are compared with the results of field observations. The transverse settlement curve values obtained from the proposed equation have good agreement with the actual results for the Otogar–Kirazli metro case studies.

Performance prediction of roadheaders using ensemble machine learning techniques

Mechanical excavators are widely used in mining, tunneling and civil engineering projects. There are several types of mechanical excavators, such as a roadheader, tunnel boring machine and impact hammer. This is because these tools can bring productivity to the project quickly, accurately and safely. Among these, roadheaders have some advantages like selective mining, mobility, less over excavation, minimal ground disturbances, elimination of blast vibration, reduced ventilation requirements and initial investment cost. A critical issue in successful roadheader application is the ability to evaluate and predict the machine performance named instantaneous (net) cutting rate. Although there are several prediction methods in the literature, for the prediction of roadheader performance, only a few of them have been developed via artificial neural network techniques. In this study, for this purpose, 333 data sets including uniaxial compressive strength and power on cutting boom, 103 data set including RQD, and 125 data sets including machine weight are accumulated from the literature. This paper focuses on roadheader performance prediction using six different machine learning algorithms and a combination of various machine learning algorithms via ensemble techniques. Algorithms are ZeroR, random forest (RF), Gaussian process, linear regression, logistic regression and multi-layer perceptron (MLP). As a result, MLP and RF give better results than the other algorithms also the best solution achieved was bagging technique on RF and principle component analysis (PCA). The best success rate obtained in this study is 90.2% successful prediction, and it is relatively better than contemporary research.

Discussion on article “Surface subsidence induced by twin subway tunneling in soft ground conditions in Istanbul” by Yılmaz Mahmutoğlu, Bulletin of Engineering Geology and the Environment, 2011 70(1):115–131

I have read the paper with interest, and would like to congratulate the author for providing valuable information on the EPBM tunneling, which is a highly complicated subject and much discussed over the years. Generally, result of this study—both in terms of surface settlement and environmental effect—is in agreement with former studies published with regard to same project (Ocak 2009; Ocak and Bilgin 2009; Ercelebi et al. 2011). But I would like to submit some minor contributions to paper. Physical gap value which is named as tail void in the literature, have accepted different from other authors by Mahmutoglu (Ercelebi et al. 2011; Talmon and Bezuijen 2009). Tail void is accepted as distance between segment outer diameter and external shield diameter by the Mahmutoglu. It is clear that; tail void isn’t distance between segment outer diameter and external shield diameter. As shown from the literature above the tail void is the space between the segment outer diameter and the excavation diameter. Tail void is given as 150 and 220 mm for Lovat and Herrenknecht, respectively, by Mahmutoglu. Whereas according to literature, tail void from excavation diameter—external shield diameter is 6,564 mm 6,300 mm = 264 mm for lovat and 6,500 mm 6,300 mm = 200 mm for Herrenknecht. Additionally, with accepting of tunnel diameter as 6.500 and 6.564 m (Ocak 2009; Ocak and Bilgin 2009), some minor changes will occur in Table 3 and Table 4 like Z0/D ratio. It is a fact that, face pressure applied by the EPBM, as one of the important surface settlement data, plays an important role on settlement process in EPBM excavations. Therefore, using of face pressure on an EPBM tunneling paper associated with surface settlement is vital. According to this project specification, face pressure haven’t to less than 250 kPa. Various face pressure rates were applied each step of EPBM excavation for this reason on this project. Author said ‘‘in view of the shallow overburden, the face pressure was limited’’ on the paper. Face pressure influence on surface settlements should have studied more detailed on the paper. Also, Formulas used for obtaining material constant (K) (Table 4) should have given. Author have used together building measurement points (BMP) and surface measurement points (SMP) as SMP (Table 3 and Table 4). In my opinion, BMP don’t reflect surface settlements accurately. Because, BMP were installed on building 4–5 m above the surface. Therefore, BMP is different than from SMP. Therefore, using of both BMP and SMP are controversial for same purposes. Another difference Mahmutoglu’s paper from literature (Ocak and Bilgin 2009) is about average tunnel advancement. Average tunnel advancement is given as 10 m/day by Mahmutoglu. The overall performances have examined detailed by Ocak and Bilgin 2009. The overall performances of Herrenknecht and Lovat are founded as; the best daily advance rates 25.2 and 23.8 m/day, the best weakly advance rates 102.2 and 118.5 m/day, the best monthly advance rates 415.9 and 418.6 m/day, the mean daily advance (including waiting due to excessive deformations) 5.6 and 5.4 m/day, the mean daily advance (excluded waiting) is 11.1 and 11.3 m/day, respectively, (Ocak and Bilgin 2009). Also, there is a difference between Mahmutoglu’s paper and the literature (Ocak 2009) about increasing of project cost due to surface settlements. This issue was studied in detail by Ocak (2009). This project awarded the contract for construction 225.36 million USD by the Istanbul I. Ocak (&) Mining Engineering Department, Istanbul University, Istanbul, Avcılar, Turkey e-mail: iocak@istanbul.edu.tr