Performance prediction of tunnel boring machine is one of the geotechnical problems that commonly have complexity and ambiguity. Over the last few years, many researchers have made attempt to set up accurate models for predicting TBM performance prediction. This issue is crucial because a precise estimation of machine performance can considerably decrease the capital costs of mechanical excavation project. Performance prediction of TBM strictly relies on the estimation of the rate of penetration (ROP), defined as the ratio of excavated distance to the operating time during continuous excavation phase, and advance rate (AR), the ratio of both mined and supported actual distance to the total time. Many attempts were made for the development of the accurate prediction models [1–10]. In addition to these models in recent years some prediction models have been developed using artificial intelligences including artificial neural network (ANN), fuzzy logic and neuro-fuzzy [11–23]. Taking into consideration the nature of problem, the main purpose of the present study is to develop a model by utilizing the ANN for predicting TBM performance. In order to achieve this aim, a database composed of rock mass properties such as fabric indices of four rock mass classification and the angle between plane of weakness and tunnel axis, intact rock properties including uniaxial compressive strength, machine specification including net thrust per cutter together with actual measured TBM penetration rate, was compiled along the 6.5 km bored Alborz service tunnel.
Alborz service tunnel is the longest tunnel section (6.5 km) along Tehran-Shomal Freeway, situated in the high elevation portions of Alborz Mountain Range, connecting the capital city of Tehran to the Caspian Sea in the North (Fig.1). The service tunnel with diameter of 5.20 m was excavated by an open gripper TBM in advance of two main tunnel tubes to be excavated subsequently. The purpose of the service tunnel is site investigation, drainage of the rock mass, providing access for main tunnel excavations and service, ventilation and drainage during operation of the complete tunnel system.
Prediction of TBM penetration rate is the most crucial issue to estimate machine performance and contains a large number of influencing parameters, in general, four main categories including rock material and rock mass parameters, machine characteristics and operational parameters given as follows:
Intact rock strength (e.g. uniaxial compressive strength "UCS", Brazilian tensile strength "BTS", Point load index "Is (50)")
Toughness (Punch penetration index, Fracture toughness index)
Hardness and drillability (Siever's J-value, Total & Taber hardness index, Schmidt hammer hardness)
Brittleness (Swedish brittleness number "S20", brittleness indices; B1= σc / σt and B2= [(σc – σt) / (σc + σt)]), where σc and σt are uniaxial compressive and tensile strength of intact rock, respectively
Abrasiveness indices (Cerchar Abrasivity Index "CAI", Abrasion Value "AV")
Others (Poisson ratio "ν", Elasticity modulus "E", Internal friction angle "φ", Porosity, Grain size etc.)
