Advance prediction method for rock mass stability of tunnel boring based on deep neural network of time series

Geological layers excavated using tunnel boring machines are buried deeply and sampled difficultly, and the geological behavior exhibits high diversity and complexity. Excavating in uncertain geology conditions bears the risks of excessive damage to the equipment and facing geologic hazards. Many scholars have used various signals to predict the advance geology conditions, but accurate prediction of these conditions in real-time and without effecting operations has not been realized yet. In this article, based on a large amount of corresponding data, an advance prediction model of the rock mass category (RMC) is formulated. First, the problem is divided into two parts, which are modeled separately to reduce the complexity of design and training. Then, the two models are combined in a pre-trained model, which is retrained to as the final prediction model to avoid the problem of error accumulation. The final model can predict the advance RMC in real-time and without affecting operations. The accuracy of the prediction model reaches 99% at an advance time of 60 min. The advance RMC can be used to guide the selection of support modes and control parameters without additional detection equipment and excavation down-time.

Assessment of rate of penetration of a tunnel boring machine in the longest railway tunnel of Turkey

One of the most important issues in tunnels to be constructed with tunnel boring machines (TBMs) is to predict the excavation time. Excavation time directly affects tunnel costs and feasibility. For this reason, studies on the prediction of TBM performance have always been interesting for tunnel engineers. Therefore, the purpose of the study is to develop models to predict the rate of penetration (ROP) of TBMs. In accordance with the purpose of the study, a new database including 5334 cases is obtained from the longest railway tunnel of Turkey. Each case includes uniaxial compressive strength, Cerchar Abrasivity Index, α angle, weathering degree and water conditions as input or independent variables. Two multiple regression models and two ANN models are developed in the study. The performances of the ANN models are considerably better than those of the multiple regression equations. Before deep tunnel construction in a metamorphic rock medium, the ANN models developed in the study are reliable and can be used. In contrast, the performances of the multiple regression equations are promising, but they predict lower ROP values than the measured ROP values. Consequently, the prediction models for ROP are open to development depending on the new data and new prediction algorithms.

A perspective of pipejacking works by tunnel boring machines in Hong Kong: part II – applications, problems encountered, cost and prospects

In this paper, application of special techniques such as pipe roof construction, retractable TBM, and rectangular TBM to address site difficulties and constraints is highlighted. Cost and prospects of TBM pipejacking works are discussed. Some observations, including problems encountered and overcome, are also highlighted. This paper should be read together with part I, which covers development, selection of design, operations and types of techniques.

A perspective of pipejacking works by tunnel boring machines in Hong Kong: part I – development, selection of design, operations and types of techniques

In this paper, the history of pipejacking works by tunnel boring machines in Hong Kong, their development, selection of design method and tunnel boring machine (TBM), operations, length and alignment of a pipeline, types of techniques to suit different site constraints and ground conditions, and performance of the works are discussed. This paper should be read together with part II, which covers applications, problems encountered, cost and prospects.

MULTIDISCIPLINARY DESIGN OPTIMIZATION OF A MOBILE MINER USING THE OPENMDAO PLATFORM

This paper proposes an optimization framework based on the OpenMDAO software library intended for engineer-to-order products and applies it to the conceptual design of a Mobile Miner. A Mobile Miner is a complex machine and a flexible alternative to Tunnel Boring Machines for small-scale tunneling and mining applications. The proposed framework is intended for use in early design and quotation stages with the objective to get fast estimates of important product characteristics, such as excavation rate and cutter lifetime. The ability to respond fast to customer requests is vital when offering customized products for specific applications and thereby to stay competitive on the global market. This is true for most engineer-to-order products and especially for mining equipment where each construction project is unique with different tunnel geometries and rock properties. The presented framework is applied to a specific use-case where the design of the miner's cutter wheel is in focus and a set of Pareto optimal designs are obtained. Furthermore, the framework extends the capabilities of OpenMDAO by including support for mixed-variable formulations and it supports an exploratory approach to design optimization.