Presently the survey of absolute displacements of targets fixed at the tunnel wall is the state of the art in performance monitoring of tunnels. Monitoring data are used to assess the stabilization process of the tunnel, more recently also for the short term prediction of the ground conditions ahead of the face.
The displacements do not substantially vary in rock mass conditions with nearly constant properties and influencing factors. On the other hand, changes in the rock mass structure or properties result in changes in the displacement characteristics.
For nearly constant conditions, the displacement trends show minor fluctuations within a certain normal range, due to minor variations in the rock mass properties. Deviations from this range are clear indicators for changing conditions ahead of the face or outside of the tunnel. To identify such trend deviations the normal range of the trend lines becomes crucial. Geostatistical methods allow an automatic identification of trends along the tunnel. Using data from completed tunnel projects a “reference trend table” can be established. By comparing actual observed trend characteristics to this reference table changing ground conditions ahead of the face can be identified and hence, the change in the displacement characteristics and magnitudes can be predicted.
The uncertainties in the geological conditions and ground parameters require an observational approach for safe and economical tunnel construction. Several conditions must be fulfilled for a successful application of the observational method (Peck 1969, Schubert 2008). One of the requirements is the assessment of possible behaviors and the establishment of their acceptable limits during design. This includes the identification of potential failure modes, as well as the determination of deformation characteristics and magnitudes. As the ground in general is all but homogeneous, continuous, and isotropic, simple homogeneous models usually do not provide enough insight to establish a realistic “normal behavior” for structured and heterogeneous ground conditions. It is also unrealistic to think that sophisticated numerical models can be used for an entire project during the design. A reasonable way to produce expected realistic ground behaviors is to first use simplified models to determine the range of expected displacements, and then modify the results with the help of expert knowledge.
During construction the measurement results contain all influences of the ground structure, stresses, and interaction between ground and support. The previously established characteristic behaviors for certain conditions are compared to the monitoring results. In case of agreement it can be established that the observed behavior is “normal”. Deviations from the expected behavior can have various reasons. One may be that the behavior during design was not assessed correctly. In this case, a refinement of the model is required. Another reason for behavior deviating from the expected can be a change in the ground conditions ahead of the face. It is meanwhile well known that trends of displacement vector orientations can be used to predict changing ground conditions ahead of the face (Schubert & Budil 1995, Steindorfer 1997, Jeon et al. 2005).
In many circumstances, our fundamental understanding of soil and rock behavior still falls short of being able to predict how the ground will behave. Cause-wise analysis of mine accidents reveals that roof falls continue to remain the single largest killer. Ground control operation is an ‘imprecise’ area of engineering due to the fact that we are dealing with a material produced by nature (the ground). Under these circumstances, expert judgement plays an important role, and empirical approaches to design are widely used. Thus, such accidents can be obviated using the accurate measurement, optimization and analysis of data a predictions based on previous results using one of the Artificial Intelligence technique i.e. Neural Networking. It is a simple computational model, which is analogous to that of neural system in human brain.
In this paper we have given a brief study on Neural Network Technology including Back Propagation Neural Network (BPNN) to train the network for optimization the mine support parameters. Some of the variable parameters associated with the underground excavation work have been taken as input/output parameter for the network. The technique of simulation of the result has also been discussed.
Safely exploitation of coal has been a big problem since years. In terms of the method of winning coal, the share of opencast mining, which was as low as 14% in 1951, increased to current high level of above 80% whereas the share of underground mining declined from 77% in 1971 to current 20%. Even if, we can’t ignore the underground mine coal production due to its good quality of coal as well as for societal reasons. In underground operation ground control problem is an important factor affecting safety, production and efficiency .A view of underground mines with sufficient support and drilling operation have been shown in fig.1.
(Figure in full paper)
In terms number of mines, out of about 595 operating mines, about 384 are underground mines. In underground coal mining technology, bord and pillar mining method is one of the major technology being used in India, with about 91% of the underground coal production, employing about 57% of total work force. As per statistics of accident data “fall of roof / sides” is one of the major cause of mine accidents. A major consideration in supporting mine roofs is limiting the movement and expansion of the rock strata immediately above the roof. Cause-wise analysis of mine accidents reveals that roof falls continue to remain the single largest killer, As many as 61% of the incidences, which is 28.5% of total fatalities are due to roof fall. Such accidents can be obviated using the accurate measurement and optimization of data and its analysis using Artificial Intelligence. Since artificial intelligence (AI) techniques can make use of heuristic knowledge (rules of thumb) or pattern matching techniques, as opposed to solving a set of mathematical equations, they should be ideally suited for application in the field of geotechnical engineering. Many aspects of mine design are based upon empirical data.