• Journal of Korean Tunnelling and Underground Space Association
• /
• v.11 no.2
• /
• pp.151-162
• /
• 2009

This study proposes a new methodology for quantitative rock classification in unsampled rock zone, which occupies the most of tunnel design area. This methodology is to train an ANN (artificial neural network) by using results from a drilling investigation combined with electric resistivity survey in sampled zone, and then apply the trained ANN to making a prediction of grade of rock classification in unsampled zone. The prediction is made at the center point of a shifting window by using a number of electric resistivity values within the window as input reference information. The ANN training in this study was carried out by the RPROP (Resilient backpropagation) training algorithm and Early-Stopping method for achieving a generalized training. The proposed methodology is then applied to generate a rock grade distribution on a real tunnel site where drilling investigation and resistivity survey were undertaken. The result from the ANN based prediction is compared with one from a conventional kriging method. In the comparison, the proposed ANN method shows a better agreement with the electric resistivity distribution obtained by field survey. And it is also seen that the proposed method produces a more realistic and more understandable rock grade distribution.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2003.06b
• /
• pp.141-152
• /
• 2003

터널 설계를 위한 조사 있어서, 요사이 시추공 조사는 물론 탄성파 탐사, 전기 비저항 탐사 등의 물리탐사가 빈번히 행해지고 있는 실정이다. 따라서 최적의 지반평가(암반 등급 등)를 위해 조사에서 얻어지는 모든 자료를 체계적으로 최대한 활용할 수 있는 방법이 절실히 요구되고 있다. 많은 연구자들이 정량적 데이터가 부족한 경우에 대처하기 위해 정상적 데이터의 이용을 적극 제안해 왔다. 본 연구에서는 신뢰도가 다른 두 종류의 자료, 즉 시추공자료와 물리탐사 자료를 활용하여 시추가 되지 않은 구간의 암반등급을 추정하는 방법을 지구통계학적 이론에 근거하여 소개하고자 한다.

• Tunnel and Underground Space
• /
• v.16 no.2 s.61
• /
• pp.135-145
• /
• 2006

In this study, conditional simulation was conducted to estimate rock mass rating(RMR) in unsurveyed regions. Sequential Gaussian simulation(SGS) and sequential indicator simulation(SIS) were applied for estimating RMR from the bore hole logging data. The uncertainty of SGS and SIS was verified by sample cross validation. A subset composed of 5 bore hole logging data among the original 30 bore hole logging data was set aside as test data. The remainder was training data. The quality of SGS and SIS estimation on the testing data reflects how well it would perform in an unsupervised setting. SGS and SIS were useful stochastic methods to estimate the spatial distribution of rock mass classes correctly and assess the uncertainty of estimation quantitatively. The result of conditional simulation can offer useful information of rock mass classes such as RMR in unsurveyed regions.

• Journal of the Korean GEO-environmental Society
• /
• v.20 no.3
• /
• pp.15-20
• /
• 2019

A large amount of site investigation data is essential to obtain reliable design value. However, site investigations are generally insufficient due to economic problems. It is important to estimate the ground profile information in unboring region for accurate earthwork-volume prediction, and such ground profile information can be estimated by using the geo-statistical approach. Furthermore, the ground profile information in unboring region can be estimated by training a model via machine learning technique such as artificial neural network. In this paper, artificial neural network-based model estimated the ground profile information in unboring region, and this results were compared with that of ordinary kriging technique, which is referred to the geo-statistical approach. Accordingly, a total of 84 ground profile information in an actual bridge environment was split into 75 training and 9 test databases. The observed ground profile information of the test database was compared with those of the ordinary kriging technique and artificial neural network.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.6 no.2
• /
• pp.101-111
• /
• 2004

In order to guarantee the stability of a tunnel and its optimum design, it is very important to obtain enough ground investigation data. In realty, however, it is not the case due to the limitation of measuring spatially distributed data and economical reasons. Especially, there are regions where drilling is impossible due to civil appeal and mountainous topology, and it is also difficult to estimate rock mass classes quantitatively with only geophysical exploration data. In this study, therefore, 3 dimensional multiple indicator kriging (3D-MI kriging), which can incorporate geophysical exploration data and drill core data off a tunnel center line, is proposed to cope with such problems. To this end, two dimensional mutiple indicator kriging, which is one of the geostatistical techniques, is extended for three dimensional analysis. Also, the proposed 3D-MI kriging was applied to determine the rock mass classes by RMR system for the design of a Kyungbu express rail way tunnel.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.6
• /
• pp.867-878
• /
• 2018

As the era of AVs (Automated Vehicles) comes to a close, many researches related to AVs have been conducted. Up until now, research on traffic flow impact of AVs has been the main topic, and research on traffic management for AVs is still in beginning stage. This study analyzed the effect of Dynamic Merge Control (DMC) in manual vehicle (MV) and AV environment at work zone. Dynamic Late Merge (DLM) and DLM with Dynamic Early Merge (DEM) are compared by simulation. Simulation results showed that DLM improves travel time and work zone throughput compared to no merge control case in both MV and AV environment. In the case of additional operation of DEM, the improvement effect was not observed in MV environment, but it was improved in AV environment. As a result, DMC operation in AV environment was as effective as the improvement in transition from MV to AV environment. Therefore congestion reduction at freeway work zone by DMC will be possible in future AV environment, and the improvement of DMC can be suggested.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2005.09a
• /
• pp.37-42
• /
• 2005

Remote seismic tomography is regarded as one of the most valuable geophysical technique for the estimation of the rock mass classification in the tunnel area where hard data information such as drill logs are absent. But the results of rock mass classification based on the remote seismic tomography tend to be overestimated in practice. In this study, we propose the effective method to implement the seismic reference velocity ratio based on semblance for the improvement of rock mass classification. Also, to verify its feasibility, proposed technique was tested by using the real field data.

• Geophysics and Geophysical Exploration
• /
• v.11 no.4
• /
• pp.350-360
• /
• 2008

Electrical resistivity is one of physical property of the earth and measured by electrical resistivity survey, electrical resistivity logging and laboratory test. Recently, electrical resistivity is widely used in determination of rock quality in support pattern design of road and railway tunnel construction sites. To get more reliable rock quality data from electrical resistivity, it needs a lot of test and study on correlation of resistivity and rock quality. Firstly, we did rock property test in laboratory, such as P wave velocity, Young's modulus, uniaxial compressive strength (UCS) and electrical resistivity. We correlate each test results and we found out that electrical resistivity has highly related to P wave velocity, Young's modulus and UCS. Next, we accomplished electrical resistivity survey in field site and carried out electrical resistivity logging at in-situ area. We also performed rock classification, such as RQD, RMR and Q-system and we correlate electrical resistivity to RMR data. We found out that electrical resistivity logging data are highly correlate to RMR. Also we found out that electrical resistivity survey data are lower than electrical resistivity logging data when there are faults or fractures. And it cause electrical resistivity survey data to lowly correlate to RMR.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.14 no.5
• /
• pp.503-516
• /
• 2012

Rock mass rating (RMR) is the key factor when designing the appropriate support pattern of tunnel projects. Borehole drilling is usually performed along the tunnel route in order to determine the rock mass rating to be used for tunnel design. The rock mass rating at the non-boring region between boreholes is usually assessed through geophysical surveys such as electrical prospecting, seismic prospecting, etc. Many studies were carried out to find out the correlation between electrical resistivity and rock mass rating. However, most researches were aimed at obtaining the relationship between the two parameters utilizing experimental results obtained from laboratory tests or electrical prospectings. In this paper, efforts were made to analyze and obtain relationships between the electrical resistivity obtained from in-situ electrical resistivity logging data and the rock mass rating. Correlation studies using field data showed that the electrical resistivity is highly correlated with the rock mass rating with the determination coefficient more than 90%. The correlation analysis was also carried out between RMR classification parameters and the electrical resistivity. It was shown that the correlation between the condition of discontinuities and the electrical resistivity was very high with the determination coefficient more than 80%; that between the groundwater condition and the electrical resistivity was very low with the determination coefficient less than 57%.

• The Journal of Engineering Geology
• /
• v.32 no.2
• /
• pp.221-239
• /
• 2022

Unknown geotechnical characteristics are key challenges in the design of piles for the plant, civil and building works. Although the N-values which were read through the standard penetration test are important, those N-values of the whole area are not likely acquired in common practice. In this study, the N-value is predicted by means of regression analysis with artificial intelligence (AI). Big data is important to improve learning performance of AI, so circular augmentation method is applied to build up the big data at the current study. The optimal model was chosen among applied AI algorithms, such as artificial neural network, decision tree and auto machine learning. To select optimal model among the above three AI algorithms is to minimize the margin of error. To evaluate the method, actual data and predicted data of six performed projects in Poland, Indonesia and Malaysia were compared. As a result of this study, the AI prediction of this method is proven to be reliable. Therefore, it is realized that the geotechnical characteristics of non-boring points were predictable and the optimal arrangement of structure could be achieved utilizing three dimensional N-value distribution map.