• Tunnel and Underground Space
• /
• v.31 no.1
• /
• pp.25-40
• /
• 2021

As the era of Industry 4.0 arrives, the researches using artificial intelligence in the field of rock engineering as well have increased. For a better understanding and availability of AI, this paper analyzed the types of algorithms and how to apply them to the research papers where AI is applied among domestic and international studies related to tunnels, blasting and mines that are major objects in which rock engineering techniques are applied. The analysis results show that the main specific fields in which AI is applied are rock mass classification and prediction of TBM advance rate as well as geological condition ahead of TBM in a tunnel field, prediction of fragmentation and flyrock in a blasting field, and the evaluation of subsidence risk in abandoned mines. Of various AI algorithms, an artificial neural network is overwhelmingly applied among investigated fields. To enhance the credibility and accuracy of a study result, an accurate and thorough understanding on AI algorithms that a researcher wants to use is essential, and it is expected that to solve various problems in the rock engineering fields which have difficulty in approaching or analyzing at present, research ideas using not only machine learning but also deep learning such as CNN or RNN will increase.

• Journal of Aerospace System Engineering
• /
• v.17 no.6
• /
• pp.16-26
• /
• 2023

With the growing demand for unmanned aerial vehicles (UAVs), various collision avoidance methods have been proposed, mainly using LiDAR and stereo cameras. However, it is difficult to apply these sensors to small UAVs due to heavy weight or lack of space. The recently proposed methods use a combination of object recognition models and distance sensors, but they lack information on the obstacle size. This disadvantage makes distance determination and obstacle coordination complicated in an early-stage collision avoidance. We propose a method for estimating obstacle sizes using a monocular camera-YOLO and infrared sensor. Our experimental results confirmed that the accuracy was 86.39% within the distance of 40 cm. In addition, the proposed method was applied to a small UAV to confirm whether it was possible to avoid obstacle collisions.

• Tunnel and Underground Space
• /
• v.32 no.6
• /
• pp.502-517
• /
• 2022

As the use of TBM increases, research has recently increased to to analyze TBM data with machine learning techniques to predict the exchange cycle of disc cutters, and predict the advance rate of TBM. In this study, a regression prediction of disc cutte wear of slurry shield TBM site was made by combining machine learning based on the machine data and the geotechnical data obtained during the excavation. The data were divided into 7:3 for training and testing the prediction of disc cutter wear, and the hyper-parameters are optimized by cross-validated grid-search over a parameter grid. As a result, gradient boosting based on the ensemble model showed good performance with a determination coefficient of 0.852 and a root-mean-square-error of 3.111 and especially excellent results in fit times along with learning performance. Based on the results, it is judged that the suitability of the prediction model using data including mechanical data and geotechnical information is high. In addition, research is needed to increase the diversity of ground conditions and the amount of disc cutter data.

• Proceedings of the IEEK Conference
• /
• 2004.08c
• /
• pp.743-746
• /
• 2004

This study is generally purposed to develop a mobile augmentative and alternative communication system (hereinafter referred to as 'a mobile AAC system'). Also the device is aimed as a mobile AAC system for helping communication of persons who suffer from speech disorders in a free and convenient manner. One thing firstly considered for a successful AAC system is selection of vocabularies. For this purpose, the above researchers collected utterances in actual situations to select vocabularies of higher frequencies. For generation of sentences by maximizing the use of limited space in the AAC system, this study specifically presents a method of predicting predictions. This method includes selecting vocabulary and classifying it by domains so as to meet the characteristics of the mobile AAC communication, using a noun thesaurus for semantic analysis, and building a sub-category dictionary. Predicting predicates by selecting symbols in accordance with this method is tested and the utility thereof is confirmed.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.1019-1029
• /
• 2020

In this paper, we propose a novel knowledge distillation algorithm to create an compressed deep ensemble network coupled with the combined use of local and global features of face images. In order to transfer the capability of high-level recognition performances of the ensemble deep networks to a single deep network, the probability for class prediction, which is the softmax output of the ensemble network, is used as soft target for training a single deep network. By applying the knowledge distillation algorithm, the local feature informations obtained by training the deep ensemble network using facial subregions of the face image as input are transmitted to a single deep network to create a so-called compressed ensemble DCNN. The experimental results demonstrate that our proposed compressed ensemble deep network can maintain the recognition performance of the complex ensemble deep networks and is superior to the recognition performance of a single deep network. In addition, our proposed method can significantly reduce the storage(memory) space and execution time, compared to the conventional ensemble deep networks developed for face recognition.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.19 no.10
• /
• pp.1075-1082
• /
• 2009

Using GIS data of C-si as basic data when making noise map of road traffic, we estimated exactly the noise excess areas and consequently suggested the population and the area exposed to road traffic noise accurately. We made 3D noise map to assess regional distribution of noise quantitatively. The noise map consists of noise prediction model based on data base such as traffic volume and speed changes for estimating quantitatively the noise and 3D urban space model which includes locations of noise sources, 3D buildings, topography and roads. We made noise standard map according to land use conditions and compared this map to road traffic noise map, and consequently made excess noise map. Using excess noise map, we assessed areas which exceed environmental noise level standards and noise guidelines quantitatively and effectively through GIS spatial analysis, and consequently more accurate noise exposed area and noise exposed population could be estimated. To show buildings' outer walls noise exposure, we analyzed 3D urban noise distributions using 3D-analysis of GIS.

• Proceedings of KOSOMES biannual meeting
• /
• 2006.11a
• /
• pp.131-135
• /
• 2006

The objective of the project is to model and study the geomagnetic field structure and its secular variation in space and in time due to sources in the dynamic fluid outer core. the Earth's spherical harmonic model of the main field and of the secular variation gives the intensity and geomagnetic structure at any location around the Earth, assuming an undistorted, steady state field that no external sources or localized earth anomalies. To consider the practical use of a ship's digital compass in Earth's magnetic field, Earth's spherical harmonic model is searched for the related practical methods and procedures as a basic study in this work.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.25 no.7
• /
• pp.398-405
• /
• 2013

Small and medium-size buildings employ a multi-distributed individual air-conditioning system that utilizes package air conditioners instead of centralized cooling systems, which can allow easier building management and maintenance, along with a diversification of facility use. Inverter driven system heat pumps have been developed to achieve not only an easy distribution control, allowing free combination of indoor units with different models and different capacities, but also wide applications to intelligent air conditioning. However, the control algorithms of the system heat pump are limited in the open literature, due to complicated operating conditions. In this paper, an inverter-driven system heat pump having two indoor units with electronic expansion valves (EEV) was simulated in the cooling mode. An integral optimum regulating controller employing the state space control method was also simulated, and applied to the system-heat pump system, to obtain efficient control of the MIMO (multi input multi output) system. The simulation model for the controller yielded satisfactory prediction results. The new control model can be successfully utilized as a basic tool in controller design.

• Journal of Integrative Natural Science
• /
• v.3 no.1
• /
• pp.49-53
• /
• 2010

Molecular docking is a critical event which mostly forms Van der waals complex in molecular recognition. Since the majority of developed drugs are small molecules, docking them into proteins has been a prime concern in drug discovery community. Since the binding pose space is too vast to cover completely, many search algorithms such as genetic algorithm, Monte Carlo, simulated annealing, distance geometry have been developed. Proper evaluation of the quality of binding is an essential problem. Scoring functions derived from force fields handle the ligand binding prediction with the use of potential energies and sometimes in combination with solvation and entropy contributions. Knowledge-based scoring functions are based on atom pair potentials derived from structural databases. Forces and potentials are collected from known protein-ligand complexes to get a score for their binding affinities (e.g. PME). Empirical scoring functions are derived from training sets of protein-ligand complexes with determined affinity data. Because non of any single scoring function performs generally better than others, some other approaches have been tried. Although numerous scoring functions have been developed to locate the correct binding poses, it still remains a major hurdle to derive an accurate scoring function for general targets. Recently, consensus scoring functions and target specific scoring functions have been studied to overcome the current limitations.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2013.11a
• /
• pp.139-140
• /
• 2013

Ground source heat pump systems can achieve the energy saving of building and reduce CO₂ emission by utilizing stable ground temperature. However, they have many barriers such as high cost of installation, incompletion of design tool, lack of recognition as heating and cooling systems. In order to solve the problems, the building integrated geothermal system (BIGS) developed by several researches which use building foundation as a heat exchanger. In order to establish the optimum design tool of BIGS with the horizontal heat exchanger, the prediction method of ground heat exchange rate developed with numerical simulation model. In this study, the economic analysis for BIGS was conducted based on simulation results and the optimal design method was suggested. As a result, it was found that the case of 32 A, piping space 0.3 m, piping deep 0.5 m and flow rate 9.52 L/min was the best case as 50.1 W/m2 of heat exchange rate. In this case the initial cost was reduced to 115 million won.