• Journal of Korean Tunnelling and Underground Space Association
• /
• v.9 no.1
• /
• pp.19-28
• /
• 2007

Tunnel management is a time-consuming and expensive task. In particular, visual analysis of tunnel inspection often requires extended time and cost and shows problems on data gathering, storage and analysis. This study proposes a new approach to extract information for tunnel management by using a laser scanning technology. A prototype tunnel laser scanner developed was used to obtain point clouds of a railway tunnel surface. Initial processing of laser scanning data was to separate those laser pulses returned from the installations attached to tunnel liner using radiometric and geometric characteristics of laser returns. Once the laser returns from the installations were separated and removed, physically damaged parts on tunnel lining are detected. Based on the plane formed by laser scanner data, damaged parts are detected by analysis of proximity. The algorithms presented in this study successfully detect the physically damaged parts which can be verified by the digital photography of the corresponding location on the tunnel surface.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.5
• /
• pp.155-162
• /
• 2021

n-step TD learning is a combination of Monte Carlo method and one-step TD learning. If appropriate n is selected, n-step TD learning is known as an algorithm that performs better than Monte Carlo method and 1-step TD learning, but it is difficult to select the best values of n. In order to solve the difficulty of selecting the values of n in n-step TD learning, in this paper, using the characteristic that overestimation of Q can improve the performance of initial learning and that all n-step returns have similar values for Q ≈ Q^*, we propose a new learning target, which is composed of the maximum and the mean of all k-step returns for 1 ≤ k ≤ n. Finally, in OpenAI Gym's Atari game environment, we compare the proposed algorithm with n-step TD learning and proved that the proposed algorithm is superior to n-step TD learning algorithm.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.13 no.1
• /
• pp.203-211
• /
• 1993

Experiment on the instability and breaking of the initially modulated deep-water wave train (in wave amplitude or in wave frequency) is performed to investigate the effect of the initial modulation on nonlinear wave evolution. Wave amplitude and frequency modulations are developed earlier and larger than in the case of the uniform deep-water wave trains. However, for small wave steepness in the initially amplitude-modulated wave train, the wave train becomes demodulated and nearly returns to the original wave form at the end of the wave evolution far downstream from the breaking region, with energy returning to the fundamental wave frequency.

• The Korean Journal of Financial Studies
• /
• v.6 no.1
• /
• pp.289-312
• /
• 2000

This paper proposes a pricing model for IPOs which can reconcile the average underpricing phenomenon with the expected wealth maximizing behaviors of market participants. Under the usual informational asymmetry, the optimal offer price for best efforts IPOs is derived as a function of the uncertainty about market's valuation, the expected return on proposed projects and the size of offerings relative to the firm's market value. Depending on these firm-specific characteristics, best efforts IPOs can be underpriced, fairly priced, or overpriced. Introducing the investment banker as an outside information producer, the model is extended to provide empirical implications for pricing and underwriting contract choice decisions which are consistent with the existing empirical evidences. The model predicts that the issuers with greater uncertainty about market's valuation choose best efforts contract over firm commitment contract and the dispersion of initial returns would be greater for best efforts IPOs than for firm commitment IPOs.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2007.08b
• /
• pp.1467-1469
• /
• 2007

We report on the effect of ambient gas on the OLED degradation. The operating voltage and quantum efficiency increases when the device is exposed to the atmospheric gas and then returns to the initial level of the device in vacuum when the atmospheric gas is evacuated. These changes in the OLED performance can be attributed to the ambient gas pressure.

• ETRI Journal
• /
• v.35 no.6
• /
• pp.1126-1133
• /
• 2013

The first spectrum auction in the Republic of Korea is attention-worthy owing to the fierce competition for the only 1.8-GHz spectrum license, the winning bidder of which was suspected of overpaying for its acquisition. This study empirically investigates the existence of a "winner's curse" in the first Korean spectrum auction by using a standard event study methodology. The results show that both the winner and loser experienced significant positive returns on the completion day of the auction. The results imply that there was no winner's curse in the auction and that the losing firm might increase its competitive advantage by acquiring other spectrum licenses despite failing to achieve its initial target spectrum. Therefore, these results suggest that regulators may need to consider bringing about positive short-term wealth benefits to all bidders by appropriately designing a spectrum auction, such as by performing multiband auctions.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.16 no.3
• /
• pp.537-541
• /
• 2002

When a Kigong master concentrates the Qi at Yintang, Laogong(P8), Qihai(CV6) meridian points during Kigong state, the change of magnetic field around acupoints Yintang, Laogong points has been measured using DROS-SQUID apparatus. After smoothing process of the continuously measured magnetic signal around acupoints for a few minutes, we could observe that a series of peaks, magnitude of 1～2 pT and period of 5 sec, appeared and find that these peaks were clearly changed as if switch on and off according to Qi concentration state. Before Qi radiation, a series of the peaks measured on Yintang or Laogong point of a Kigong master shows one of either SW-ON state or SW-OFF state as initial state. During Qi radiation, its state becomes inverse of initial state. After Qi radiation, it returns to the initial state for some cases (called P type ; push button switch type) or it remains inversion state for other cases (called T type; toggle switch type). From the data of peaks measured at different position from the Qi concentration acupoint, we found that the Qi radiation on an acupoint makes the switching effect even not at the acupoint that Qi is concentrated but at the other acupoints that Qi is not concentrated.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.3 no.2
• /
• pp.313-319
• /
• 1999

We devise the 3 models of Multilayered Recurrent Prediction Neural Network(MLRPNN), which are obtained by modifying the Multilayered Perceptron(MLP) with 4 layers. We experimentally study the speech recognition performance of 3 models by a comparative method, according to the variation of the prediction order, the number of neurons in two hidden layers, initial values of connecting weights and transfer function, respectively. By the experiment, the recognition performance of each MLRPNN is better than that of MLP. At the model that returns the output of the upper hidden layer to the lower hidden layer, the recognition performance shows the best value. All MLRPNNs, which have 10 or 15 neurons in the upper and lower hidden layer and is predicted by 3rd or 4th order, show the improved speech recognition rate. On learning, these MLRPNNs have a better recognition rate when we set the initial weights between -0.5 and 0.5, and use the unipolar sigmoid transfer function in the lower hidden layer.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.40 no.1
• /
• pp.50-56
• /
• 2017

The IRR (internal rate of return) is often used by investors for the evaluation of engineering projects. Unfortunately, it is widely known that it has serial flaws. Also, External rate of returns (ERRs) such as ARR (Average Rate of Return) or MIRR (MIRR, Modified Internal Rate of Return) do not differentiate between the real investment and the expenditure. The PRR (Productive rate of return) is faithful to the conception of the return on investment. The PRR uses the effective investment instead of the initial investment. In this paper, we examined two cases of the engineering project. the one is a traditional engineering project with financing activity, another is the project with R&D. Although the IRR has only one value, it overestimates or underestimate profitabilities of Engineering Projects. The ARR and the MARR assume that a returned cash reinvest other projects or assets instead of the project currently executing. Thus they are only one value of a project's profitability, unlike the IRR. But the ARR does not classify into the effective investment and non-investment expenditure. It only accepts an initial expenditure as for an investment. The MIRR also fails to classify into the investment and the expenditure. It has an error of making a loss down as the investment. The IRR works as efficiently as a NPW (Net Present Worth). It clearly expresses a rate of return in respect of an investment in an engineering project with a loan. And it shows its ability in an engineering project with a R&D investment.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.48 no.6
• /
• pp.31-41
• /
• 2006

The analysis used in this work was cost-benefit analysis method. All future costs and returns of a given mushroom house were discounted to the time of initial investment (present) by means of 3.5% discount rate. Then the cost of ownership was compared to the return from the system. This analysis method has been developed and coded into a balance sheet for use on a EXCEL program. Using this programmed analysis,a large number of the case studies were examined using different combinations of economic conditions. These results will be very useful to individuals considering investment in a mushroom house, or any similar production system. By the way of the sensitivity analysis for each important parameter, the change of the marginal cost-benefit period could be finally determined. These parameters were typically construction cost of mushroom house, cost of cooling system, required cooling and heating energy amounts, unit price of mushroom media bottle, growing number of media bottles, production weight per unit bottle, sale price of mushroom, and annual number of growing period, etc.