• Journal of the military operations research society of Korea
• /
• v.34 no.2
• /
• pp.79-101
• /
• 2008

This study is for improvement of repair part demand forecasting method of Republic of Korea Air Force aircraft. Recently, demand prediction methods are Weighted moving average, Linear moving average, Trend analysis, Simple exponential smoothing, Linear exponential smoothing. But these use fixed weight and moving average range. Also, NORS(Not Operationally Ready upply) is increasing. Recommended method of Box-Jenkins' ARIMA can solve problems of these method and improve estimate accuracy. To compare recent prediction method and ARIMA that use mean squared error(MSE) is reacted sensitively in change of error. ARIMA has high accuracy than existing forecasting method. If apply this method of study in other several Items, can prove demand forecast Capability.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.635-642
• /
• 2003

Reinforcement learning got successful results in a lot of applications such as control and scheduling. Various function approximation methods have been studied in order to improve the learning speed and to solve the shortage of storage in the standard reinforcement learning algorithm of Q-Learning. Most function approximation methods remove some special quality of reinforcement learning and need prior knowledge and preprocessing. Fuzzy Q-Learning needs preprocessing to define fuzzy variables and Local Weighted Regression uses training examples. In this paper, we propose a function approximation method, Fuzzy Q-Map that is based on on-line fuzzy clustering. Fuzzy Q-Map classifies a query state and predicts a suitable action according to the membership degree. We applied the Fuzzy Q-Map, CMAC and LWR to the mountain car problem. Fuzzy Q-Map reached the optimal prediction rate faster than CMAC and the lower prediction rate was seen than LWR that uses training example.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.20 no.4
• /
• pp.65-76
• /
• 2017

The purpose of this study was to present a prediction model that reflects crime risk area analysis, including factors and spatial characteristics, as a precursor to preparing an alternative plan for crime prevention and design. This analysis of criminal cases in high-risk areas revealed clusters in which approximately 25% of the cases within the study area occurred, distributed evenly throughout the region. This means that using a multiple linear regression model might overestimate the crime rate in some regions and underestimate in others. It also suggests that the number of deserted houses in an analyzed region has a negative relationship with the dependent variable, based on the multiple linear regression model results, and can also have different influences depending on the region. These results reveal that closure signs in a study area affect the dependent variable differently, depending on the region, rather than a simple or direct relationship with the dependent variable, as indicated by the results of the multiple linear regression model.

• Structural Engineering and Mechanics
• /
• v.74 no.1
• /
• pp.55-67
• /
• 2020

The common practice to predict the characteristic structural load effects (LEs) in long reference periods is to employ the extreme value theory (EVT) for building limit distributions. However, most applications ignore that LEs are driven by multiple loading events and thus do not have the identical distribution, a prerequisite for EVT. In this study, we propose the composite extreme value modeling approach using clustering to (a) cluster initial blended samples into finite identical distributed subsamples using the finite mixture model, expectation-maximization algorithm, and the Akaike information criterion; (b) combine limit distributions of subsamples into a composite prediction equation using the generalized Pareto distribution based on a joint threshold. The proposed approach was validated both through numerical examples with known solutions and engineering applications of bridge traffic LEs on a long-span bridge. The results indicate that a joint threshold largely benefits the composite extreme value modeling, many appropriate tail approaching models can be used, and the equation form is simply the sum of the weighted models. In numerical examples, the proposed approach using clustering generated accurate extrema prediction of any reference period compared with the known solutions, whereas the common practice of employing EVT without clustering on the mixture data showed large deviations. Real-world bridge traffic LEs are driven by multi-events and present multipeak distributions, and the proposed approach is more capable of capturing the tendency of tailed LEs than the conventional approach. The proposed approach is expected to have wide applications to general problems such as samples that are driven by multiple events and that do not have the identical distribution.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.984-989
• /
• 2003

The integrated global position system (GPS) and inertial navigation system (INS) has been considered as a cost-effective way of providing an accurate and reliable navigation system for civil and military system. Even the integration of a navigation sensor as a supporting device requires the development of non-traditional approaches and algorithms. The objective of this paper is to assess the feasibility of integrated with GPS and INS information, to provide the navigation capability for long term accuracy of the integrated system. Advanced algorithms are used to integrate the GPS and INS sensor data. That is fuzzy inference system based Weighted Extended Kalman Filter(FWEKF) algorithm INS signal corrections to provided an accurate navigation system of the integrated GPS and INS. Repeatedly, these include INS error, calculated platform corrections using GPS outputs, velocity corrections, position correction and error model estimation for prediction. Therefore, the paper introduces the newly developed technology which is aimed at achieving high accuracy results with integrated system. Finally, in this paper are given the results of simulation tests of the integrated system and the results show very good performance

• International Journal of Aeronautical and Space Sciences
• /
• v.16 no.2
• /
• pp.295-310
• /
• 2015

In the present study, an unstructured mixed mesh flow solver was used to conduct a numerical prediction of the aerodynamic performance of the S-76 rotor in hover. For the present mixed mesh methodology, the near-body flow domain was modeled by using body-fitted prismatic/tetrahedral cells while Cartesian mesh cells were filled in the off-body region. A high-order accurate weighted essentially non-oscillatory (WENO) scheme was employed to better resolve the flow characteristics in the off-body flow region. An overset mesh technique was adopted to transfer the flow variables between the two different mesh regions, and computations were carried out for three different blade configurations including swept-taper, rectangular, and swept-taper-anhedral tip shapes. The results of the simulation were compared against experimental data, and the computations were also made to investigate the effect of the blade tip Mach number. The detailed flow characteristics were also examined, including the tip-vortex trajectory, vortex core size, and first-passing tip vortex position that depended on the tip shape.

• Earthquakes and Structures
• /
• v.11 no.2
• /
• pp.245-264
• /
• 2016

A suitable ground motion intensity measure (IM) plays a crucial role in the seismic performance assessment of a structure. In this paper, we introduce a scalar IM for use in evaluating the seismic response of single-layer reticulated domes. This IM is defined as the weighted geometric mean of the spectral acceleration ordinates at the periods of the dominant vibration modes of the structure considered, and the modal strain energy ratio of each dominant vibration mode is the corresponding weight. Its applicability and superiority to 11 other existing IMs are firstly investigated in terms of correlation with the nonlinear seismic response, efficiency and sufficiency using the results of incremental dynamic analyses which are performed for a typical single-layer reticulated dome. The hazard computability of this newly proposed IM is also briefly discussed and illustrated. A conclusion is drawn that this dominant vibration mode-based scalar IM has the characteristics of strong correlation, high efficiency, good sufficiency as well as hazard computability, and thereby is appropriate for use in the prediction of seismic response of single-layer reticulated domes.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2007.02a
• /
• pp.9-12
• /
• 2007

다시점 동영상 부호화기는 서로 다른 카메라에 의해 영상을 획득하므로 카메라 내부 파라미터의 차이나 조명의 차이 및 변화 등에 의한 시점 간 명도 성분의 불균형을 가지고 있다. 이로 인해 잘못된 변이 추정이 이루어질 수 있으며, 따라서 전체적인 다시점 동영상 부호화의 성능을 크게 저하시킬 수 있다. 본 논문에서는 레지듀얼이 가지고 있는 밝기 차 정보를 이용하여 시점 간의 불균형을 해소하는 가중치 예측 알고리듬을 제안한다. 주변의 인과적인 블록의 레지듀얼 정보를 이용하여 현재 블록과 참조 블록의 밝기 차를 예측하고, 이 값을 이용해 시점 간 불균형을 보정 한 후 변이 추정을 수행한다. 변이 보상 후 계산된 현재 블록의 레지듀얼 평균값을 앞에서 예측된 밝기 차의 값에 누적하여 다음 블록의 밝기 차 예측에 사용한다. 제안된 방법을 실험 영상에 적용한 결과 평균적으로 약 0.2dB의 이득을 얻었다.

• Journal of the Korean Society of Safety
• /
• v.20 no.2 s.70
• /
• pp.6-10
• /
• 2005

With the rapid industrial development, railways have become a main traffic means of transportation. However, rail traffic noise and vibration have become a major problem in urban areas which is a very serious issue for the living environment. Especially, railway noise induced by rail operations has influenced on the residents living near railway tracks. The purpose of this paper is to investigate the Raynoise modeling in railway applications. Generally, my acoustics have been used to investigate the effectiveness of noise barriers in railway applications and barriers are modeled using the commercial software Raynoise. A-weighted sound pressure level have been measured at six locations, 4m from the track and are compared with experimental values. Based on the analysis of the results, Comparison between numerical and experimental values are within 1dB (A). Also, when a train is m through the Raynoise modeling, the general influential sphere of railway noise can be determined. Therefore, this study will be using basic data in establishing effective railway noise prevention plans far the future. Also, we could know that is applicable of Raynoise modeling at railway noise.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2002.10a
• /
• pp.237-240
• /
• 2002

Design rainfall using LH-moments following the consecutive duration were derived by the regional and at-site analysis using the observed and simulated data resulted from Monte Carlo techniques. RRMSE, RBIAS and RR in RRMSE for the design rainfall were computed and compared in the regional and at-site frequency analysis. Consequently, it was shown that the regional analysis can substantially more reduce the RRMSE, RBIAS and RR in RRMSE than at-site analysis in the prediction of design rainfall. RE for an optimal order of L-moments was also computed by the methods of L, L1, L2, L3 and L4-moments for GEV distribution. It was found that the method of L-moments is more effective than the others for getting optimal design rainfall according to the regions and consecutive durations in the regional frequency analysis. Diagrams for the design rainfall derived by the regional frequency analysis using L-moments were drawn according to the regions and consecutive durations by GIS techniques.