• Title/Summary/Keyword: 개선 모델

Search Result 6,513, Processing Time 0.041 seconds

Development for Estimation Improvement Model of Wind Velocity using Deep Neural Network (심층신경망을 활용한 풍속 예측 개선 모델 개발)

  • Ku, SungKwan;Hong, SeokMin;Kim, Ki-Young;Kwon, Jaeil
    • Journal of Advanced Navigation Technology
    • /
    • v.23 no.6
    • /
    • pp.597-604
    • /
    • 2019
  • Artificial neural networks are algorithms that simulate learning through interaction and experience in neurons in the brain and that are a method that can be used to produce accurate results through learning that reflects the characteristics of data. In this study, a model using deep neural network was presented to improve the predicted wind speed values in the meteorological dynamic model. The wind speed prediction improvement model using the deep neural network presented in the study constructed a model to recalibrate the predicted values of the meteorological dynamics model and carried out the verification and testing process and Separate data confirm that the accuracy of the predictions can be increased. In order to improve the prediction of wind speed, an in-depth neural network was established using the predicted values of general weather data such as time, temperature, air pressure, humidity, atmospheric conditions, and wind speed. Some of the data in the entire data were divided into data for checking the adequacy of the model, and the separate accuracy was checked rather than being used for model building and learning to confirm the suitability of the methods presented in the study.

Deep Learning Model for Weather Forecast based on Knowledge Distillation using Numerical Simulation Model (수치 모델을 활용한 지식 증류 기반 기상 예측 딥러닝 모델)

  • 유선희;정은성
    • Annual Conference of KIPS
    • /
    • 2023.05a
    • /
    • pp.530-531
    • /
    • 2023
  • 딥러닝에서 지식 증류 기법은 큰 모델의 지식을 작은 모델로 전달하여 작은 모델의 성능을 개선하는 방식이다. 지식 증류 기법은 모델 경량화, 학습 속도 향상, 학습 정확도 향상 등에 활용될 수 있는데, 교사 모델이라 불리는 큰 모델은 일반적으로 학습된 딥러닝 모델을 사용한다. 본 연구에서는 학습된 딥러닝 모델 대신에 수치 기반 시뮬레이션 모델을 사용함으로써 어떠한 효과가 있는지 검증하였으며, 수치 모델을 활용한 기상 예측 모델에서의 지식 증류는 기존 단독 딥러닝 모델 학습 대비 더 작은 학습 횟수(epoch)에서도 동일한 에러 수준(RMSE)까지 도달하여, 학습 속도 측면에서 이득이 있음을 확인하였다.

The Performance Improvement by Using a Role Model (역할 모델(Role model)을 이용한 업무 개선 - 업무의 달인(達人)을 찾아라 -)

  • Jung, Woo-Young;Ryu, Jae-Kwang;Kim, Kyung-Sik
    • The Korean Journal of Nuclear Medicine Technology
    • /
    • v.12 no.3
    • /
    • pp.199-207
    • /
    • 2008
  • Purpose: The purpose of a role model is to find a master or expert who conduct their business effectively and work as a model that helps staff in charge to conduct their works effectively. This process is an effort to find out the deviation of work performance capability among staffs in charge and settle the work process to be optimized to practical business, the standardization of protocol. Materials and Methods: Seoul Asan medical center nuclear department developed 6~7 role modeling items and has applied it to work. I report two operation improvement cases we had applied a role model to Seoul ASAN medical center. Result: By comparison with none application of a role model, we could improve operation efficiency. Conclusion: The main purpose of role modeling is to make the employees who do the business in their charge effectively as a role model and contribute to work performance of employees who are lack of work performance although they performed their works hard. Our team is going to develop role model items continuously and to make a use of it to practical work then improve work performance. We regard it as a useful tool to performance improvement. The performance improvement by using a role model was very useful instrument.

  • PDF

An Enhanced Floor Field based Pedestrian Simulation Model (개선된 Floor Field 기반 보행 시뮬레이션 모델)

  • Jun, Chul-Min
    • Journal of Korea Spatial Information System Society
    • /
    • v.12 no.1
    • /
    • pp.76-84
    • /
    • 2010
  • Many pedestrian simulation models for micro-scale spaces as building indoor areas have been proposed for the last decade and two models - social force model and floor field model - are getting attention. Among these, CA-based floor field model is viewed more favourable for computer simulations than computationally complex social force model. However, Kirchner's floor field model has limitations in capturing the differences in dynamic values of different agents and this study proposes an enhanced algorithm. This study improved the floor field model in order for an agent to be able to exclude the influences of its own dynamic values by changing the data structure, and, also modified the initial dynamic value problem in order to fit more realistic environment. In the simulations, real 3D building data stored in a spatial DBMS were used considering future integration with indoor localization sensors and real time applications.

A Study on Identification using Particle Swarm Optimization for 3-DOF Helicopter System (3-자유도 헬리콥터 시스템의 입자군집최적화 기법을 이용한 시스템 식별)

  • Lee, Ho-Woon;Kim, Tae-Woo;Kim, Tae-Hyoung
    • Journal of the Korean Institute of Intelligent Systems
    • /
    • v.25 no.2
    • /
    • pp.105-110
    • /
    • 2015
  • This study proposes the more improved mathematical model than conventional that for the 3-DOF Helicopter System in Quanser Inc., and checks the validity about the proposed model by performance comparison between the controller based on the conventional model and that based on the proposed model. Research process is next : First, analyze the dynamics for the 3-DOF helicopter system and establish the linear mathematical model. Second, check the eliminated nonlinear-elements in linearization process for establishing the linear mathematical model. And establish the improved mathematical model including the parameters corresponding to the eliminated nonlinear-elements. At that time, it is used for modeling that Particle Swarm Optimization algorithm the meta-heuristic global optimization method. Finally, design the controller based on the proposed model, and verify the validity of the proposed model by comparison about the experimental results between the designed controller and the controller based on the conventional model.

Modified-stoichiometric Model for Describing Hydration of Alkali-Activated Slag (알칼리 활성 슬래그의 수화에 대한 개선된 화학양론적 모델)

  • Abate, Selamu Yihune;Park, Solmoi;Song, Keum-Il;Lee, Bang-Yeon;Kim, Hyeong-Ki
    • Journal of the Korean Recycled Construction Resources Institute
    • /
    • v.9 no.1
    • /
    • pp.1-12
    • /
    • 2021
  • The present study proposes the modified-stoichiometric model for describing hydration of sodium silicate-based alkaliactivated slag(AAS), and compares the results with the thermodynamic modelling-based calculations. The proposed model is based on Chen and Brouwers(2007a) model with updated database as reported in recent studies. In addition, the calculated results for AAS are compared to those for hydrated portland cement. The maximum difference between the proposed model and the thermodynamic calculation for AAS was at most 20%, and the effects of water-to-binder ratio and activator dosages were identically described by both approaches. In particular, the amount of non-evaporable water was within 10% difference, and was in excellent agreement with the experimental results. Nevertheless, notable deviation was observed for the chemical shrinkage, which is largely dependent on the volume of hydrates and pores.

Application of HSPF Model for Effect Analyses of Watershed Management Plans on Receiving Water Qualities (유역관리에 따른 수질개선 효과분석을 위한 HSPF 모델 적용)

  • Song, Hye-Won;Lee, Hye-Won;Choi, Jung-Hyun;Park, Seok-Soon
    • Journal of Korean Society of Environmental Engineers
    • /
    • v.31 no.5
    • /
    • pp.358-363
    • /
    • 2009
  • The HSPF (Hydrological Simulation Program-Fortran) model was applied to the Kyoungan stream watershed to analyze effects of watershed management plans on receiving water qualities. Utilizing BASINS 3.1 GIS program, the Kyoungan stream watershed was divided into 57 sub-basins and model input parameters were obtained, from DEM (Digital Elevation Model), land use type, stream map, and wastewater treatment facilities, etc.. The hydrologic module of the model was validated based on the measured meteorological data and stream flow data. Then the model was calibrated and verified against the field measurements of water qualities, including temperature, DO, BOD, $NO_3-N$, $NH_3-N$, Org-N, TN and TP. In most cases, there were reasonable agreements between measurements and predictions. The validated model was used to analyze the water quality improvements in the main stream of Kyoungan stream according to the watershed management plans in sub-basins, which are three different scenarios: water quality improvement in tributaries through watershed management activities, expansion and up-grade of wastewater treatment plants, and application of first and second scenarios together. It was concluded that expansion and upgrade of wastewater treatment plants would be more effective than watershed management activities. In order to improve water qualities to the satisfactory level, both watershed management and point source control must be required in the Kyoungan stream.

A Study on Models for Technical Security Maturity Level Based on SSE-CMM (SSE-CMM 기반 기술적 보안 성숙도 수준 측정 모델 연구)

  • Kim, Jeom Goo;Noh, Si Choon
    • Convergence Security Journal
    • /
    • v.12 no.4
    • /
    • pp.25-31
    • /
    • 2012
  • The SSE-CMM model is how to verify the level of information protection as a process-centric information security products, systems and services to develop the ability to assess the organization's development. The CMM is a model for software developers the ability to assess the development of the entire organization, improving the model's maturity level measuring. However, this method of security engineering process improvement and the ability to asses s the individual rather than organizational level to evaluate the ability of the processes are stopped. In this research project based on their existing research information from the technical point of view is to define the maturity level of protection. How to diagnose an information security vulnerabilities, technical security system, verification, and implementation of technical security shall consist of diagnostic status. The proposed methodology, the scope of the work place and the current state of information systems at the level of vulnerability, status, information protection are implemented to assess the level of satisfaction and function. It is possible that measures to improve information security evaluation based on established reference model as a basis for improving information security by utilizing leverage.