• Title/Summary/Keyword: HyperParameter

Search Result 111, Processing Time 0.038 seconds

Quantitative Structure-Activity Relationship Study on Phenylcyclohexylamine (Phenylcyclohexylamine의 정량적 구조-작용 상관관계에 관한 연구)

  • Kim, Ja Hong;Sohn, Sung Ho;Yang, Kee Soo;Hong, Sung Wan
    • Journal of the Korean Chemical Society
    • /
    • v.42 no.4
    • /
    • pp.378-382
    • /
    • 1998
  • A Quantitative Structure-Activity Relationship of 1-phenylcyclohexyl amine(PCA) and dexoxadral as a receptor has been investigated using semiempirical PM3 MO and Hyper Chem calculation. A set of 19 analogues of PCA was chosen for the study using a selection procedure aimed at minimizing the interparameter correlations, while ensuring that the frontier orbital covered the maximum possible range of LogP. The results show that the FOS and LogP is a good structural parameter to predict the maximum electroshock effective dose ($MES\;ED_{50}$) and toxicity dose ($TD_{50}$) for PCA derivatives.

  • PDF

Hyper-Geometric Distribution Software Reliability Growth Model : Generalizatio, Estimation and Prediction (초기하분포 소프트웨어 신뢰성 성장 모델 : 일반화, 추정과 예측)

  • Park, Jung-Yang;Yu, Chang-Yeol;Park, Jae-Hong
    • The Transactions of the Korea Information Processing Society
    • /
    • v.6 no.9
    • /
    • pp.2343-2349
    • /
    • 1999
  • The hyper-geometric distribution software reliability growth model (HGDM) was recently developed and successfully applied to real data sets. The HGDM considers the sensitivity factor as a parameter to be estimated. In order to reflect the random behavior of the test-and-debug process, this paper generalizes the HGDM by assuming that the sensitivity factor is a binomial random variable. Such a generalization enables us to easily understand the statistical characteristics of the HGDM. It is shown that the least squares method produces the identical results for both the HGDM and the generalized HGDM. Methods for computing the maximum likelihood estimates and predicting the future outcomes are also presented.

  • PDF

A Study on the Fatigue Life Prediction and Evaluation of Rubber Components for Automobile Vehicle (자동차 방진고무부품의 피로수명 예측 및 평가)

  • Woo, Chang-Su;Kim, Wan-Doo;Kwon, Jae-Do
    • Transactions of the Korean Society of Automotive Engineers
    • /
    • v.13 no.6
    • /
    • pp.56-62
    • /
    • 2005
  • The fatigue analysis and lifetime evaluation are very important in design procedure to assure the safety and reliability of the rubber components. Fatigue lifetime prediction methodology of the rubber component was proposed by incorporating the finite element analysis and fatigue damage parameter from fatigue test. Finite element analysis of 3D dumbbell specimen and rubber component were performed based on a hyper-elastic material model determined from material test. The Green-Lagrange strain at the critical location determined from the FEM was used for evaluating the fatigue damaged parameter of the natural rubber. Fatigue life of the rubber component are predicted by using the fatigue damage parameter at the critical location. Predicted fatigue lifes of the rubber component agreed fairly well the experimental fatigue lives.

Performance Evaluation of Machine Learning and Deep Learning Algorithms in Crop Classification: Impact of Hyper-parameters and Training Sample Size (작물분류에서 기계학습 및 딥러닝 알고리즘의 분류 성능 평가: 하이퍼파라미터와 훈련자료 크기의 영향 분석)

  • Kim, Yeseul;Kwak, Geun-Ho;Lee, Kyung-Do;Na, Sang-Il;Park, Chan-Won;Park, No-Wook
    • Korean Journal of Remote Sensing
    • /
    • v.34 no.5
    • /
    • pp.811-827
    • /
    • 2018
  • The purpose of this study is to compare machine learning algorithm and deep learning algorithm in crop classification using multi-temporal remote sensing data. For this, impacts of machine learning and deep learning algorithms on (a) hyper-parameter and (2) training sample size were compared and analyzed for Haenam-gun, Korea and Illinois State, USA. In the comparison experiment, support vector machine (SVM) was applied as machine learning algorithm and convolutional neural network (CNN) was applied as deep learning algorithm. In particular, 2D-CNN considering 2-dimensional spatial information and 3D-CNN with extended time dimension from 2D-CNN were applied as CNN. As a result of the experiment, it was found that the hyper-parameter values of CNN, considering various hyper-parameter, defined in the two study areas were similar compared with SVM. Based on this result, although it takes much time to optimize the model in CNN, it is considered that it is possible to apply transfer learning that can extend optimized CNN model to other regions. Then, in the experiment results with various training sample size, the impact of that on CNN was larger than SVM. In particular, this impact was exaggerated in Illinois State with heterogeneous spatial patterns. In addition, the lowest classification performance of 3D-CNN was presented in Illinois State, which is considered to be due to over-fitting as complexity of the model. That is, the classification performance was relatively degraded due to heterogeneous patterns and noise effect of input data, although the training accuracy of 3D-CNN model was high. This result simply that a proper classification algorithms should be selected considering spatial characteristics of study areas. Also, a large amount of training samples is necessary to guarantee higher classification performance in CNN, particularly in 3D-CNN.

A Study on the Prediction of Optimized Injection Molding Condition using Artificial Neural Network (ANN) (인공신경망을 활용한 최적 사출성형조건 예측에 관한 연구)

  • Yang, D.C.;Lee, J.H.;Yoon, K.H.;Kim, J.S.
    • Transactions of Materials Processing
    • /
    • v.29 no.4
    • /
    • pp.218-228
    • /
    • 2020
  • The prediction of final mass and optimized process conditions of injection molded products using Artificial Neural Network (ANN) were demonstrated. The ANN was modeled with 10 input parameters and one output parameter (mass). The input parameters, i.e.; melt temperature, mold temperature, injection speed, packing pressure, packing time, cooling time, back pressure, plastification speed, V/P switchover, and suck back were selected. To generate training data for the ANN model, 77 experiments based on the combination of orthogonal sampling and random sampling were performed. The collected training data were normalized to eliminate scale differences between factors to improve the prediction performance of the ANN model. Grid search and random search method were used to find the optimized hyper-parameter of the ANN model. After the training of ANN model, optimized process conditions that satisfied the target mass of 41.14 g were predicted. The predicted process conditions were verified through actual injection molding experiments. Through the verification, it was found that the average deviation in the optimized conditions was 0.15±0.07 g. This value confirms that our proposed procedure can successfully predict the optimized process conditions for the target mass of injection molded products.

A MRAS Speed and Stator Flux Linkage Estimator for Permanent Magnet Synchronous Motor drives with parameter identification (파라미터 산정과 영구자석 동기전동기 제어를 위한 MRAS Speed 와 Stator Flux Linkage 추정량)

  • Lin, Hai;Kwon, Byung-Il
    • Proceedings of the KIEE Conference
    • /
    • 2011.07a
    • /
    • pp.830-831
    • /
    • 2011
  • The paper makes an investigation on a speed and stator flux linkage estimator for permanent magnet synchronous motor (PMSM) sensorless drives using the technology of model reference adaptive system (MRAS). The designed estimator including two models and two adaptive estimating laws is proved to be stable by the Popov hyper-stability theory. The speed, the stator flux linkage and the resistance are estimated accurately by the proposed estimator while overcoming the shortcoming of the traditional one. The experiment results demonstrate its effectiveness.

  • PDF

A Design of Stable Continuous-time Model Reference Adaptive Controllers by a Hyperstability Method (초안정도 방법에 의한 안정한 시연속 기준모델 적응제어기의 설계)

  • 이호진;정종대;최계근
    • Journal of the Korean Institute of Telematics and Electronics
    • /
    • v.26 no.10
    • /
    • pp.1488-1497
    • /
    • 1989
  • In this paper, a new adaptive control scheme is proposed that uses a special form of rational function-type linear operator in the parameter adaptation and that removes the augmenting signal terms of the control input components. This adaptation scheme is applied to the MRAC of continuous-time, linear time-invariant, minimum-phase plants whose relative degrees are arbitrary. This scheme can be applied without any change of the controller structure to the adaptive systems regardless of the relative degree if it is greater than 1. And this scheme does not require any signal augmentation for arbitrary relative-degree plants if the reference model has no zeros. The asymptotic stability of the adaptive systems controlled by this scheme is shown by a hyper-stability method.

  • PDF

Application of Convolution Neural Network to Flare Forecasting using solar full disk images

  • Yi, Kangwoo;Moon, Yong-Jae;Park, Eunsu;Shin, Seulki
    • The Bulletin of The Korean Astronomical Society
    • /
    • v.42 no.2
    • /
    • pp.60.1-60.1
    • /
    • 2017
  • In this study we apply Convolution Neural Network(CNN) to solar flare occurrence prediction with various parameter options using the 00:00 UT MDI images from 1996 to 2010 (total 4962 images). We assume that only X, M and C class flares correspond to "flare occurrence" and the others to "non-flare". We have attempted to look for the best options for the models with two CNN pre-trained models (AlexNet and GoogLeNet), by modifying training images and changing hyper parameters. Our major results from this study are as follows. First, the flare occurrence predictions are relatively good with about 80 % accuracies. Second, both flare prediction models based on AlexNet and GoogLeNet have similar results but AlexNet is faster than GoogLeNet. Third, modifying the training images to reduce the projection effect is not effective. Fourth, skill scores of our flare occurrence model are mostly better than those of the previous models.

  • PDF

An Improved Flux Observer for Sensorless Permanent Magnet Synchronous Motor Drives with Parameter Identification

  • Lin, Hai;Hwang, Kyu-Yun;Kwon, Byung-Il
    • Journal of Electrical Engineering and Technology
    • /
    • v.8 no.3
    • /
    • pp.516-523
    • /
    • 2013
  • This paper investigates an improved stator flux linkage observer for sensorless permanent magnet synchronous motor (PMSM) drives using a voltage-based flux linkage model and an adaptive sliding mode variable structure. We propose a new observer design that employs an improved sliding mode reaching law to achieve better estimation accuracy. The design includes two models and two adaptive estimating laws, and we illustrate that the design is stable using the Popov hyper-stability theory. Simulation and experimental results demonstrate that the proposed estimator accurately calculates the speed, the stator flux linkage, and the resistance while overcoming the shortcomings of traditional estimators.

Claims Reserving via Kernel Machine

  • Kim, Mal-Suk;Park, He-Jung;Hwang, Chang-Ha;Shim, Joo-Yong
    • Journal of the Korean Data and Information Science Society
    • /
    • v.19 no.4
    • /
    • pp.1419-1427
    • /
    • 2008
  • This paper shows the kernel Poisson regression which can be applied in the claims reserving, where the row effect is assumed to be a nonlinear function of the row index. The paper concentrates on the chain-ladder technique, within the framework of the chain-ladder linear model. It is shown that the proposed method can provide better reserve estimates than the Poisson model. The cross validation function is introduced to choose optimal hyper-parameters in the procedure. Experimental results are then presented which indicate the performance of the proposed model.

  • PDF