• 제목/요약/키워드: Error Back Propagation

검색결과 463건 처리시간 0.03초

A New Application of Human Visual Simulated Images in Optometry Services

  • Chang, Lin-Song;Wu, Bo-Wen
    • Journal of the Optical Society of Korea
    • /
    • 제17권4호
    • /
    • pp.328-335
    • /
    • 2013
  • Due to the rapid advancement of auto-refractor technology, most optometry shops provide refraction services. Despite their speed and convenience, the measurement values provided by auto-refractors include a significant degree of error due to psychological and physical factors. Therefore, there is a need for repetitive testing to obtain a smaller mean error value. However, even repetitive testing itself might not be sufficient to ensure accurate measurements. Therefore, research on a method of measurement that can complement auto-refractor measurements and provide confirmation of refraction results needs to be conducted. The customized optometry model described herein can satisfy the above requirements. With existing technologies, using human eye measurement devices to obtain relevant individual optical feature parameters is no longer difficult, and these parameters allow us to construct an optometry model for individual eyeballs. They also allow us to compute visual images produced from the optometry model using the CODE V macro programming language before recognizing the diffraction effects visual images with the neural network algorithm to obtain the accurate refractive diopter. This study attempts to combine the optometry model with the back-propagation neural network and achieve a double check recognition effect by complementing the auto-refractor. Results show that the accuracy achieved was above 98% and that this application could significantly enhance the service quality of refraction.

Real-time estimation of break sizes during LOCA in nuclear power plants using NARX neural network

  • Saghafi, Mahdi;Ghofrani, Mohammad B.
    • Nuclear Engineering and Technology
    • /
    • 제51권3호
    • /
    • pp.702-708
    • /
    • 2019
  • This paper deals with break size estimation of loss of coolant accidents (LOCA) using a nonlinear autoregressive with exogenous inputs (NARX) neural network. Previous studies used static approaches, requiring time-integrated parameters and independent firing algorithms. NARX neural network is able to directly deal with time-dependent signals for dynamic estimation of break sizes in real-time. The case studied is a LOCA in the primary system of Bushehr nuclear power plant (NPP). In this study, number of hidden layers, neurons, feedbacks, inputs, and training duration of transients are selected by performing parametric studies to determine the network architecture with minimum error. The developed NARX neural network is trained by error back propagation algorithm with different break sizes, covering 5% -100% of main coolant pipeline area. This database of LOCA scenarios is developed using RELAP5 thermal-hydraulic code. The results are satisfactory and indicate feasibility of implementing NARX neural network for break size estimation in NPPs. It is able to find a general solution for break size estimation problem in real-time, using a limited number of training data sets. This study has been performed in the framework of a research project, aiming to develop an appropriate accident management support tool for Bushehr NPP.

A Prediction Model of the Sum of Container Based on Combined BP Neural Network and SVM

  • Ding, Min-jie;Zhang, Shao-zhong;Zhong, Hai-dong;Wu, Yao-hui;Zhang, Liang-bin
    • Journal of Information Processing Systems
    • /
    • 제15권2호
    • /
    • pp.305-319
    • /
    • 2019
  • The prediction of the sum of container is very important in the field of container transport. Many influencing factors can affect the prediction results. These factors are usually composed of many variables, whose composition is often very complex. In this paper, we use gray relational analysis to set up a proper forecast index system for the prediction of the sum of containers in foreign trade. To address the issue of the low accuracy of the traditional prediction models and the problem of the difficulty of fully considering all the factors and other issues, this paper puts forward a prediction model which is combined with a back-propagation (BP) neural networks and the support vector machine (SVM). First, it gives the prediction with the data normalized by the BP neural network and generates a preliminary forecast data. Second, it employs SVM for the residual correction calculation for the results based on the preliminary data. The results of practical examples show that the overall relative error of the combined prediction model is no more than 1.5%, which is less than the relative error of the single prediction models. It is hoped that the research can provide a useful reference for the prediction of the sum of container and related studies.

Prediction of removal percentage and adsorption capacity of activated red mud for removal of cyanide by artificial neural network

  • Deihimi, Nazanin;Irannajad, Mehdi;Rezai, Bahram
    • Geosystem Engineering
    • /
    • 제21권5호
    • /
    • pp.273-281
    • /
    • 2018
  • In this study, the activated red mud was used as a new and appropriate adsorbent for the removal of ferrocyanide and ferricyanide from aqueous solution. Predicting the removal percentage and adsorption capacity of ferro-ferricyanide by activated red mud during the adsorption process is necessary which has been done by modeling and simulation. The artificial neural network (ANN) was used to develop new models for the predictions. A back propagation algorithm model was trained to develop a predictive model. The effective variables including pH, absorbent amount, absorbent type, ionic strength, stirring rate, time, adsorbate type, and adsorbate dosage were considered as inputs of the models. The correlation coefficient value ($R^2$) and root mean square error (RMSE) values of the testing data for the removal percentage and adsorption capacity using ANN models were 0.8560, 12.5667, 0.9329, and 10.8117, respectively. The results showed that the proposed ANN models can be used to predict the removal percentage and adsorption capacity of activated red mud for the removal of ferrocyanide and ferricyanide with reasonable error.

PSO based neural network to predict torsional strength of FRP strengthened RC beams

  • Narayana, Harish;Janardhan, Prashanth
    • Computers and Concrete
    • /
    • 제28권6호
    • /
    • pp.635-642
    • /
    • 2021
  • In this paper, soft learning techniques are used to predict the ultimate torsional capacity of Reinforced Concrete beams strengthened with Fiber Reinforced Polymer. Soft computing techniques, namely Artificial Neural Network, trained by various back propagation algorithms, and Particle Swarm Optimization (PSO) algorithm, have been used to model and predict the torsional strength of Reinforced Concrete beams strengthened with Fiber Reinforced Polymer. The performance of each model has been evaluated by using statistical parameters such as coefficient of determination (R2), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The hybrid PSO NN model resulted in an R2 of 0.9292 with an RMSE of 5.35 for training and an R2 of 0.9328 with an RMSE of 4.57 for testing. Another model, ANN BP, produced an R2 of 0.9125 with an RMSE of 6.17 for training and an R2 of 0.8951 with an RMSE of 5.79 for testing. The results of the PSO NN model were in close agreement with the experimental values. Thus, the PSO NN model can be used to predict the ultimate torsional capacity of RC beams strengthened with FRP with greater acceptable accuracy.

인공신경망모델을 이용한 산사태 예측 (Prediction of Landslide Using Artificial Neural Network Model)

  • 홍원표;김원영;송영석;임석규
    • 한국지반공학회논문집
    • /
    • 제20권8호
    • /
    • pp.67-75
    • /
    • 2004
  • 산사태는 인간의 생명과 재산을 위협하는 가장 주요한 자연재해중의 하나이다. 일반적으로 산사태는 토질물성, 지질학적 및 지형학적 특성과 같은 복잡한 문제로 인하여 발생하게 된다. 인공신경망모델은 많은 연구분야에서 적용되고 있으며, 복잡한 문제를 해결하는데 사용되는 유용한 계산방법이다. 본 논문에서는 자연사면의 산사태 발생여부를 조사하기 위하여 오류역전파를 이용한 인공신경망모델을 제안하였다. 제안된 인공신경망 모델은 두가지 경우에 대한 산사태 발생여부의 평가가 가능하다. 한가지는 토질물성데이터만을 적용한 경우이고, 다른 한가지는 토질물성, 지형 및 지질데이터를 적용한 경우이다. 사면의 안정성을 합리적으로 평가하기 위하여, 인공신경망모델을 적용한 SlideEval(Ver. 1.0)을 개발하였다. 인공신경망모델을 이용한 사면의 안정성 평가는 매우 정확한 것으로 나타났다. 특히, 인공신경망모델을 이용한 산사태 예측은 토질물성데이터만을 적용한 경우보다 토질물성, 지형 및 지질데이터를 적용한 경우가 안정하고 정확한 것으로 나타났다. 그리고, 산사태 발생예측에 대한 통계적인 분석결과(한국지질자원 연구원, 2003)와 비교 검토하여 보면 인공신경망 예측결과와 거의 일치하는 것으로 나타났다. 따라서, 인공신경망을 이용한 SlideEval (Ver. 1.0)프로그램은 산사태를 예측하여 사면의 안정성을 평가하는데 적용이 가능하다.

신경회로망에 의한 의료영상 질환인식 (Disease Recognition on Medical Images Using Neural Network)

  • 이준행;이흥만;김태식;이상복
    • 한국방사선학회논문지
    • /
    • 제3권1호
    • /
    • pp.29-39
    • /
    • 2009
  • 본 논문에서는 신경회로망을 이용한 의료영상의 질환부위 인식방법을 제안하였다. 질환부위 인식을 위한 신경회로망은 입력층, 은닉층, 출력층으로 구성하여 적응 오차 역전파 알고리즘으로 학습하였다. 신경회로망에 입력된 의료영상의 특징 파라미터는 웨이브릿 변환에 의하여 분해된 저주파 영역을 행렬식으로 표현하여 특성 다항식의 계수값(n+1)개로 하였다. 추출된 특징 파라미터는 탄젠트시그모이드 전달함수의 범위로 정규화하여 신경회로망의 입력 벡터로 이용하였다. 제안된 방법의 타당성을 입증하기 위해서 실험에 사용된 입력 의료영상을 가지고 모사실험을 통해 질환부위의 인식률을 평가하였다. 실험 결과 4레벨 DWT로 변환된 저주파영역 행렬의 특성 다항식 계수를 탄젠트시그모이드 전달함수의 범위로 정규화하여 신경회로망의 입력 벡터로 이용했을 때 최적의 학습 횟수를 보였다. 신경회로망의 학습은 적응 오차 역전파 알고리즘을 사용하였고, 학습계수를 0.01, 모우멘텀을 0.95로 하였을 때, 위영상에 대해서는 55회, 가슴영상은 55회, CT영상은 46회, 초음파영상은 55회 그리고 혈관영상에 대해서는 157회 등의 최적의 학습 횟수를 보이며 100%의 인식률을 보였다.

  • PDF

신경회로망을 이용한 SVC 계통의 안정화에 관한 연구 (A Study on the SVC System Stabilization Using a Neural Network)

  • 정형환;허동렬;김상효
    • 조명전기설비학회논문지
    • /
    • 제14권3호
    • /
    • pp.49-58
    • /
    • 2000
  • 본 논문에서는 FACTS(Flexible AC Transnission System)로 분류되는 여라 기기중 기존의 전압제어 및 무효 전력보상기들이 가지고 있는 바속응성과 불연속성 문제를 해결해줄 수 있는 정지형 무효전력 보상가(Static Var Compensator : SVC)를 포함한 전력계통에 신경회로망 제어기를 적용하여 안정화에 관하여 연구하였다. 제안된 신경회로망 제어기는 오차와 오차변화량을 입력하는 오차역전과 학습 알고리즘을 사용하고, 학습시간올 단축하기 위해 모멘텀 방법을 사용하였다. 제안된 방법의 강인섬을 입증하기 위해 중부하시 및 정상부하시에 초기 전력을 변동시킨 경우와 초기에 회천자각을 변동시킨 경우에 대하여 시스렘의 회전자각, 각속도 편차 특성 및 단 자전압의 동특성을 고찰하여 다른 시스템보다 응답특성이 우수합을 보였다.

  • PDF

Prediction of unconfined compressive and Brazilian tensile strength of fiber reinforced cement stabilized fly ash mixes using multiple linear regression and artificial neural network

  • Chore, H.S.;Magar, R.B.
    • Advances in Computational Design
    • /
    • 제2권3호
    • /
    • pp.225-240
    • /
    • 2017
  • This paper presents the application of multiple linear regression (MLR) and artificial neural network (ANN) techniques for developing the models to predict the unconfined compressive strength (UCS) and Brazilian tensile strength (BTS) of the fiber reinforced cement stabilized fly ash mixes. UCS and BTS is a highly nonlinear function of its constituents, thereby, making its modeling and prediction a difficult task. To establish relationship between the independent and dependent variables, a computational technique like ANN is employed which provides an efficient and easy approach to model the complex and nonlinear relationship. The data generated in the laboratory through systematic experimental programme for evaluating UCS and BTS of fiber reinforced cement fly ash mixes with respect to 7, 14 and 28 days' curing is used for development of the MLR and ANN model. The data used in the models is arranged in the format of four input parameters that cover the contents of cement and fibers along with maximum dry density (MDD) and optimum moisture contents (OMC), respectively and one dependent variable as unconfined compressive as well as Brazilian tensile strength. ANN models are trained and tested for various combinations of input and output data sets. Performance of networks is checked with the statistical error criteria of correlation coefficient (R), mean square error (MSE) and mean absolute error (MAE). It is observed that the ANN model predicts both, the unconfined compressive and Brazilian tensile, strength quite well in the form of R, RMSE and MAE. This study shows that as an alternative to classical modeling techniques, ANN approach can be used accurately for predicting the unconfined compressive strength and Brazilian tensile strength of fiber reinforced cement stabilized fly ash mixes.

유전알고리즘을 이용한 신경망 구조 및 파라미터 최적화 (Neural Network Structure and Parameter Optimization via Genetic Algorithms)

  • 한승수
    • 한국지능시스템학회논문지
    • /
    • 제11권3호
    • /
    • pp.215-222
    • /
    • 2001
  • 신경망은 선형 시스템뿐만 아니라 비선형 시스템에 있어서도 탁월한 모델링 및 예측 성능을 갖고 있다. 하지만 좋은 성능을 갖는 신경망을 구현하기 위해서는 최적화 해야할 파라미터들이 있다. 은닉층의 뉴런의 수, 학습율, 모멘텀, 학습오차 등이 그것인데 이러한 파라미터들은 경험에 의해서, 또는 문헌들에서 제시하는 값들을 선택하여 사용하는 것이 일반적인 경향이다. 하지만 신경망의 전체적인 성능은 이러한 파라미터들의 값에 의해서 결정되기 때문에 이 값들의 선택은 보다 체계적인 방법을 사용하여 구하여야 한다. 본 논문은 유전 알고리즘을 이용하여 이러한 신경망 파라미터들의 최적 값을 찾는데 목적이 있다. 유전 알고리즘을 이용하여 찾은 파라미터들을 사용하여 학습된 신경망의 학습오차와 예측오차들을 심플렉스 알고리즘을 이용하여 찾는 파라미터들을 사용하여 학습된 신경망의 오차들과 비교하여 본 결과 유전 알고리즘을 이용하여 찾을 파라미터들을 이용했을 때의 신경망의 성능이 더욱 우수함을 알 수 있다.

  • PDF