• 전자공학회논문지B
• /
• 제31B권10호
• /
• pp.44-52
• /
• 1994

In this paper, we assume that the dynamics of DC motor and nonlinear load are unknown. We propose an inverse dynamic model of DC motor and nonlinear load using the artificial neural network and construck speed control system based on the proposed dynamic model. We also propose another dynamic model with speed prediction scheme using the artificial neural network that removes the undesirable time delay effect caused by the computation time during the real-time control. We suggest a dynamic model which has arbitrary number of speed arguments and is especially effective when the motor and load has large moment of inertia. Next, we suggest a controller that combine the neurocontrol and PID control with constant gain. We show that the proposed neurocontrol systems have capabilities of noise rejection and generalization to have good velocity tracking through computer simulations and experiments.

• Steel and Composite Structures
• /
• 제46권6호
• /
• pp.731-744
• /
• 2023

Vehicle load information is an important role in operating and ensuring the structural health of cable-stayed bridges. In this regard, an efficient and economic method is proposed for vehicle load detection based on the observed cable tension and vehicle position using a graph neural network (GNN). Datasets are first generated using the practical advanced analysis program (PAAP), a robust program for modeling and considering both geometric and material nonlinearities of bridge structures subjected to vehicle load with low computational costs. With the superiority of GNN, the proposed model is demonstrated to precisely capture complex nonlinear correlations between the input features and vehicle load in the output. Four popular machine learning methods including artificial neural network (ANN), decision tree (DT), random forest (RF), and support vector machines (SVM) are refereed in a comparison. A case study of a cable-stayed bridge with the typical truck is considered to evaluate the model's performance. The results demonstrate that the GNN-based model provides high accuracy and efficiency in prediction with satisfactory correlation coefficients, efficient determination values, and very small errors; and is a novel approach for vehicle load detection with the input data of the existing monitoring system.

• Tribology and Lubricants
• /
• 제36권6호
• /
• pp.371-377
• /
• 2020

In In this study, we compared the results of a ball bearing life prediction model based on rolling element loads with the results of fatigue life prediction of ball bearings when a stress-based contact fatigue life prediction technique is applied to the ball bearing. We calculate the load acting on each rolling element by the external load of the bearing and apply the result to the Lundberg-Palmgren (LP) theory to calculate ball bearing life based on the rolling element. We also calculate stress-based ball bearing life through contact and fatigue analyses based on contact modeling of the ball and raceway while considering the fatigue test results of AISI 52100 steel. In stress-based life prediction, we use three high-cycle fatigue-determination equations that can predict the fatigue life when multi-axis proportional loads such as rolling-slide contact conditions are applied. These equations are derived from the stress invariant and critical plane methods and the mesoscopic approach. Life expectancy results are compared with those of the LP model. Results of the analysis indicated that the fatigue life was predicted to be lower in the order of the Crossland, Dang Van, and Matake models. Of the three, the Dang Van fatigue model was found to be the closest to the LP life.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2004년도 추계학술대회
• /
• pp.359-364
• /
• 2004

In this study, experiments were tried on the mixed-mode I+II single overloading model which changes the loading mode of overload and fatigue load. Aspects of deformation field in front of the crack which is formed by mixed-mode I+II single overloading were experimentally studied. Then the shape and size of mixed-mode plastic zone were approximately calculated. The propagation behavior of fatigue crack was examined under the test conditions combined by changing the loading mode. The behavior of fatigue cracks were greatly affected by shapes of plastic deformation field and applying mode of fatigue load. Accuracy of prediction and evaluation for fatigue life may be improved by considering all aspects of deformation and behavior of fatigue cracks.

• Industrial Engineering and Management Systems
• /
• 제14권3호
• /
• pp.318-324
• /
• 2015

This paper proposes a dynamic short-term load forecasting method that utilizes a new sequential learning algorithm based on Relevance Vector Machine (RVM). The method performs general optimization of weights and hyperparameters using the current relevance vectors and newly arriving data. By doing so, the proposed algorithm is trained with the most recent data. Consequently, it extends the RVM algorithm to real-time and nonstationary learning processes. The results of application of the proposed algorithm to prediction of electrical loads indicate that its accuracy is comparable to that of existing nonparametric learning algorithms. Further, the proposed model reduces computational complexity.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1995년도 추계학술대회 논문집
• /
• pp.961-964
• /
• 1995

For geometric accuracy in the net shape machining, the problem of tool deflection should be resolved in some fashion. In particular, this is crucial in finish cut operation where slim tools are used. The purpose of this paper is to verify the validity and effectiveness of the prediction model of the machined surface. Experimental results are presented for the cut of steel material with HSS endmill of diameter 6mm on machining center. The results shows that 1) the machining error due totool deflection is serious even in the low cutting load, 2) by using the mechanistic simulation model with experimental coefficients, the machining error was predicted with maximum prediction error of 10% which was significantly reduced to the desired level by the path modification method.

• KEPCO Journal on Electric Power and Energy
• /
• 제6권4호
• /
• pp.413-418
• /
• 2020

초기 학습 데이터의 과적합으로 인한 전력망 상태예측 모델의 성능 감소를 방지하고 예측모델의 예측 정확도 유지를 통한 계속적인 현장활용을 위해서는 기계학습 모델의 예측 정확도를 지속적으로 관리할 필요가 있다. 이를 위해, 본 논문에서는 다양한 요인에 의해 끊임없이 변화하는 전력망 상태 데이터의 특성을 고려하여 예측모델의 정확성과 신뢰성을 높이고 현장 적용 가능한 수준의 품질을 유지하기 위한 기계학습 기반 전력망 상태예측 모델의 성능 유지관리 자동화 기법을 제안한다. 제안 기법은 워크플로우 관리 기술의 적용을 통해 전력망 상태예측 모델 성능 유지관리를 위한 일련의 태스크들을 워크플로우의 형태로 모델링하고 이를 자동화하여 업무를 효율화 하였다. 또한, 기존 기술에서는 시도되지 않았던 학습데이터의 통계적 특성 변화 정도와 예측의 일반화 수준을 모두 고려한 예측모델의 성능 평가를 통해 성능 결과의 신뢰성을 확보하고 이를 통해 예측 모델의 정확도를 일정 수준으로 유지관리하고 더욱 성능이 우수한 예측모델의 신규 개발이 가능하다. 결과적으로 본 논문에서 제안하는 전력망 상태예측 모델 성능 유지관리 자동화 기법을 통해 예측모델의 성능 저하문제를 해결하여 분산자원 연계 등 외부 환경의 변화에 유연한 예측모델 관리를 통해 정확성과 신뢰성이 보장된 예측 모델의 지속적인 활용이 가능하다.

• Steel and Composite Structures
• /
• 제10권6호
• /
• pp.501-516
• /
• 2010

This paper is aiming to study the performances of reinforced high-strength concrete (HSC) short columns confined with aramid fibre-reinforced polymer (AFRP) sheets. An experimental program, which involved 45 confined columns and nine unconfined columns, was carried out in this study. All the columns were circular in cross section and tested under axial compressive load. The considered parameters included the concrete strength, amount of AFRP layers, and ratio of hoop reinforcements. Based on the experimental results, a prediction model for the axial stress-strain curves of the confined columns was proposed. It was observed from the experiment that there was a great increment in the compressive strength of the columns when the amount of AFRP layers increases, similar as the ultimate strain. However, these increments were reduced as the concrete strength increasing. Comparisons with other existing prediction models present that the proposed model can provide more accurate predictions.

• KCI Concrete Journal
• /
• 제11권3호
• /
• pp.99-107
• /
• 1999

In designing the Prestressed concrete box-girder bridge. dead load, prestressing force, creep and shrinkage of concrete are the main factors which influence the camber and deflection of segmental concrete structure under construction. Among these factors the creep and shrinkage are the functions of the time-dependent property which. therefore, must be considered with time. The prediction model for estimating creep and shrinkage of concrete has been suggested by ACI, CEB/FIP, JSCE and KSCE design code and EMM, AEMM, RCM, IDM and SSM has been suggested for analytical method in consideration of time-dependent characteristics. In this study the creep test was carried out for four different curing ages of concrete which were applied to the Prestressed concrete structure at the construction site, and the results of test were compared with the values of creep prediction proposed by the design code. Also the creep test was performed with step-wise incremental stresses and the results were compared to the analytical values.

• 한국인터넷방송통신학회논문지
• /
• 제19권5호
• /
• pp.93-99
• /
• 2019

본 논문에서는 효율적 에너지 관리를 위해 인공 신경망 중 하나인 GRU 모델을 사용하여 전력사용량을 예측하고 예측된 전력사용량과 실제 전력사용량의 비교를 통해 부하를 자동 제어 하는 스마트 미터기를 제안한다. 제안한 스마트 미터기를 통해 GRU 모델을 학습시키기 위해 필요한 전력사용량 데이터를 수집했다. 구현된 스마트 미터기가 전력사용량 자동측정 및 실시간 관찰 기능과 전력사용량 예측을 통한 부하 제어 기능을 가지고 있음을 보여준다. 성능평가 지표 중 하나인 Root Mean Squared Error (RMSE) 값에 약 20%의 마진 값을 이용하여 부하 자동 제어를 위한 기준 값으로 설정했다. 부하 자동 제어 기능을 가진 스마트 미터기로 인해 에너지 관리의 효율성이 증대되는 것을 확인하였다.