• Structural Engineering and Mechanics
• /
• 제51권2호
• /
• pp.299-313
• /
• 2014

In the first part of this paper, the influences of some of crack parameters on natural frequencies of a cracked cantilever Functionally Graded Beam (FGB) are studied. A cantilever beam is modeled using Finite Element Method (FEM) and its natural frequencies are obtained for different conditions of cracks. Then effect of variation of depth and location of cracks on natural frequencies of FGB with single and multiple cracks are investigated. In the second part, two Multi-Layer Feed Forward (MLFF) Artificial Neural Networks (ANNs) are designed for prediction of FGB's Cracks' location and depth. Particle Swarm Optimization (PSO) and Back-Error Propagation (BEP) algorithms are applied for training ANNs. The accuracy of two training methods' results are investigated.

• KIEE International Transactions on Power Engineering
• /
• 제2A권4호
• /
• pp.136-144
• /
• 2002

Load models are very important for improving the accuracy of stability analysis and load flow studies. Various loads are connected to a power bus and their characteristics of power consumption change with voltage and frequency. Thus, the effect of voltage/frequency changes must be considered in load modeling. In this work, artificial neural networks-ANNs- were used to construct the component load models for more accurate modeling. A typical residential load was selected and subjected to a test under variable voltage/frequency conditions. Acquired data were used to construct component models by ANNs. The aggregation process of separately determined load models is also presented in the paper. Furthermore, this paper proposes a method to transform a single load model constructed by the aggregation method into a mathematical load model that can be used in traditional power system analysis software.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1997년도 하계학술대회 논문집 B
• /
• pp.485-487
• /
• 1997

Schemes for detecting and diagnosing faults are presented. Faults are detected when residuals change significantly and thresholds are exceed. Two stage artificial neural networks are applied to diagnose faults. The idealized steady state patterns of residuals are defined and learned by ANNs using back propagation algorithm. The first stage ANN is trained to classify the subsystem in which the various faults are located. The first stage ANN could be also used to detect faults with threshold, checking. The second stage ANNs are trained to discriminate the specific cause of a fault at the subsystem level.

• Bulletin of the Korean Chemical Society
• /
• 제28권9호
• /
• pp.1477-1484
• /
• 2007

Artificial neural networks (ANNs), for a first time, were successfully developed for the prediction partial molar heat capacity of aqueous solutions at infinite dilution for various polar aromatic compounds over wide range of temperatures (303.55-623.20 K) and pressures (0.1-30.2 MPa). Two three-layered feed forward ANNs with back-propagation of error were generated using three (the heat capacity in T = 303.55 K and P = 0.1 MPa, temperature and pressure) and six parameters (four theoretical descriptors, temperature and pressure) as inputs and its output is partial molar heat capacity at infinite dilution. It was found that properly selected and trained neural networks could fairly represent dependence of the heat capacity on the molecular descriptors, temperature and pressure. Mean percentage deviations (MPD) for prediction set by the models are 4.755 and 4.642, respectively.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 추계학술대회 논문집 전력기술부문
• /
• pp.91-93
• /
• 2004

Load forecasting is essential in order to supply electrical energy stably and economically in power systems. ANNs have flexibility to predict a nonlinear feature of load profiles. In this paper, we selected just the necessary input variables used in the paper(2) which is based on the phase-space embedding of a load time-series and reviewing others. So only 5 input variables were selected to forecast for spring, fall and winter season and another input considering temperature sensitivity is added during the summer season. The training cases are also selected from all previous data composed training cases of a 7-day, 14-day and 30-day period. Finally, we selected the training case of a 7-day period because it can be used in STLF without sacrificing the accuracy of the forecast. This allows more compact ANNs, smaller training cases. Consequently, test results show that compact neural networks can be forecasted without sacrificing the accuracy.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2019년도 학술발표회
• /
• pp.159-159
• /
• 2019

In this study, the simulation of sediment deposition at Sangju weir reservoir, South Korea, was carried out using artificial neural networks. The ANNs have typically been used in water resources engineering problems for their robustness and high degree of accuracy. Three basic variables namely turbid water inflow, outflow, and water stage have been used as input variables. It was found that ANNs were able to establish valid relationship between input variables and target variable of sedimentation. The R value was 0.9806, 0.9091, and 0.8758 for training, validation, and testing phase respectively. Comparative analysis was also performed to find optimum structure of ANN for sediment deposition prediction. 3-14-1 network architecture using BR algorithm outperformed all other combinations. It was concluded that ANN possess mapping capabilities for complex, non-linear phenomenon of reservoir sedimentation.

• Computers and Concrete
• /
• 제24권5호
• /
• pp.399-412
• /
• 2019

In this investigation, Elman neural networks were utilized for predicting the mechanical properties of Self-Compacting Concretes (SCCs). Elman models were designed by using experimental data of many different concrete mixdesigns of various types of SCC that were collected from the literature. In order to investigate the effectiveness of the selected input variables on the network performance in predicting intended properties, utilized data in artificial neural networks were considered in two sets of 8 and 140 input variables. The obtained outcomes showed that not only can the developed Elman ANNs predict the mechanical properties of SCCs with high accuracy, but also for all of the desired outputs, networks with 140 inputs, compared to ones with 8, have a remarkable percent improvement in the obtained prediction results. The prediction accuracy can significantly be improved by using a more complete and accurate set of key factors affecting the desired outputs, as input variables, in the networks, which is leading to more similarity of the predicted results gained from networks to experimental results.

• 한국터널지하공간학회 논문집
• /
• 제16권3호
• /
• pp.321-339
• /
• 2014

본 논문에서는 인공신경망 기술을 이용한 TBM 터널 세그먼트 라이닝의 설계 시스템 개발에 관한 내용을 다루었다. 먼저 개발 시스템에 대한 개념 및 개발 과정과 시스템을 구성하는 각 요소기술 및 개별 모듈 개발에 관한 내용을 기술하였다. 본 시스템의 요소기술인 ANN-기반의 세그먼트 라이닝 부재력 예측 시스템에 대해 그 개념과 ANN 학습과정 및 검증과정을 기술하였다. ANN-기반의 세그먼트 라이닝 부재력 예측은 유한요소해석을 토대로 구축한 DB를 ANN을 통해 일반화 한 후 개발된 엔진을 세부 모듈에 접목시켜 별도의 해석 없이 유사 단면 혹은 현장에 적용이 가능하도록 하였다. 또한 해석 대상 단면에 대하여 상용 유한요소해석 프로그램과 연계하여 해석 Input파일의 자동생성이 가능하도록 하였으며 유한요소해석 결과를 통한 단면 검토가 가능하도록 하였다.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2004년도 춘계학술발표대회 논문집 제23권 1호
• /
• pp.337-340
• /
• 2004

한국어 문-음성 합성 시스템(TTS: Text-To-Speech)은 합성음의 자연스러움을 증가시키기 위해 운율 발생 알고리듬을 만들어 시스템에 적용하고 있다. 운율 법칙은 각국의 언어에 대한 언어학적 정보나 자연음에서 구한 운율에 대한 지식을 기반으로 음성 합성 시스템에 적용하고 있다. 그러나 이렇게 구한 운율 법칙이 자연음에 존재하는 모든 운율 법칙을 포함할 수도 없고, 또 추출한 운율 법칙이 틀린 법칙이라면, 합성음의 자연감이나 이해도는 떨어질 것이므로, TTS의 실용화에 장애가 될 수 있다. 이러한 점을 감안하여 본 논문에서는 자연음에 내재하는 운율을 학습할 수 있는 인공 신경망을 이용한 운율발생 신경망을 제안하였다. 훈련단계에서 인공 신경망의 입력 단에 한국어 문장의 음소 열을 차례로 이동시켜 인가하면 입력 단의 중앙에 해당하는 음소의 운율 정보가 출력되도록 훈련시킬 때, 목표 패턴을 이용한 감독학습을 통해, 자연음에 내재하는 운율을 학습하도록 하였다. 평가 단계에서 문장의 음소 열을 입력하고, 추정율을 측정하여 인공 신경망이 한국어 문장에 내재하는 운율을 학습하여 발생시킬 수 있음을 살펴보았다.

• 한국터널지하공간학회 논문집
• /
• 제16권1호
• /
• pp.13-24
• /
• 2014

본 논문에서는 TBM 터널의 세그먼트 라이닝 설계 자동화 기술 개발의 일환으로 인공신경망기법을 이용한 세그먼트 라이닝 부재력 산정기법 개발에 관한 내용을 다루었다. 부재력 평가가 가능한 인공신경망을 개발하기 위해 먼저 다양한 설계조건을 도출하고 이에 대해 2-Ring Beam 모델을 이용한 유한요소해석을 수행하여 인공신경망 학습에 필요한 설계조건별 부재력에 관한 DB를 구축하였다. 구축된 DB를 활용하여 인공신경망의 최적화 과정을 통해 최대 부재력 및 분포도를 예측할 수 있는 인공신경망을 구축하였다. 검토 결과 구축된 인공신경망은 유한요소해석과 동일한 정밀도의 부재력 산정 기능을 확보하는 것으로 검토되었으며 따라서 TBM 세그먼트 라이닝 설계시 필요한 부재력 평가를 위한 효율적인 수단으로 활용될 수 있는 것으로 판단된다.