• Smart Structures and Systems
• /
• 제29권4호
• /
• pp.599-616
• /
• 2022

Bridge displacement contains vital information for bridge condition and performance. Due to the limits of direct displacement measurement methods, the indirect displacement reconstruction methods based on the strain or acceleration data are also developed in engineering applications. There are still some deficiencies of the displacement reconstruction methods based on strain or acceleration in practice. This paper proposed a novel method based on long short-term memory (LSTM) networks to reconstruct the bridge dynamic displacements with the strain and acceleration data source. The LSTM networks with three hidden layers are utilized to map the relationships between the measured responses and the bridge displacement. To achieve the data fusion, the input strain and acceleration data need to be preprocessed by normalization and then the corresponding dynamic displacement responses can be reconstructed by the LSTM networks. In the numerical simulation, the errors of the displacement reconstruction are below 9% for different load cases, and the proposed method is robust when the input strain and acceleration data contains additive noise. The hyper-parameter effect is analyzed and the displacement reconstruction accuracies of different machine learning methods are compared. For experimental verification, the errors are below 6% for the simply supported beam and continuous beam cases. Both the numerical and experimental results indicate that the proposed data fusion method can accurately reconstruct the displacement.

• 한국군사과학기술학회지
• /
• 제25권4호
• /
• pp.412-424
• /
• 2022

Since traditional derivative-based optimization for noisy simulation shows bad performance, evolutionary algorithms are considered as substitutes. Especially in case when outputs are binary, more simulation trials are needed to get near-optimal solution since the outputs are discrete and have high and heterogeneous variance. In this paper, we propose a genetic algorithm called SARAGA which adopts dynamic resampling and fitness approximation using surrogate. SARAGA reduces unnecessary numbers of expensive simulations to estimate success probabilities estimated from binary simulation outputs. SARAGA allocates number of samples to each solution dynamically and sometimes approximates the fitness without additional expensive experiments. Experimental results show that this novel approach is effective and proper hyper parameter choice of surrogate and resampling can improve the performance of algorithm.

• 한국재료학회지
• /
• 제33권4호
• /
• pp.164-174
• /
• 2023

The bandgap characteristics of semiconductor materials are an important factor when utilizing semiconductor materials for various applications. In this study, based on data provided by AFLOW (Automatic-FLOW for Materials Discovery), the bandgap of a semiconductor material was predicted using only the material's compositional features. The compositional features were generated using the python module of 'Pymatgen' and 'Matminer'. Pearson's correlation coefficients (PCC) between the compositional features were calculated and those with a correlation coefficient value larger than 0.95 were removed in order to avoid overfitting. The bandgap prediction performance was compared using the metrics of R² score and root-mean-squared error. By predicting the bandgap with randomforest and xgboost as representatives of the ensemble algorithm, it was found that xgboost gave better results after cross-validation and hyper-parameter tuning. To investigate the effect of compositional feature selection on the bandgap prediction of the machine learning model, the prediction performance was studied according to the number of features based on feature importance methods. It was found that there were no significant changes in prediction performance beyond the appropriate feature. Furthermore, artificial neural networks were employed to compare the prediction performance by adjusting the number of features guided by the PCC values, resulting in the best R² score of 0.811. By comparing and analyzing the bandgap distribution and prediction performance according to the material group containing specific elements (F, N, Yb, Eu, Zn, B, Si, Ge, Fe Al), various information for material design was obtained.

• 한국가시화정보학회지
• /
• 제21권1호
• /
• pp.103-109
• /
• 2023

This research developed a measurement technique that can measure the oxygen temperature inside a high temperature furnace. Instead of measuring only changes in frequency components within a small range used in the existing variable laser absorption spectroscopy, laser spectroscopy technology was used to spread out wavelength of the light source passing through the gas Based on a total of 20,000 image data, research was conducted to predict the temperature of a high-temperature furnace using CNN with black and white images in the form of spectral bands by temperature of 25 to 800 degrees. The optimal model was found through Hyper parameter optimization, R2 score is 0.89, and the accuracy of the test data is 88.73%. Based on this research, it is expected that concentration measurement and air-fuel ratio control technology can be applied.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 추계학술대회
• /
• pp.127-130
• /
• 2022

본 논문은 생성자 손실함수를 이용한 가창 음성합성 모델링에 대한 연구로서 기존 이미지 생성에 최적화된 딥러닝 알고리즘 중 BEGAN모델을 오디오 생성모델(SVS모델)에 적용시킬 때 발생할 수 있는 여러 요인에 대해 분석하고 최적의 품질을 도출하기 위한 실험을 수행하였다. 특히 BEGAN 기반 모델에서 제안된 L1 loss가 어느 시점에서 감마(𝛾)파라미터의 역할을 상실하게 한다는 점을 개선하고자 알파(𝛼)파라미터를 추가한 후 각 파라미터 값들의 구간별 실험을 통해 최적의 값을 찾아냄으로써 가창합성 생성물의 품질향상에 기여할 수 있음을 확인하였다.

• International Journal of Computer Science & Network Security
• /
• 제23권10호
• /
• pp.115-128
• /
• 2023

Deep learning has been incorporating various optimization techniques motivated by new pragmatic optimizing algorithm advancements and their usage has a central role in Machine learning. In recent past, new avatars of various optimizers are being put into practice and their suitability and applicability has been reported on various domains. The resurgence of novelty starts from Stochastic Gradient Descent to convex and non-convex and derivative-free approaches. In the contemporary of these horizons of optimizers, choosing a best-fit or appropriate optimizer is an important consideration in deep learning theme as these working-horse engines determines the final performance predicted by the model. Moreover with increasing number of deep layers tantamount higher complexity with hyper-parameter tuning and consequently need to delve for a befitting optimizer. We empirically examine most popular and widely used optimizers on various data sets and networks-like MNIST and GAN plus others. The pragmatic comparison focuses on their similarities, differences and possibilities of their suitability for a given application. Additionally, the recent optimizer variants are highlighted with their subtlety. The article emphasizes on their critical role and pinpoints buttress options while choosing among them.

• ETRI Journal
• /
• 제45권5호
• /
• pp.822-835
• /
• 2023

Recent semantic segmentation frameworks usually combine low-level and high-level context information to achieve improved performance. In addition, postlevel context information is also considered. In this study, we present a Context ReFinement Network (CRFNet) and its training method to improve the semantic predictions of segmentation models of the encoder-decoder structure. Our study is based on postprocessing, which directly considers the relationship between spatially neighboring pixels of a label map, such as Markov and conditional random fields. CRFNet comprises two modules: a refiner and a combiner that, respectively, refine the context information from the output features of the conventional semantic segmentation network model and combine the refined features with the intermediate features from the decoding process of the segmentation model to produce the final output. To train CRFNet to refine the semantic predictions more accurately, we proposed a sequential training scheme. Using various backbone networks (ENet, ERFNet, and HyperSeg), we extensively evaluated our model on three large-scale, real-world datasets to demonstrate the effectiveness of our approach.

• 대한임베디드공학회논문지
• /
• 제18권6호
• /
• pp.319-325
• /
• 2023

In this study, we proposed a method for hyperparameter optimization in the building and training of a deep learning model designed to process point cloud data collected by a millimeter-wave radar system. The primary aim of this study is to facilitate the deployment of a baseline model in resource-constrained IoT devices. We evaluated a RadHAR baseline deep learning model trained on a public dataset composed of point clouds representing five distinct human activities. Additionally, we introduced a coarse-to-fine hyperparameter optimization procedure, showing substantial potential to enhance model efficiency without compromising predictive performance. Experimental results show the feasibility of significantly reducing model size without adversely impacting performance. Specifically, the optimized model demonstrated a 3.3% improvement in classification accuracy despite a 16.8% reduction in number of parameters compared th the baseline model. In conclusion, this research offers valuable insights for the development of deep learning models for resource-constrained IoT devices, underscoring the potential of hyperparameter optimization and model size reduction strategies. This work contributes to enhancing the practicality and usability of deep learning models in real-world environments, where high levels of accuracy and efficiency in data processing and classification tasks are required.

• 한국산학기술학회논문지
• /
• 제15권9호
• /
• pp.5585-5593
• /
• 2014

본 연구는 기대 교통사고건수 추정을 위해 사용되는 SPF의 이질적 분산계수의 유의성이 이질적 사전분포에 직접적인 영향을 받는다는 가설을 검증하고, 이질적 사전분포에 대한 모형 오설정이 교통 안전개선 사업의 평가결과에 주는 영향의 특성을 분석하기 위해 수행되었다. 구체적으로 본 연구에서는 이질적 분산계수의 유의성과 이질적 사전분포의 연관성을 검증하기 위해 모의실험을 통해 이질적 사전분포를 발생시킨 후 이를 NB모형과 HNB모형을 이용하여 SPF를 추정하여 이질적 과분산계수가 SPF의 평균함수 및 분산함수에 주는 영향을 분석하였다. 또한 추정된 계수추정치를 이용한 사전분포의 초모수 추정치의 오차특성과 이질적 과분산계수를 고려하지 않았을 경우 발생하는 교통사고감소계수(CRF)의 오차 부호와 크기를 상세 분석하여 제시하였다. 모의실험 자료 분석결과 이질적 분산계수의 오추정은 포아송 사전분포의 평균에는 큰 영향을 주지 않으나 분산의 크기를 변화시켜 궁극적으로는 기대교통사고건수의 추정량인 사후평균의 값에 오차를 발생시킬 수 있으며, 구체적으로 이질적 분산함수를 NB모형으로 오설정할 경우 CRF의 값은 참값에 비해 최대 120%의 오차를 발생시키는 것으로 나타났다.

• Korean Chemical Engineering Research
• /
• 제58권4호
• /
• pp.588-595
• /
• 2020

지속적으로 강화되는 환경오염 물질 배출 규제로 인해, 질소 산화물(NO_x)의 배출량 예측 및 관리는 산업 현장에서 많은 관심을 받고 있다. 본 연구에서는 인공지능 기반 질소산화물 배출량 예측모델 개발을 위한 연구모형을 제안하였다. 제안된 연구모형은 데이터의 전처리 과정부터 인공지능 모델의 학습 및 평가까지 모두 포함하고 있으며, 시계열 특성을 가지는 NO_x 배출량을 예측하기 위하여 순환 신경망 중 하나인 Long Short-Term Memory (LSTM) 모델을 활용하였다. 또한 의사결정나무 기법을 활용하여 LSTM의 time window를 모델 학습 이전에 선정하는 방법을 채택하였다. 본 연구에서 제안된 연구모형의 NO_x 배출량 예측 모델은 가열로에서 확보한 조업 데이터로 학습되었으며, 최적 모델은 hyper-parameter를 조절하여 개발되었다. 개발된 LSTM 모델은 학습 데이터 및 평가 데이터에 대하여 모두 93% 이상의 NO_x 배출량 예측 정확도를 나타내었다. 본 연구에 제안된 연구모형은 시계열 특성을 가지는 다양한 대기오염 물질의 배출량 예측모델 개발에 응용될 수 있을 것으로 기대된다.