• Proceedings of the Korea Concrete Institute Conference
• /
• 2006.11a
• /
• pp.1-12
• /
• 2006

For advanced civil engineering structures a service life of hundred up to hundred fifty and even two hundred years is sometimes required. The prediction of the performance of concrete structures over such a long period requires accurate and reliable predictive models. Most of the presently used, mostly experience based models don't have the quality and reliability that is required for reliable long-term predictions. The models designers are searching for should be based on an accurate description of the relevant degradation mechanisms. The starting point of such models is a realistic description of the microstructure of the concrete. In this presentation the need and the role of fundamental microstructural models for predicting the performance of concrete structures will be presented. An example will be given of a microstructural model with a proven potential for long-term predictions. Besides this also the role of models in general, i.e. in the whole design and execution process of concrete structures, will be dealt with. Finally recent trends in concrete research will be presented, like the research on self-healing cement-bases systems.

• Communications for Statistical Applications and Methods
• /
• v.22 no.5
• /
• pp.507-518
• /
• 2015

It is not easy to select a linear mixed model since the main interest for model building could be different and the number of parameters in the model could not be clearly defined. In this paper, performance of conditional Akaike Information Criteria and its bias-corrected version are compared with marginal Bayesian and Akaike Information Criteria through a simulation study. The results from the simulation study indicate that bias-corrected conditional Akaike Information Criteria shows promising performance when candidate models exclude large models containing the true model, but bias-corrected one prefers over-parametrized models more intensively when a set of candidate models increases. Marginal Bayesian and Akaike Information Criteria also have some difficulty to select the true model when the design for random effects is nested.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.6
• /
• pp.319-327
• /
• 2022

Transformers are the most famous deep learning models that has achieved great success in natural language processing and also showed good performance on computer vision. In this survey, we categorized transformer-based models for computer vision, particularly object detection tasks and perform comprehensive comparative experiments to understand the characteristics of each model. Next, we evaluated the models subdivided into standard transformer, with key point attention, and adding attention with coordinates by performance comparison in terms of object detection accuracy and real-time performance. For performance comparison, we used two metrics: frame per second (FPS) and mean average precision (mAP). Finally, we confirmed the trends and relationships related to the detection and real-time performance of objects in several transformer models using various experiments.

• Smart Structures and Systems
• /
• v.33 no.2
• /
• pp.133-143
• /
• 2024

Piezoelectric (PZT) actuators have been widely used in precision positioning fields for their excellent displacement resolution. However, due to the inherent characteristics of piezoelectric actuators, hysteresis has been proven to greatly reduce positioning performance. In this paper, five mathematical hysteretic models based on activation function are proposed to characterize the nonlinear hysteresis characteristics of piezoelectric actuators. Then the performance of the proposed models is verified by particle swarm optimization (PSO) algorithm and the experiment data. Thirdly, the fitting performance of the proposed models is compared with the classical Bouc-Wen model. Finally, the performance of the five proposed models in modelling hysteresis nonlinearity of piezoelectric drivers is compared, in terms of RMSE, MAPE, SAPE and operation efficiency, and relevant suggestions are given.

• Korean Management Science Review
• /
• v.11 no.1
• /
• pp.107-115
• /
• 1994

Stochastic manufacturing systems are generally formulated as performance models of discrete event systems. In this paper, logical models(as opposed to performance models) of kanban systems are presented which are deterministic and untimed but not stochastic and timed. As a result, the first and second order properties of kanban systems are showed which can be fruitfully applied to the analysis and design of kanban systems.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.578-581
• /
• 2006

Usually to achieve precise geolocation of satellite images, we need to get GCPs (Ground control points) from individual scenes. This requirement greatly increases the cost and processing time for satellite mapping. In this article, we focus on finding appropriate sensor models for entire image strips composing of several adjacent scenes. We tested the feasibility of modelling whole satellite image strips by establishing sensor models of one scene with GCPs and by applying the models to neighboring scenes without GCPs. For this, we developed two types of sensor models: collinearity-based type and orbit-based type and tested them using different sets of unknowns. Results indicated that although the performance of two types was very similar, for modelling individual scenes, it was not for modelling the whole strips. Moreover, the performance of sensor models was remarkably sensitive to different sets of unknowns. It was found that the orbit-based model using attitude biases as unknowns can be used to model SPOT image strips of 420 Km in length.

• Earthquakes and Structures
• /
• v.24 no.5
• /
• pp.317-331
• /
• 2023

The buckling-restrained braced frames with eccentric configurations (BRBFECs) are stable cyclic behavior and high energy absorption capacity. Furthermore, they have an architectural advantage for creating openings like eccentrically braced frames (EBFs). In the present study, it has been suggested to use the performance-based plastic design (PBPD) method to calculate the design base shear of the BRBFEC systems. Moreover, in this study, to reduce the required steel material, it has been suggested to use the performance-based practical design (PBPD) method instead of the force-based design (FBD) method for the design of this system. The 3-, 6-, and 9-story buildings with the BRBFEC system were designed, and the finite element models were modeled. The seismic performance of the models was investigated using two suits of ground motions representing the maximum considered earthquake (MCE) and design basis earthquake (DBE) seismic hazard levels. The results showed that the models designed with the suggested method, which had lower weights compared to those designed with the FBD method, had a desirable seismic performance in terms of maximum story drift and ductility demand under earthquakes at both MCE and DBE seismic hazard levels. This suggests that the steel weights of the structures designed with the PBPD method are about 13% to 18% lesser than the FBD method. However, the residual drifts in these models were higher than those in the models designed with the FBD method. Also, in earthquakes at the DBE hazard level, the residual drifts in all models except the PBPD-6s and PBPD-9s models were less than the allowable reparability limit.

• Smart Media Journal
• /
• v.12 no.6
• /
• pp.35-40
• /
• 2023

In this study paper, we examine pre-trained language models used in Multi-Goal Conversational Recommender Systems (MG-CRS), comparing and analyzing their performances of various pre-trained language models. Specifically, we investigates the impact of the sizes of language models on the performance of MG-CRS. The study targets three types of language models - of BERT, GPT2, and BART, and measures and compares their accuracy in two tasks of 'type prediction' and 'topic prediction' on the MG-CRS dataset, DuRecDial 2.0. Experimental results show that all models demonstrated excellent performance in the type prediction task, but there were notable provide significant performance differences in performance depending on among the models or based on their sizes in the topic prediction task. Based on these findings, the study provides directions for improving the performance of MG-CRS.

• Geomechanics and Engineering
• /
• v.25 no.1
• /
• pp.1-16
• /
• 2021

Machine learning models have been widely used for landslide susceptibility assessment (LSA) in recent years. The large number of inputs or conditioning factors for these models, however, can reduce the computation efficiency and increase the difficulty in collecting data. Feature selection is a good tool to address this problem by selecting the most important features among all factors to reduce the size of the input variables. However, two important questions need to be solved: (1) how do feature selection methods affect the performance of machine learning models? and (2) which feature selection method is the most suitable for a given machine learning model? This paper aims to address these two questions by comparing the predictive performance of 13 feature selection-based machine learning (FS-ML) models and 5 ordinary machine learning models on LSA. First, five commonly used machine learning models (i.e., logistic regression, support vector machine, artificial neural network, Gaussian process and random forest) and six typical feature selection methods in the literature are adopted to constitute the proposed models. Then, fifteen conditioning factors are chosen as input variables and 1,017 landslides are used as recorded data. Next, feature selection methods are used to obtain the importance of the conditioning factors to create feature subsets, based on which 13 FS-ML models are constructed. For each of the machine learning models, a best optimized FS-ML model is selected according to the area under curve value. Finally, five optimal FS-ML models are obtained and applied to the LSA of the studied area. The predictive abilities of the FS-ML models on LSA are verified and compared through the receive operating characteristic curve and statistical indicators such as sensitivity, specificity and accuracy. The results showed that different feature selection methods have different effects on the performance of LSA machine learning models. FS-ML models generally outperform the ordinary machine learning models. The best FS-ML model is the recursive feature elimination (RFE) optimized RF, and RFE is an optimal method for feature selection.

• International Journal of Fluid Machinery and Systems
• /
• v.8 no.4
• /
• pp.221-229
• /
• 2015

For efficient design process of regenerative blower, the present study provides new generalized pressure and leakage flow loss models, which can be used in the performance analysis method of regenerative blower. The present performance analysis on designed blower is made by incorporating momentum exchange theory between impellers and side channel with mean line analysis method, and its pressure loss and leakage flow models are generalized from the related fluid mechanics correlations which can be expressed in terms of blower design variables. The present performance analysis method is applied to four existing models for verifying its prediction accuracy, and the prediction and the test results agreed well within a few percentage of relative error. Furthermore, the present performance analysis method is also applied in developing a new blower used for fuel cell application, and the newly designed blower is manufactured and tested through chamber-type test facility. The performance prediction by the present method agreed well with the test result and also with the CFD simulation results. From the comparison results, the present performance analysis method is shown to be suitable for the actual design practice of regenerative blower.