• International Journal of Concrete Structures and Materials
• /
• v.11 no.2
• /
• pp.261-273
• /
• 2017

The control of the deformative behaviour of pre-stressed concrete roof elements for a satisfactory service performance is a main issue of their structural design. Slender light-weight wing-shaped roof elements, typical of the European heritage, are particularly sensitive to this problem. The paper presents the results of deformation measurements during storage and of both torsional-flexural and purely flexural load tests carried out on a full-scale 40.5 m long innovative wing-shaped roof element. An element-based simplified integral procedure that de-couples the evolution of the deflection profile with the progressive shortening of the beam is adopted to catch the experimental visco-elastic behaviour of the element and the predictions are compared with normative close-form solutions. A linear 3D fem model is developed to investigate the torsional-flexural behaviour of the member. A mechanical non-linear beam model is used to predict the purely flexural behaviour of the roof member in the pre- and post-cracking phases and to validate the loss prediction of the adopted procedure. Both experimental and numerical results highlight that the adopted analysis method is viable and sound for an accurate simulation of the service behaviour of precast roof elements.

• Journal of Korean Elementary Science Education
• /
• v.24 no.1
• /
• pp.43-50
• /
• 2005

This study examined the effects of the result of ascertaining predictions on cognitive conflict and conceptual change when students teamed the concept of weightlessness. Participants were 200 pre-service elementary teachers. They answered the pretest composed of two items. Through the demonstration on either of two items of the pretest, they identified whether their predictions were correct or not. In addition, students' cognitive conflicts were measured. After brief instructional treatment, the posttest was conducted. The results of this study are as follows: The more students who identified their own predictions on the experiment were incorrect there were, the more effective it was on cognitive conflict and conceptual change. And cognitive conflicts and conceptual changes of students who identified that their predictions were incorrect were generated meaningfully more than those of students who identified that their predictions were correct. From these results, it is concluded that students who identified that their predictions were correct experience cognitive conflicts, but their cognitive conflicts and conceptual changes were smaller than those of students who identified that their predictions were incorrect.

• Journal of Korean Society for Quality Management
• /
• v.50 no.3
• /
• pp.459-471
• /
• 2022

Purpose: The purpose of this study was to increase prediction accuracy for an anomaly interval identified using an artificial intelligence-based time series anomaly detection technique by establishing a pre-processing process. Methods: Significant variables were extracted by applying feature selection techniques, and anomalies were derived using the TadGAN time series anomaly detection algorithm. After applying machine learning and deep learning methodologies using normal section data (excluding anomaly sections), the explanatory power of the anomaly sections was demonstrated through performance comparison. Results: The results of the machine learning methodology, the performance was the best when SHAP and TadGAN were applied, and the results in the deep learning, the performance was excellent when Chi-square Test and TadGAN were applied. Comparing each performance with the papers applied with a Conventional methodology using the same data, it can be seen that the performance of the MLR was significantly improved to 15%, Random Forest to 24%, XGBoost to 30%, Lasso Regression to 73%, LSTM to 17% and GRU to 19%. Conclusion: Based on the proposed process, when detecting unsupervised learning anomalies of data that are not actually labeled in various fields such as cyber security, financial sector, behavior pattern field, SNS. It is expected to prove the accuracy and explanation of the anomaly detection section and improve the performance of the model.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.559-561
• /
• 2022

In order to secure the stability of tap water production and supply, we have built a system that can be pre-verified before applying AI-based composite sensors to the water purification plant, which is a demonstration site. We have collected and analyzed data related to the drug input of the GO-RYEONG water purification plant for about two years from December 2019 to December 2021. The outliers of each tag were removed through data preprocessing such as outliers and derived variable, and the cycle was set as average data for 60 minutes of each one-minute period, and the model was learned using the PLS model.

• Journal of the Korean Solar Energy Society
• /
• v.36 no.2
• /
• pp.19-29
• /
• 2016

Whilst there are growing interests in pursuing energy efficiency and zero-energy buildings in built environment, it is widely recognised that Building-Integrated Photovoltaic (BIPV) is one of the most promising and required technologies to achieve these goals in recent years. Although BIPV is a broadly utilized technique in variety of fields in built environments, it is required that generation of BIVP should be analysed and calculated by external specialists. The aim of this research is to focus on developing a new diagram for prediction of the pre-estimation model in early design stage to harness solar radiation data, PV types, slopes, azimuth and so forth. The results of this study show as follows: 1) We analysed 162 districts in a national level and the examined areas were categorised into five zones. The standard deviation of the results was 2.9 per cent; 2) The increased value of solar radiation on a vertical plane in five categorised zones was 42kWh/m3, and the result was similar to the average value of 43.8kWh/m3; and 3) The pre-estimation of diagram was developed based on the categorisation of zones and azimuth as well as the results of the developed diagram showed little difference compared to the previously utilised method. The suggested diagram in this paper will contribute to estimate BIPV without any external contribution to calculate the value. Even though the result of this study shows little difference, it is required to investigate a number of different variables such as BIPV types, modules, slope angle and so forth in order to develop an integrated pre-estimation diagram.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.6B
• /
• pp.534-543
• /
• 2006

As the topology frequently varies, more cluster reconstructing is needed and also management overheads increase in the wireless ad hoc/sensor networks. In this paper, we propose a multi-hop clustering algorithm for wireless sensor network topology management using dynamic pre-clusterhead scheme to solve cluster reconstruction and load balancing problems. The proposed scheme uses weight map that is composed with power level and mobility, to choose pre-clusterhead and construct multi-hop cluster. A clusterhead has a weight map and threshold to hand over functions of clusterhead to pre-clusterhead. As a result of simulation, our algorithm can reduce overheads and provide more load balancing well. Moreover, our scheme can maintain the proper number of clusters and cluster members regardless of topology changes.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.4
• /
• pp.145-152
• /
• 2020

MRC (Machine reading comprehension) is the AI NLP task that predict the answer for user's query by understanding of the relevant document and which can be used in automated consult services such as chatbots. Recently, the BERT (Pre-training of Deep Bidirectional Transformers for Language Understanding) model, which shows high performance in various fields of natural language processing, have two phases. First phase is Pre-training the big data of each domain. And second phase is fine-tuning the model for solving each NLP tasks as a prediction. In this paper, we have made the Patent MRC dataset and shown that how to build the patent consultation training data for MRC task. And we propose the method to improve the performance of the MRC task using the Pre-trained Patent-BERT model by the patent consultation corpus and the language processing algorithm suitable for the machine learning of the patent counseling data. As a result of experiment, we show that the performance of the method proposed in this paper is improved to answer the patent counseling query.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.25 no.3
• /
• pp.305-314
• /
• 2001

A modified model is proposed for calculation of transitional boundary layer flows. In order to develop the eddy viscosity model for the problem, the flow is divided into three regions; namely, pre-transition region, transition region and fully turbulent region. The pre-transition eddy-viscosity is formulated by extending the mixing length concept. In the transition region, the eddy-viscosity model employs two length scales, i.e., pre-transition length scale and turbulent length scale pertaining to the regions upstream and the downstream, respectively, and a universal model of stream-wise intermittency variation is used as a function bridging the pre-transition region and the fully turbulent region. The proposed model is applied to calculate three benchmark cases of the transitional boundary layer flows with different free-stream turbulent intensity (1%∼6%) under zero-pressure gradient. It was found that the profiles of mean velocity and turbulent intensity, local maximum of velocity fluctuations, their locations as well as the stream-wise variation of integral properties such as skin friction, shape factor and maximum velocity fluctuations are very satisfactorily predicted throughout the flow regions.

• The Journal of the Acoustical Society of Korea
• /
• v.27 no.3
• /
• pp.154-162
• /
• 2008

Signals in an underwater channel get distorted by multipath propagation. In this paper, pre-coding method is suggested which helps comprehending the signals with minimum equalization. The signals are transformed based on the knowledge of the impulse response of the channel. Proposed pre-coding method is tested by simulations based on the ray theory and through water tank experiments. In weak multipath environment, in case of an SNR of about 20 dB, BER is $10^{-3}{\sim}10^{-4}$, while in strong multipath environment, similar BER is achieved with SNR of about 30 dB. In order for the pre-coding method to be used for underwater vehicles, channel prediction method utilizing the waveguide invariant is suggested and tested.

• Magazine of the Korea Concrete Institute
• /
• v.6 no.6
• /
• pp.180-189
• /
• 1994

A practical methodology to detect and localm da~nagc in rcinforced concrete structures by utilizing modal response parameters of as built and tiamaged states is presented. First, a damage detection algorithm which yields information on the, location of damage directly from changes in mode shapes of structures is outlined. Next, the algorithm is implemented to detec and localize damage in a real, 1 1/3 scale, reinforced concrete structure. A set of pre-damage and post damage modal parameters are used for I he damage detection exercise. The results of the damage prediction show that the proposed algorithm can correctly locate the damage inflicted in the test structure.