• Wind and Structures
• /
• v.33 no.6
• /
• pp.423-435
• /
• 2021

Although existing algorithms can predict wind speed using historical observation data, for engineering feasibility, most use moment methods and probability density functions to estimate fitted parameters. However, extreme wind speed prediction accuracy for long-term return periods is not always dependent on how the optimized frequency distribution curves are obtained; long-term return periods emphasize general distribution effects rather than marginal distributions, which are closely related to potential extreme values. Moreover, there are different wind speed parent sample types; how to theoretically select the proper extreme value distribution is uncertain. The influence of different sampling time intervals has not been evaluated in the fitting process. To overcome these shortcomings, updated steps are introduced, involving parameter sensitivity analysis for different sampling time intervals. The extreme value prediction accuracy of unknown parent samples is also discussed. Probability analysis of mean wind is combined with estimation of the probability plot correlation coefficient and the maximum likelihood method; an iterative estimation algorithm is proposed. With the updated steps and comparison using a Monte Carlo simulation, a fitting policy suitable for different parent distributions is proposed; its feasibility is demonstrated in extreme wind speed evaluations at Longhua and Chuansha meteorological stations in Shanghai, China.

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.3
• /
• pp.33-40
• /
• 2011

This paper presents the prediction and evaluation of landslide using LRA(logistic regression analysis) and ANN (Artificial Neural Network) methods. In order to assess the landslide, we selected Sarabong, Byeoldobong area and Mt. Song-ak in Jeju Island. Five factors which affect the landslide were selected as: slope angle, elevation, porosity, dry density, permeability. So as to predict and evaluate the landslide, firstly the weight value of each factor was analyzed by LRA(logistic regression analysis) and ANN(Artificial Neural Network) methods. Then we got two prediction maps using AcrView software through GIS(Geographic Information System) method. The comparative analysis reveals that the slope angle and porosity play important roles in landslide. Prediction map generated by LRA method is more accurate than ANN method in Jeju. From the prediction map, we found that the most dangerous area is distributed around the road and path.

• Nuclear Engineering and Technology
• /
• v.52 no.11
• /
• pp.2471-2478
• /
• 2020

Pumps are essential machinery in the various industries. With the development of high-speed and large-scale pumps, especially high energy density, high requirements have been imposed on the vibration and noise performance of pumps, and cavitation is an important source of vibration and noise excitation in pumps, so it is necessary to improve pumps cavitation performance. The modern pump optimization design method mainly adopts parameterization and artificial intelligence coupling optimization, which requires direct correlation between geometric parameters and pump performance. The existing cavitation performance calculation method is difficult to be integrated into multi-objective automatic coupling optimization. Therefore, a fast prediction method for pump cavitation performance is urgently needed. This paper proposes a novel cavitation prediction method based on impeller pressure isosurface at single-phase media. When the cavitation occurs, the area of pressure isosurface S_iso increases linearly with the NPSH_a decrease. This demonstrates that with the development of cavitation, the variation law of the head with the NPSH_a and the variation law of the head with the area of pressure isosurface are consistent. Therefore, the area of pressure isosurface S_iso can be used to predict cavitation performance. For a certain impeller blade, since the area ratio R_s is proportional to the area of pressure isosurface S_iso, the cavitation performance can be predicted by the R_s. In this paper, a new cavitation performance prediction method is proposed, and the feasibility of this method is demonstrated in combination with experiments, which will greatly accelerate the pump hydraulic optimization design.

• Journal of Mechanical Science and Technology
• /
• v.19 no.6
• /
• pp.1229-1242
• /
• 2005

When natural rubber is used for a long period of time, it becomes aged; it usually becomes hardened and loses its damping capability. This aging process affects not only the material property but also the (fatigue) life of natural rubber. In this paper the aging effects on the material property and the fatigue life were experimentally investigated. In addition, several fatigue life prediction equations for natural rubber were proposed. In order to investigate the aging effects on the material property, the load-stretch ratio curves were plotted from the results of the tensile test, the compression test and the simple shear test for virgin and heat-aged rubber specimens. Rubber specimens were heat-aged in an oven at a temperature ranging from $50^{\circ}C$ to $90^{\circ}C$ for a period ranging from 2 days to 16 days. In order to investigate the aging effects on the fatigue life, fatigue tests were conducted for differently heat-aged hourglass-shaped and simple shear specimens. Moreover, finite element simulations were conducted for the specimens to calculate physical quantities occurring in the specimens such as the maximum value of the effective stress, the strain energy density, the first invariant of the Cauchy-Green deformation tensor and the maximum principal nominal strain. Then, four fatigue life prediction equations based on one of the physical quantities could be obtained by fitting the equations to the test data. Finally, the fatigue life of a rubber bush used in an automobile was predicted by using the prediction equations, and it was compared with the test data of the bush to evaluate the reliability of those equations.

• Journal of Radiation Protection and Research
• /
• v.46 no.4
• /
• pp.204-212
• /
• 2021

Background: Identification of radioisotopes for plastic scintillation detectors is challenging because their spectra have poor energy resolutions and lack photo peaks. To overcome this weakness, many researchers have conducted radioisotope identification studies using machine learning algorithms; however, the effect of data normalization on radioisotope identification has not been addressed yet. Furthermore, studies on machine learning-based radioisotope identifiers for plastic scintillation detectors are limited. Materials and Methods: In this study, machine learning-based radioisotope identifiers were implemented, and their performances according to data normalization methods were compared. Eight classes of radioisotopes consisting of combinations of ²²Na, ⁶⁰Co, and ¹³⁷Cs, and the background, were defined. The training set was generated by the random sampling technique based on probabilistic density functions acquired by experiments and simulations, and test set was acquired by experiments. Support vector machine (SVM), artificial neural network (ANN), and convolutional neural network (CNN) were implemented as radioisotope identifiers with six data normalization methods, and trained using the generated training set. Results and Discussion: The implemented identifiers were evaluated by test sets acquired by experiments with and without gain shifts to confirm the robustness of the identifiers against the gain shift effect. Among the three machine learning-based radioisotope identifiers, prediction accuracy followed the order SVM > ANN > CNN, while the training time followed the order SVM > ANN > CNN. Conclusion: The prediction accuracy for the combined test sets was highest with the SVM. The CNN exhibited a minimum variation in prediction accuracy for each class, even though it had the lowest prediction accuracy for the combined test sets among three identifiers. The SVM exhibited the highest prediction accuracy for the combined test sets, and its training time was the shortest among three identifiers.

• Journal of Preventive Medicine and Public Health
• /
• v.49 no.6
• /
• pp.367-375
• /
• 2016

Objectives: The established theory that breast density is an independent predictor of breast cancer risk is based on studies targeting white women in the West. More Asian women than Western women have dense breasts, but the incidence of breast cancer is lower among Asian women. This meta-analysis investigated the association between breast density in mammography and breast cancer risk in Asian women. Methods: PubMed and Scopus were searched, and the final date of publication was set as December 31, 2015. The effect size in each article was calculated using the interval-collapse method. Summary effect sizes (sESs) and 95% confidence intervals (CIs) were calculated by conducting a meta-analysis applying a random effect model. To investigate the dose-response relationship, random effect dose-response meta-regression (RE-DRMR) was conducted. Results: Six analytical epidemiology studies in total were selected, including one cohort study and five case-control studies. A total of 17 datasets were constructed by type of breast density index and menopausal status. In analyzing the subgroups of premenopausal vs. postmenopausal women, the percent density (PD) index was confirmed to be associated with a significantly elevated risk for breast cancer (sES, 2.21; 95% CI, 1.52 to 3.21; $I^2=50.0%$). The RE-DRMR results showed that the risk of breast cancer increased 1.73 times for each 25% increase in PD in postmenopausal women (95% CI, 1.20 to 2.47). Conclusions: In Asian women, breast cancer risk increased with breast density measured using the PD index, regardless of menopausal status. We propose the further development of a breast cancer risk prediction model based on the application of PD in Asian women.

• Applied Chemistry for Engineering
• /
• v.18 no.6
• /
• pp.636-642
• /
• 2007

Software program to predict the packing density of multi-sized and multi-component particulate system was developed. For this purpose, the experiment to measure the packing density of AP (ammonium perchlorate) and Al (aluminum) particles with different sizes and their mixtures was carried out. The packing densities obtained from various experiments were compared with the predicted data from the developed software program. In the case of the packing density of the binary system, which is comprised of two different size particles and/or two different components, the relative errors were ranged 0.25~13.13%, and in the same venue the relative errors of the ternary system were 0.25~13.13%. Agreement between experimental data and the predicted results is reasonably accurate. In order to achieve the targeted packing density, the software program calculated the contour of the component particles and this will contribute the formulation of optimal packing systems.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.9 no.4
• /
• pp.217-227
• /
• 2007

The western cherry fruit fly, Rhagoletis indifferens Curran (Diptera:Tephritidae), is the most important pest of cultivated cherries in the Pacific Northwest area of the United States, being widely distributed throughout Oregon, Washington, Montana, Utah, Idaho, Colorado and parts of Nevada. The control of R. indifferens has been based on calendar sprays after its first emergence because of their zero tolerance for quarantine. Therefore, a good prediction model is needed for the spray timing. This study was conducted to obtain the empirical population dynamic information of R. indifferens after overwintering in the major cherry growing area of the Pacific Northwest of the United States, where the information is critically needed to develop and validate the prediction model of the fruit fly. Adult fly populations were monitored by using yellow sticky and emergence traps. Larvae growth and density in fruits were observed by fruit sampling and the pupal growth and density were monitored by pupal collection traps. The first adult was emerged around mid May and a large number of adults were caught in early June. A fruit had more than one larva from mid June to early July. A large number of pupae were caught in early July. The pupae were collected in various period of time to determine the effect of pupation timing and the soil moisture content during the winter. A series of population density data collected in each of the developmental stage were analyzed and organized to provide more reliable validation information for the population dynamic models.

• Journal of the Korean GEO-environmental Society
• /
• v.24 no.4
• /
• pp.23-29
• /
• 2023

Ground subsidence shows a mechanism in which the upper ground collapses due to the formation of a cavity due to the movement of soil particles in the ground due to the formation of a waterway because of damage to the water supply/sewer pipes. As a result, cavity is created in the ground and the upper ground is collapsing. Therefore, ground subsidence frequently occurs mainly in downtown areas where a large amount of underground facilities are buried. Accordingly, research to predict the risk of ground subsidence is continuously being conducted. This study tried to present a ground subsidence risk prediction model for two districts of ○○ city. After constructing a data set and performing preprocessing, using the property data of underground facilities in the target area (year of service, pipe diameter), density of underground facilities, and ground subsidence history data. By applying the dataset to the machine learning model, it is evaluated the reliability of the selected model and the importance of the influencing factors used in predicting the ground subsidence risk derived from the model is presented.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.6
• /
• pp.665-671
• /
• 2004

A mathematical model is presented to predict the frost properties and heat and mass transfer within the frost layer formed on a cold plate. The model consists of the laminar flow equations for air-side and the empirical correlation of local frost density. The correlation of local frost density used in this study is obtained from various experimental conditions by considering frosting parameters. The numerical results are compared with experimental data to validate the model, and agree well with experimental data within a maximum error of 9%.