• 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.

• 한국전산유체공학회:학술대회논문집
• /
• 2008.03b
• /
• pp.92-95
• /
• 2008

We performed the theoretical calculation of CN Violet radiation using the code SPRADIAN07 to predict the Lee et al.'s experimental measurements and to reinvestigate $C_2$ dissociation rate. CN Violet radiations are calculated under the Boltzmann and non-Boltzmann distribution using two chemical reaction sets: Park-Losev-G\"{o}kcen-Tsang and Park-Losev-G\"{o}kcen-Tsang-Lee models. Our SPRADIAN07 calculations show improvement in prediction of absolute radiation intensity of CN Violet and its decay rate by Park-Losev-G\"{o}kcen-Tsang reaction set with $C_2$ dissociation rate coefficient of $k_f$ = 1.5${\times}$10$^{16}$ exp(-71,600/$T_x$) cm$^3$ mole$^{-1}$ s$^{-1}$.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.10a
• /
• pp.514-517
• /
• 2017

In solar photovoltaic systems, power generation is greatly affected by the weather conditions, so it is essential to predict solar energy for stable load operation. Therefore, data on weather conditions are needed as inputs to machine learning algorithms for solar energy prediction. In this paper, we use 15 kinds of weather data such as the precipitation accumulated during the 3 hours of the surface, upward and downward longwave radiation average, upward and downward shortwave radiation average, the temperature during the past 3 hours at 2 m above from the ground and temperature from the ground surface as input data to the algorithm. We analyzed the statistical characteristics and correlations of weather data and extracted the downward and upward shortwave radiation averages as a major elements of a feature vector with high correlation of 70% or more with solar energy.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.24 no.9
• /
• pp.695-705
• /
• 2014

In this paper, two types of techniques for the prediction of radiated sound pressure due to vibration of a structure are investigated. The prediction performance using wave-number sensing technique is compared to that of conventional prediction method, such as Rayleigh's integral method, for the prediction of far-field radiated sound pressure. For a coupled plate, wave-number components are predicted by the vibration response of plate and the prediction performance of far-field sound is verified. In addition, the applicability of distributed sensors that are not allowable to Rayleigh's integral method is considered and these can replace point sensors. Experimental implementation verified the prediction accuracy of far-field sound radiation by the wave-number sensing technique. Prediction results from the technique are as good as those of Rayleigh's integral method and with distributed sensors, more reduced computation time is expected. To predict the radiated sound by the efficient configuration of structural sensors, composed(synthesized) mode considering sound power contribution is determined and from this size and location of sensors are chosen. Four types of sensor configuration are suggested, simulated and compared.

• Journal of Radiation Industry
• /
• v.17 no.3
• /
• pp.265-273
• /
• 2023

To effectively and safely manage the radiation exposure to nuclear power plant (NPP) workers in accidents, major overseas NPP operators such as the United States, Germany, and France have developed and applied realistic 3D model radiation dose assessment software for workers. Continuous research and development have recently been conducted, such as performing NPP accident management using 3D-VR based on As Low As Reasonably Achievable (ALARA) planning tool. In line with this global trend, it is also required to secure technology to manage radiation exposure of workers in Korea efficiently. Therefore, in this paper, it is described the application method and assessment results of radiation exposure scenarios for workers in response to accidents assessment technology, which is one of the fundamental technologies for constructing a realistic platform to be utilized for radiation exposure prediction, diagnosis, management, and training simulations following accidents. First, the post-accident sampling after the Loss of Coolant Accident(LOCA) was selected as the accident and response scenario, and the assessment area related to this work was established. Subsequently, the structures within the assessment area were modeled using MCNP, and the radiation source of the equipment was inputted. Based on this, the radiation dose distribution in the assessment area was assessed. Afterward, considering the three principles of external radiation protection (time, distance, and shielding) detailed work scenarios were developed by varying the number of workers, the presence or absence of a shield, and the location of the shield. The radiation exposure doses received by workers were compared and analyzed for each scenario, and based on the results, the optimal accident response scenario was derived. The results of this study plan to be utilized as a fundamental technology to ensure the safety of workers through simulations targeting various reactor types and accident response scenarios in the future. Furthermore, it is expected to secure the possibility of developing a data-based ALARA decision support system for predicting radiation exposure dose at NPP sites.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.92.4-92
• /
• 2001

Estimation of the general along-track acceleration was performed in the KOMPSAT-1 orbit determination process. Several sets of the atmospheric drag and solar radiation pressure coefficients were also derived with the different spacecraft area. State vectors in the orbit determination with the different spacecraft area were compared in the time frame. The orbit prediction using the estimated coefficients was performed and compared with the orbit determination results. The orbit prediction with the different general acceleration values was also carried out for the comparison.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.06a
• /
• pp.78-83
• /
• 2000

Exterior noise generated by the aircraft induces a serious noise pollution near the airport. For the prediction of an exterior noise radiation of aircraft an empirical formula is employed to model the acoustic sources. It is shown that the fan/compressor noise is the most dominant part of the acoustic sources in all cases.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.24 no.12
• /
• pp.992-998
• /
• 2014

Acoustic radiation efficiency is one of the important factors in the prediction of underwater radiated noise of ships. A ship has much equipment to operate successful mission in a ship. Most of equipment is running simultaneously as multi-excitation and becomes the source of underwater radiated noise. In many cases of multi-excitation, phase difference between multi-excitation is not considered. Because vibration response under multi-excitation is the vector sum of each single excitation, acoustic radiation efficiency based on surface velocity field can be affected by phase of excitation. In this study, acoustic radiation efficiency of a plate on air and a stiffened cylindrical model in water under multi-excitation with phase difference is investigated.

• KIEAE Journal
• /
• v.12 no.6
• /
• pp.93-98
• /
• 2012

The concerns on high ozone concentration phenomenon is significantly growing in Seoul metropolitan area including the industry complex area, like Shiwha Banwol area. The aims of this research is the analysis of relationship between high concentrations of $O_3$ and solar radiation parameters in atmosphere. The understanding of the effects of solar radiation intensity, humidity, high air temperature on ozone concentration in a day is very useful to provide a direction for reducing of the high ozone concentration to a local government or a metropolitan government. The correlation analysis between maximum ozone concentration and various meteorological parameters in 2009 - 2011 carried out using IBM's SPSS program. The results showed that the mean correlations coefficient (R) between daily Ozone maximum and solar radiation resulted R = 0.64 during 2011. May - September in 10 air pollution stations. In case of correlations between daily ozone maximum and relative humidity showed negative correlation R = -0.61. The correlation analysis with mean air temperature during 1-3 PM resulted R = 0.29. This low correlation coefficient could be corrected by using of categorized data of ozone concentration. The daily maximum ozone concentration is more dependent on peak solar radiation and high air temperature during 1-3 PM than its simple daily maximum values. The results of this research would be used to develop the high ozone alert system around Seoul metropolitan area. This correlation analysis could be partially integrated to prediction of ozone peak concentration in connection with other methods like classification and regression tree(CART).

• Current Photovoltaic Research
• /
• v.4 no.2
• /
• pp.80-86
• /
• 2016

Since the last five years it has become a lot of solar power plants installed. However, by installing the large-scale solar power station it is not easy to predict the actual generation years. Because there are a variety of factors, such as changes daily solar radiation, temperature and humidity. If the power output can be measured accurately it predicts profits also we can measure efficiency for solar power plants precisely. Therefore, Prediction of power generation is forecast to be a useful research field. In this study, out discovering the factors that can improve the accuracy of the prediction of the photovoltaic power generation presents the means to apply them to the power generation amount prediction.