• Journal of Korea Water Resources Association
• /
• v.51 no.5
• /
• pp.439-449
• /
• 2018

Many potential sources of bias are used in several steps of the radar-rainfall estimation process because the hydrological and meteorological radars measure the rainfall amount indirectly. Previous studies on radar-rainfall uncertainties were performed to reduce the uncertainty of each step by using bias correction methods in the quantitative radar-rainfall estimation process. However, these studies do not provide comprehensive uncertainty for the entire process and the relative ratios of uncertainty between each step. Consequently, in this study, a suitable approach is proposed that can quantify the uncertainties at each step of the quantitative radar-rainfall estimation process and show the uncertainty propagation through the entire process. First, it is proposed that, in the suitable approach, the new concept can present the initial and final uncertainties, variation of the uncertainty as well as the relative ratio of uncertainty at each step. Second, the Maximum Entropy Method (MEM) and Uncertainty Delta Method (UDM) were applied to quantify the uncertainty and analyze the uncertainty propagation for the entire process. Third, for the uncertainty quantification of radar-rainfall estimation at each step, two quality control algorithms, two radar-rainfall estimation relations, and two bias correction methods as post-processing through the radar-rainfall estimation process in 18 rainfall cases in 2012. For the proposed approach, in the MEM results, the final uncertainty (from post-processing bias correction method step: ME = 3.81) was smaller than the initial uncertainty (from quality control step: ME = 4.28) and, in the UDM results, the initial uncertainty (UDM = 5.33) was greater than the final uncertainty (UDM = 4.75). However uncertainty of the radar-rainfall estimation step was greater because of the use of an unsuitable relation. Furthermore, it was also determined in this study that selecting the appropriate method for each stage would gradually reduce the uncertainty at each step. Therefore, the results indicate that this new approach can significantly quantify uncertainty in the radar-rainfall estimation process and contribute to more accurate estimates of radar rainfall.

• ETRI Journal
• /
• v.15 no.3
• /
• pp.27-34
• /
• 1994

There have been traditional approaches to model radio propagation path loss mechanism both theoretically ad empirically. Theoretical approach is simple to explain and effective in certain cases. Empirical approach accommodates the terrain configuration and distance between base station and mobile unit along the propagation path only. In other words, it does not accommodate natural terrain configuration over a specific area. In this paper, we propose a spatial prediction technique for the mobile radio propagation path loss accommodating complete natural terrain configuration over a specific area. Statistical uncertainty analysis is also considered.

• Journal of Mechanical Science and Technology
• /
• v.18 no.4
• /
• pp.689-698
• /
• 2004

Uncertainties generated from the individual measured variables have an influence on the uncertainty of the experimental result through a data reduction equation. In this study, a performance test of a single stage axial type turbine is conducted, and total-to-total efficiencies are measured at the various off-design points In the low pressure and cold state. Based on an experimental apparatus, a data reduction equation for turbine efficiency is formulated and six measured variables are selected. Codes are written to calculate the efficiency, the uncertainty of the efficiency, and the sensitivity of the efficiency uncertainty by each of the measured quantities. The influence of each measured variable on the experimental result is figured out. Results show that the largest uncertainty magnification factor (UMF) value is obtained by the inlet total pressure among the six measured variables, and its value is always greater than one. The UMF values of the inlet total temperature, the torque, and the RPM are always one. The uncertainty percentage contribution (UPC) of the RPM shows th, lowest influence on the uncertainty of the turbine efficiency, but the UPC of the torque has the largest influence to the result among the measured variables. These results are applied to find the correct direction for meeting an uncertainty requirement of the experimental result in the planning or development Phase of experiment, and also to offer ideas for preparing a measurement system in the planning phase.

• Structural Engineering and Mechanics
• /
• v.53 no.4
• /
• pp.819-831
• /
• 2015

In this paper, we propose a new method for taking into account uncertainties based on the projection on polynomial chaos. The new approach is used to determine the dynamic response of a spur gear system with uncertainty associated to gear system parameters and this uncertainty must be considered in the analysis of the dynamic behavior of this system. The simulation results are obtained by the polynomial chaos approach for dynamic analysis under uncertainty. The proposed method is an efficient probabilistic tool for uncertainty propagation. It was found to be an interesting alternative to the parametric studies. The polynomial chaos results are compared with Monte Carlo simulations.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.27 no.8
• /
• pp.1071-1080
• /
• 2003

A national standard system was developed in order to calibrate and test the oil flowmeters for the petroleum field. A stop valve and a gyroscopic weighing scale were employed for the primary standard of the flow quantity. It is operated by the standing start and finish mode and the static weighing method. The model equation for uncertainty evaluation was based on the calibration principle of standard system. The sources of the uncertainties were quantified and combined according to the GUM(Guide to the Expression of Uncertainty in Measurement). It was found that the standard system had the relative expanded uncertainty of 0.04 % in the range of 18 - 350 ㎥/ｈ. According to the uncertainty budget, the uncertainties of the fluid density and the volume of pipeline, which were temperature dependent, contributed 92% of final uncertainty in the oil flow standard system.

• Nuclear Engineering and Technology
• /
• v.36 no.1
• /
• pp.83-103
• /
• 2004

The methodological framework of the Level 2 PSA appears to be currently standardized in a formalized fashion, but there have been different opinions on the way the sources of uncertainty are characterized and treated. This is primarily because the Level 2 PSA deals with complex phenomenological processes that are deterministic in nature rather than random processes, and there are no probabilistic models characterizing them clearly. As a result, the probabilistic quantification of the Level 2 PSA CET / APET is often subjected to two sources of uncertainty: (a) incomplete modeling of accident pathways or different predictions for the behavior of phenomenological events and (b) expert-to-expert variation in estimating the occurrence probability of phenomenological events. While a clear definition of the two sources of uncertainty involved in the Level 2 PSA makes it possible to treat an uncertainty in a consistent manner, careless application of these different sources of uncertainty may produce different conclusions in the decision-making process. The primary purpose of this paper is to characterize typical sources of uncertainty that would often be addressed in the Level 2 PSA and to provide a formal guidance for quantifying their impacts on the PSA Level 2 risk results. An additional purpose of this paper is to give a formal approach on how to combine random uncertainties addressed in the Level 1 PSA with subjectivistic uncertainties addressed in the Level 2 PSA.

• International Journal of Precision Engineering and Manufacturing
• /
• v.5 no.4
• /
• pp.61-68
• /
• 2004

Necessity to address engineering system uncertainties in design processes has long been acknowledged. To obtain quality of product, a safety factor is traditionally used by many design engineers due to its easy of use and comprehension. However, the safety factor approach often yields either conservative or unreliable designs, since it ignores the type of probability distribution and the mechanism of uncertainty propagation from the input to the output. For a consistent reliability-based design, two fundamental issues must be investigated thoroughly. First, the design-decision process that clearly identifies a mechanism of uncertainty propagation under system uncertainties needs to be developed, which must be an efficient and accurate process. To identify the mechanism more effectively, an adaptive probability analysis is proposed by adaptively setting probability levels through a posteriori error estimation. The second is to develop the design process that not only yields a high quality design but also a cost-effective optimum design from manufacturing point of view. As a result, a response surface methodology is specially developed for RBDO, thus enhancing numerical challenges of efficiency and complicatedness. Side crashworthiness application is used to demonstrate the integrated design process for product and manufacturing process design.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.33 no.8
• /
• pp.707-719
• /
• 2009

The variables relating to the fatigue behavior have uncertainty and are random. The fatigue crack propagation is, thus, stochastic in nature. In this study, fatigue experiments are performed on the specimen of the magnesium alloy AZ31. The data of the fatigue life are scattered even in the same experimental condition. It is necessary to determine the probability distribution of the fatigue crack propagation life for the reliability analysis as well as the design and maintenance of structural components. Therefore the statistics and the probability distribution for the fatigue crack propagation life are investigated and the best fit probability distribution of that is proposed in this paper.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.54 no.6
• /
• pp.77-87
• /
• 2012

Recently, Load and Resistance Factor Design (LRFD) based on reliability analysis has become a global trend for economical and rational design. In order to implement the LRFD, quantification of uncertainty for load and resistance should be done. The reliability of result relies on input variable, and therefore, it is important to obtain exact uncertainty properties of load and resistance. Since soil stress is the main reason causing the settlement or deformation of ground and load on the underground structure, it is essential to clarify the uncertainty of soil stress distribution for accurately predict the uncertainty of load in LRFD. In this study, laboratory model test on silty sand bed under probabilistic load is performed to observe propagation of upper load uncertainty. The results show that the coefficient of variation (COV) of soil stress are varied depending on location due to non-linear relationship between upper load increment and soil pressure increment. In addition, when the load uncertainty is transmitted through ground, COV is decreased by damping effect.

• Nuclear Engineering and Technology
• /
• v.52 no.6
• /
• pp.1148-1155
• /
• 2020

This work presents two different methods for quantifying and propagating the uncertainty associated with fuel composition at end of life for cask criticality calculations. The first approach, the computational approach uses parametric uncertainty including those associated with nuclear data, fuel geometry, material composition, and plant operation to perform forward depletion on Monte-Carlo sampled inputs. These uncertainties are based on experimental and prior experience in criticality safety. The second approach, the data-driven approach relies on using radiochemcial assay data to derive code bias information. The code bias data is used to perturb the isotopic inventory in the data-driven approach. For both approaches, the uncertainty in k_eff for the cask is propagated by performing forward criticality calculations on sampled inputs using the distributions obtained from each approach. It is found that the data driven approach yielded a higher uncertainty than the computational approach by about 500 pcm. An exploration is also done to see if considering correlation between isotopes at end of life affects k_eff uncertainty, and the results demonstrate an effect of about 100 pcm.