• Journal of Biosystems Engineering
• /
• v.20 no.2
• /
• pp.133-140
• /
• 1995

The model of Poisson's ratio of the fruits was developed on the basis that the cylindrical fruits specimen became the barrel shape when it was being compressed. The model of the corrected creep compliance of the fruits was developed under considering the developed model of Poisson's ratio. Both of the Poisson's ratio and the corrected creep compliance of the samples showed the nonlinear viscoelastic behavior. Those models were a similar form, but their coefficients of the model were different, and these behaviors of the samples were well described by the nonlinear model as a function of the initial stress and time. Effects of storage condition and period on the Poisson's ratio of the samples were investigated, and comparisons between the corrected and the uncorrected creep compliance of the samples were made.

• Journal of Soil and Groundwater Environment
• /
• v.15 no.6
• /
• pp.91-98
• /
• 2010

The effective thermal conductivity of porous materials is usually determined by porosity, water content, and the conductivity of the matrix. In addition, it is also affected by the internal structure of the materials such as the size, arrangement, and connectivity of the matrix-forming grains. Based on the structural models for multi-phase materials, thermal conductivities of soils and sands measured with varying the water content were analyzed. Thermal conductivities of dry samples were likely to fall in the region between the Maxwell-Eucken model with air as the continuous phase and the matrix as the dispersed phase ($ME_{air}$) and the co-continuous (CC) model. However, water-saturated samples moved down to the region between the $ME_{wat}$ model and the series model. The predictive inconsistency of the structural models for dry and water-saturated samples may be caused by the increase of porosity for water-saturated samples, which leads to decrease of connectivity among the grains of matrix. In cases of variably saturated samples with a uniform grain size, the thermal conductivity showed progressive changes of the structural models from the $ME_{air}$ model to the $ME_{wat}$ model depending on the water content. Especially, an abrupt increase found in 0-20% of the water content, showing transition from the $ME_{air}$ model to the CC model, can be attributed to change of water from the dispersed to continuous phase. On the contrary, the undisturbed soil samples with various sizes of grains showed a gradual increase of conductivity during the transition from the $ME_{air}$ model to the CC model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.12
• /
• pp.5464-5484
• /
• 2016

Few samples face recognition has become a highly challenging task due to the limitation of available labeled samples. As two popular paradigms in face image representation, sparse component analysis is highly robust while parts-based paradigm is particularly flexible. In this paper, we propose a probabilistic generative model to incorporate the strengths of the two paradigms for face representation. This model finds a common spatial partition for given images and simultaneously learns a sparse component analysis model for each part of the partition. The two procedures are built into a probabilistic generative model. Then we derive the score function (i.e. feature mapping) from the generative score space. A similarity measure is defined over the derived score function for few samples face recognition. This model is driven by data and specifically good at representing face images. The derived generative score function and similarity measure encode information hidden in the data distribution. To validate the effectiveness of the proposed method, we perform few samples face recognition on two face datasets. The results show its advantages.

• Nuclear Engineering and Technology
• /
• v.55 no.6
• /
• pp.2096-2106
• /
• 2023

Rotating machinery is widely applied in important equipment of nuclear power plants (NPPs), such as pumps and valves. The research on intelligent fault diagnosis of rotating machinery is crucial to ensure the safe operation of related equipment in NPPs. However, in practical applications, data-driven fault diagnosis faces the problem of small and imbalanced samples, resulting in low model training efficiency and poor generalization performance. Therefore, a deep convolutional conditional generative adversarial network (DCCGAN) is constructed to mitigate the impact of imbalanced samples on fault diagnosis. First, a conditional generative adversarial model is designed based on convolutional neural networks to effectively augment imbalanced samples. The original sample features can be effectively extracted by the model based on conditional generative adversarial strategy and appropriate number of filters. In addition, high-quality generated samples are ensured through the visualization of model training process and samples features. Then, a deep convolutional neural network (DCNN) is designed to extract features of mixed samples and implement intelligent fault diagnosis. Finally, based on multi-fault experimental data of motor and bearing, the performance of DCCGAN model for data augmentation and intelligent fault diagnosis is verified. The proposed method effectively alleviates the problem of imbalanced samples, and shows its application value in intelligent fault diagnosis of actual NPPs.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1264-1264
• /
• 2001

The affect of surface temperatures of fruits on spectrum which measures actual sugar contents was observed. PLSR was applied to develop the sugar content evaluation system that was not affected by temperature. The reflected spectrum was used at the wavelengths of 654 and 1052nm with the separation distance of 2.5nm. To increase the conformance of a model using unknown samples, let the minimum value of PRESS be an optimum factor. 71 Shingo pears stored in a refrigerator were left in a room temperature for a while and these temperatures and reflected spectrums were measured. Reflected spectrums were measured at the wavelengths of 654 and 1052nm, 3 samples in one second. To measure these at different temperatures, the experiment was repeated hourly and four times. Starting temperatures of 2-3 were increased up to 17. The total number of measured spectrum was 284. To develop a sugar content evaluation system model using measured reflected spectrum, three groups of samples were considered. First group had 51 samples at 14 and second group had 141 samples with lower or higher temperatures than 14. Third group had 155 samples with well distributed temperatures. Other samples were used as validations to ensure the conformance. Measuring the sugar contents of samples with surface temperatures other than 14 were difficult with PLS model I, developed by using a sample temperature of 14. If the sugar contents were compensated using samples' temperatures, results of prediction would be close to the expected results and it would be one of the most important factors to develop this system. PLS models I and II could compensate the temperature but the precision would not come up to the standard. High precision was expected by using samples with wide ranges of temperatures and sugar contents. Both models showed the possibility of an improvement of a sugar content evaluation system disregarding the temperature. For practical use of a system, selecting samples should be done carefully to reduce the effect of the temperature.

• Journal of the Korean Society of Safety
• /
• v.20 no.3 s.71
• /
• pp.98-104
• /
• 2005

This study experimentally and theoretically examines the fire characteristics of 100- by 100- by 50-mm samples of flame retardant treated Douglas fir. Samples were exposed to a range of incident heat fluxes 10 to $50kW/m^2$. The time to ignition measurements obtained from the cone heater were used to derive characteristic properties of the materials. A one-dimensional integral model has been used to predict the, time to ignition, critical heat flux and ignition temperature of samples. Ignition data and best-fit curves confirm ${{\dot{q}}_i}^{'}{\rightarrow}{{\dot{q}}_{cr}^{'}\;then\;t_{ig}{\rightarrow}{\infty}$ and when ${{\dot{q}}_i}^'{\gg}{{\dot{q}}_{cr}^'\;then\;t_{ig}{\rightarrow}0$. And Ignition of flame retardant treated samples occurred not at incident heat flux of bellow $10kW/m^2.$. By a one-dimensional integral model, the critical heat flux of each samples was predicted $10.21kW/m^2,\;11.82kW/m^2,\;and\;14.16kW/m^2$ for the D-N, D-F2, and D-F4, respectively. In ignition temperature of each samples, flame retardant treated samples were measured high about $50^{\circ}C$ than non-treated samples. Water-soluble flame retardant used in this study finds out more effect in delay of time to ignition when incident heat flux is low than high.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.185-189
• /
• 2006

Time series analysis is a frequently effective method of constructing model and prediction in data processing of deformation monitoring. The monitoring data sample must to be as more as possible and time intervals are equal roughly so as to construct time series model accurately and achieve reliable prediction. But in the project practice of GPS deformation monitoring, the monitoring data sample can't be obtained too much and time intervals are not equal because of being restricted by all kinds of factors, and it contains many variates in the deformation model moreover. It is very important to study the data processing method for small samples and multi-variates time series in GPS deformation monitoring. A new method of establishing small samples and multi-variates deformation model and prediction model are put forward so as to resolve contradiction of small samples and multi-variates encountered in constructing deformation model and improve formerly data processing method of deformation monitoring. Based on the system theory, a deformation body is regarded as a whole organism; a time-dependence linear system model and a time-dependence bilinear system model are established. The dynamic parameters estimation is derived by means of prediction fit and least information distribution criteria. The final example demonstrates the validity and practice of this method.

• Journal of Biosystems Engineering
• /
• v.29 no.4
• /
• pp.335-340
• /
• 2004

This study was conducted to evaluate the internal quality of Korean red ginseng using VIS/NIR transmittance spectra. To classify the internal qualities, partial least squares(PLS) regression was conducted. The main results are as follows: To develop the PLS model, several wave bands were divided and incorporated into the model. Among the bands, the wavelength range of 550-1,020nm, excluded noise signal, showed the best evaluation results. Effect of step size on the performance of quality evaluation showed optimal at 15 steps. In order to enhance the accuracy of quality evaluation, the abnormal spectrum shape was considered first and then the PLS model was applied. Among the 150 samples, 12 samples were evaluated by the spectrum shape. In this study, to develop the optimal PLS regression model, among the 150 samples, 138 samples was used with exception of 12 samples which could evaluate the spectrum shape. The result of quality evaluation was promising as SEC and correlation coefficient were 1.09 and 0.967, respectively, and SEP and correlation coefficient were 1.04 and 0.958, respectively.

• Journal of Broadcast Engineering
• /
• v.23 no.2
• /
• pp.186-196
• /
• 2018

In this paper, we propose an on-line tracking method using convolutional neural network (CNN) for tracking object. It is well known that a large number of training samples are needed to train the model offline. To solve this problem, we use an untrained model and update the model by collecting training samples online directly from the test sequences. While conventional methods have been used to learn models by training samples offline, we demonstrate that a small group of samples are sufficient for online object tracking. In addition, we define a loss function containing color information, and prevent the model from being trained by wrong training samples. Experiments validate that tracking performance is equivalent to four comparative methods or outperforms them.

• Journal of the Korea Safety Management & Science
• /
• v.9 no.3
• /
• pp.165-171
• /
• 2007

Dependent models in quality statistics are classified as serially autocorrelated model, multivariate model and dependent sample model. Dependent sample model is most efficient in time and cost to obtain samples among the above models. This paper proposes to implement parametric and nonparametric models into production system depended on demand pattern. Nonparametric models have distribution free and asymptotic distribution free techniques. Quality statistical models are classified into two categories ; the number of dependent sample and the type of data. The type of data consists of nominal, ordinal, interval and ratio data. The number of dependent sample divides into 2 samples and more than 3 samples.