http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Nhan, Nguyen Huu,Loc, Nguyen Hoang The Korean Society of Plant Biotechnology 2018 식물생명공학회지 Vol.45 No.4
In this study, the effect of elicitors such as yeast extract (YE), methyl jasmonate (MeJA) and salicylic acid (SA) on the accumulation of eurycomanone in Eurycoma longifolia cell cultures were investigated. Suspension cells of E. longifolia was cultured in Murashige and Skoog (MS) medium supplemented with 30 g/L sucrose, 1.25 mg/L naphthaleneacetic acid (NAA) and 1 mg/L kinetin at a shaking speed of 120 rpm. Elicitors were added in the culture at different concentrations and times to stimulate eurycomanone accumulation in the Eurycoma longifolia cells. Eurycomanone content was determined by HPLC with a C18 column, flow rate of 0.8 mL/min, run time of 17.5 min, and a detector wavelength of 254 nm. The stationary phase was silica gel and the mobile phase was acetonitrile: $H_2O$. Non-elicited cells were used as the control. The study showed the effect of different elicitor concentrations, YE at 200 mg/L, MeJA at $20{\mu}M$ and SA at $20{\mu}M$ stimulated high production of eurycomanone. In which, treatment of $20{\mu}M$ MeJA after 4 days of culture resulted in the highest accumulation of this compound (17.36 mg/g dry weight), approximately 10-fold higher than that of untreated cells (1.70 mg/g dry weight).
A Simple SVM Technique to Eliminate Common-mode Voltage for Matrix Converters
Huu-Nhan Nguyen,Tuyen D. Nguyen,Thanh-Luan Nguyen,Hong-Hee Lee 전력전자학회 2019 ICPE(ISPE)논문집 Vol.2019 No.5
In a matrix-converter (MC) system, commonmode voltage (CMV) causes damage to the motor bearings and other equipment. When the MC is controlled by means of space vector modulation (SVM), zero-CMV can be achieved by using only rotating-vector states to drive the MC. However, as the computation becomes quite complicated, it is not easy to apply the rotating-vector states in practice. This paper presents a simple SVM technique using rotating-vector states to eliminate the CMV for MCs. Based on the conventional SVM method, the proposed SVM technique achieves the zero CMV for MCs within a few simple steps by considering the relationship between active-vector and rotating-vector states. The performance and effectiveness of the proposed technique are verified by simulation results.
Effects of Hyper-parameters and Dataset on CNN Training
Huu Nhan Nguyen,Chanho Lee 한국전기전자학회 2018 전기전자학회논문지 Vol.22 No.1
The purpose of training a convolutional neural network (CNN) is to obtain weight factors that give high classification accuracies. The initial values of hyper-parameters affect the training results, and it is important to train a CNN with a suitable hyper-parameter set of a learning rate, a batch size, the initialization of weight factors, and an optimizer. We investigate the effects of a single hyper-parameter while others are fixed in order to obtain a hyper-parameter set that gives higher classification accuracies and requires shorter training time using a proposed VGG-like CNN for training since the VGG is widely used. The CNN is trained for four datasets of CIFAR10, CIFAR100, GTSRB and DSDL-DB. The effects of the normalization and the data transformation for datasets are also investigated, and a training scheme using merged datasets is proposed.
A DSVM Method for Matrix Converters to Suppress Common-mode Voltage With Reduced Switching Losses
Huu-Nhan Nguyen,Hong-Hee Lee Institute of Electrical and Electronics Engineers 2016 IEEE transactions on power electronics Vol. No.
<P>This paper proposes a direct space-vector modulation (DSVM) method that can reduce the common-mode voltage (CMV) as well as the switching losses for matrix converters in a high voltage transfer ratio. Even though the previous DSVM improvements can reduce the CMV peak value to 42%, compared to the conventional DSVM method, they incur high switching losses. The DSVM method proposed in this paper provides the same CMV peak value as previous methods with lower switching losses. The switching losses are theoretically analyzed through the accumulated switched voltage in a switching period and compared to those of previous methods. Simulation and experimental results are given to verify the input/output waveforms, CMV reduction, and lower switching losses of the proposed DSVM method.</P>
A Modulation Scheme for Matrix Converters With Perfect Zero Common-Mode Voltage
Huu-Nhan Nguyen,Hong-Hee Lee Institute of Electrical and Electronics Engineers 2016 IEEE transactions on power electronics Vol. No.
<P>This paper presents a novel space vector modulation (SVM) method combined with a modified four-step commutation to achieve a perfect zero common-mode voltage (CMV) for matrix converters (MCs). The new SVM method was developed by using only rotating input voltage vectors, which results in the CMV for MCs being zero. However, it causes some noise in practical applications that drive MCs with traditional four-step commutation. In order to solve this noise problem, amodified four-step commutation is also proposed to perfectly eliminate the CMV for MCs without any noise. In the modified four-step commutation, a delay time is designed to deal with misjudgment of input voltages. The proposed SVM method and four-step commutation are easily implemented via software and achieve good performance in input/output current waveforms. Prototype experiments were carried out to evaluate the performance of the proposed method.</P>
Nguyen Huu Nhan,Truong Thi Nhan,Le Thi Phuong Ngoc,Nguyen Thanh Long 경남대학교 기초과학연구소 2021 Nonlinear Functional Analysis and Applications Vol.26 No.1
In this paper, we investigate an initial boundary value problem for a nonlinear pseudoparabolic equation. At first, by applying the Faedo-Galerkin, we prove local existence and uniqueness results. Next, by constructing Lyapunov functional, we establish a sufficient condition to obtain the global existence and exponential decay of weak solutions.
Effects of Hyper-parameters and Dataset on CNN Training
Nguyen, Huu Nhan,Lee, Chanho Institute of Korean Electrical and Electronics Eng 2018 전기전자학회논문지 Vol.22 No.1
The purpose of training a convolutional neural network (CNN) is to obtain weight factors that give high classification accuracies. The initial values of hyper-parameters affect the training results, and it is important to train a CNN with a suitable hyper-parameter set of a learning rate, a batch size, the initialization of weight factors, and an optimizer. We investigate the effects of a single hyper-parameter while others are fixed in order to obtain a hyper-parameter set that gives higher classification accuracies and requires shorter training time using a proposed VGG-like CNN for training since the VGG is widely used. The CNN is trained for four datasets of CIFAR10, CIFAR100, GTSRB and DSDL-DB. The effects of the normalization and the data transformation for datasets are also investigated, and a training scheme using merged datasets is proposed.
An Effective SVM Method for Matrix Converters With a Superior Output Performance
Nguyen, Huu-Nhan,Lee, Hong-Hee Institute of Electrical and Electronics Engineers 2018 IEEE transactions on industrial electronics Vol.65 No.9
<P>This paper presents the development of an effective space vector modulation (SVM) regarding matrix converters (MCs) for the full range of the voltage transfer ratio (VTR), whereby a superior output performance was achieved. The proposed method uses all of the vectors adjacent to the reference output-voltage vector in one sector to build a unique general switching pattern to drive the MC. For the purpose of pursuing the high output performance of MCs, the proposed switching pattern can be applied to low and high VTRs with a unique modulation strategy owing to generalized equations. In addition, an output current-ripple-minimization technique is also accomplished through a design of the duty cycles of zero vectors that enhances the output performance for the MC. With the unique switching pattern, the proposed SVM method can be easily implemented. The experiment results are given to confirm the effectiveness of the proposed method.</P>