• Smart Structures and Systems
• /
• v.6 no.1
• /
• pp.1-16
• /
• 2010

This paper presents a novel smart MEMS accelerometer which employs a hybrid control algorithm and an estimator. This scheme is realized by adding a sliding-mode controller to a conventional PID closed loop system to achieve higher stability and higher dynamic range and to prevent pull-in phenomena by preventing finger displacement from passing a maximum preset value as well as adding an adaptive nonlinear observer to a conventional PID closed loop system. This estimator is used for online estimation of the parameter variations for MEMS accelerometers and gives the capability of self testing to the system. The analysis of convergence and resolution show that while the proposed control scheme satisfies these criteria it also keeps resolution performance better than what is normally obtained in conventional PID controllers. The performance of the proposed hybrid controller investigated here is validated by computer simulation.

• Journal of Korea Multimedia Society
• /
• v.21 no.8
• /
• pp.969-981
• /
• 2018

Recent advances in machine learning approaches such as deep neural network and reinforcement learning offer significant performance improvements in generating detailed and varied motions in physically simulated virtual environments. The optimization methods are highly attractive because it allows for less understanding of underlying physics or mechanisms even for high-dimensional subtle control problems. In this paper, we propose an efficient learning method for stochastic policy represented as deep neural networks so that agent can generate various energetic motions adaptively to the changes of tasks and states without losing interactivity and robustness. This strategy could be realized by our novel trajectory search method motivated by the trust region policy optimization method. Our value-based trajectory smoothing technique finds stably learnable trajectories without consulting neural network responses directly. This policy is set as a trust region of the artificial neural network, so that it can learn the desired motion quickly.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.10 no.5
• /
• pp.436-442
• /
• 2005

This paper proposes an estimation algorithm to find the moment of inertia, which is essential to design high performance controller for motor drive system. The algorithm finds the moment of inertia observing the position error signal, which contains an error information of moment of inertia, generated by speed observer. Moreover, the proposed algorithm is easily realized in the observer-based speed detection method. The simulation and experimental results are also presented to confirm the performance of moment of inertia estimation method, which shows that the moment of inertia converges to the actual value within several seconds. The speed control responses and the designed speed controller performance match well.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.4
• /
• pp.387-397
• /
• 1992

The spatial domain design of 2-dimensional separable denominator digital filters(SDDF) based on the reduced dimensional decomposition can be realized when the given 2-D impulse response specifications are decomposed into a pair of 1-D specifications via singular value decompositions(SVD). Because of use of the balaned approximation and equivalent transform as 1-D design algorithm, 2-D design algorithm retains the advantage that is numerically stable and can minimize quantization errors. In this paper in order to analyze and reduce these errors, minimum comfficient quantization realization is directly derived from impulse response specification. And using the equivalent trans form relation between mininum coefficient quantization error and minimum roundoff error realizations, we optimally realize a SDDF. This algorithm is analyzed by the simulation, which shows that it is superior to direct or balanced realization in quantization errors.

• Proceedings of the KIEE Conference
• /
• 1999.11c
• /
• pp.742-744
• /
• 1999

In this paper, a new precharge type PFD for fast locking time of PLL is suggested. It is realized by inserting NMOS transistor and inverter into the precharge part of PFD for isolating the reset of the Up signal from the feedback signal. The new precharge type PFD generates the Up signal while the feedback signal is fixed at a high level. Therefore the new PFD output is increased than the conventional precharge type PFD output. As a result of the increased PFD output, fast locking of PLLs is achieved. Additionally, with control the falling time of the inverter, the dead-zone is reduced and the jitter characteristics are improved. The whole characteristics of PFD and PLL are simulated by using HSPICE. Simulation results show that the dead-zone is 20ps and the locking time of PLL using the new PFD is 38ns at the 350MHz frequency of referecne signal. This value is quite small compared with conventional PFD.

• Journal of the Korean Solar Energy Society
• /
• v.21 no.2
• /
• pp.27-34
• /
• 2001

The greenhouse heating system using solar energy has been realized in the protective agriculture in this study in order to analyse the thermal energy characteristics of the system the effects of ambient air temperature, solar radiation, relative humidities and water content of ambient air on the greenhouse air temperature were investigated through computer simulation experimental analysis for validation of the simulation. The results from this study are summarized as follows: 1) The expected values of inside air temperature for the system solar energy were very much close to the experimental values. 2) In the system using solar energy, the expected values of daytime surface temperature of soil by computer simulation were very much similar to the measured values, but those of nighttime were higher than the measured value by almost $2.5^{\circ}C$. 3) Heat loss of daytime was found to be larger than that of night time as much as 2.0 to 4.2 times for the system using solar energy. 4) In the system using solar energy. while the ambient air temperature varied between $-7^{\circ}C$ and $-3.8^{\circ}C$, the temperature of the inside air was maintained between $0^{\circ}C$ and $22^{\circ}C$. 5) At the minimum ambient temperature of $-7^{\circ}C$, the temperature of the inside air was $0^{\circ}C$.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2011.05a
• /
• pp.40.2-40.2
• /
• 2011

As the old saying 'nothing is complete unless you put it in final shape', although nanosheets (NSs) are a promising functional building block for various electrochemical applications, their true value cannot be realized until they are well woven into electrical conducting materials. As an effort to determine their ideal shape, in this study, a unique manufacturing route to build a layer-by-layer (LBL) structure of two-dimensionally ordered, free-standing ${\beta}$-nickel hydroxide nanosheets (${\beta}$-NHNSs) that are fully electrically addressed with single-wall carbon nanotube fabrics was demonstrated, and its capabilities were verified through a comparative study on the differences between a simple bulky and LBL-structured electrochemical cathode, representing two extreme cases. The LBL-structured cathode showed a discharging current peak that was 25 times larger than the bulky structured one measured in cyclic voltammetry, which implies that the LBL structure is near an ideal electrode configuration for NS-based electrochemical applications.

• Korean Journal of Materials Research
• /
• v.30 no.4
• /
• pp.155-159
• /
• 2020

Nanoporous silica aerogel insulation material is both lightweight and efficient; it has important value in the fields of aerospace, petrochemicals, electric metallurgy, shipbuilding, precision instruments, and so on. A theoretical calculation model and experimental measurement of equivalent thermal conductivity for nanoporous silica aerogel insulation material are introduced in this paper. The heat transfer characteristics and thermal insulation principle of aerogel nano are analyzed. The methods of SiO₂ aerogel production are compared. The pressure range of SiO₂ aerogel is 1Pa-atmospheric pressure; the temperature range is room temperature-900K. The pore diameter range of particle SiO₂ aerogel is about 5 to 100 nm, and the average pore diameter range of about 20 ~ 40 nm. These results show that experimental measurements are in good agreement with theoretical calculation values. For nanoporous silica aerogel insulation material, the heat transfer calculation method suitable for nanotechnology can precisely calculate the equivalent thermal conductivity of aerogel nano insulation materials. The network structure is the reason why the thermal conductivity of the aerogel is very low. Heat transfer of materials is mainly realized by convection, radiation, and heat transfer. Therefore, the thermal conductivity of the heat transfer path in aerogel can be reduced by nanotechnology.

• Proceedings of the Korean Magnestics Society Conference
• /
• 2009.12a
• /
• pp.190-190
• /
• 2009

We report the epitaxial growth of MgO and CoFe/MgO on Ge (001) substrates using molecular beam epitaxy. It was found that the epitaxial growth of a MgO film on Ge could be realized at a low growth temperature of $125{\pm}5^{\circ}C$ and the MgO matches the Ge with a cell ratio of $\sqrt{2}$:1 which renders MgO rotated by $45^{\circ}$ relative to Ge. In-situ and ex-situ structural characterizations reveal the epitaxial crystal growth of bcc CoFe/MgO on Ge with the in-plane crystallographic relationship of CoFe(001)[100] || MgO(001)[110] || Ge(001)[100], exhibiting sharp interfaces in the (001) matching planes. The saturation magnetization of the sample is $1430{\pm}20$ emu/cc, which is comparable to the value of bulk CoFe.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2008.05a
• /
• pp.25-30
• /
• 2008

This paper presents on research findings of how Visual C++ program can be used to generate codes capable of performing ramjet engine simulation an arbitrary ramjet model will be considered for which generated output values will be compared with those from a commercial program GASTURB 9 Several governing thermodynamic equations will first be discussed in order that we understand the fundamental idea behind values printed out on the GUI. The program is designed that it generates its station input value. Similar results were realized compared to those produced by gasturb 9.