• The Journal of the Convergence on Culture Technology
• /
• v.6 no.2
• /
• pp.509-514
• /
• 2020

The conventional TTS system consists of several modules, including text preprocessing, parsing analysis, grapheme-to-phoneme conversion, boundary analysis, prosody control, acoustic feature generation by acoustic model, and synthesized speech generation. But TTS system with deep learning is composed of Text2Mel process that generates spectrogram from text, and vocoder that synthesizes speech signals from spectrogram. In this paper, for the optimal Korean TTS system construction we apply Tacotron2 to Tex2Mel process, and as a vocoder we introduce the methods such as WaveNet, WaveRNN, and WaveGlow, and implement them to verify and compare their performance. Experimental results show that WaveNet has the highest MOS and the trained model is hundreds of megabytes in size, but the synthesis time is about 50 times the real time. WaveRNN shows MOS performance similar to that of WaveNet and the model size is several tens of megabytes, but this method also cannot be processed in real time. WaveGlow can handle real-time processing, but the model is several GB in size and MOS is the worst of the three vocoders. From the results of this study, the reference criteria for selecting the appropriate method according to the hardware environment in the field of applying the TTS system are presented in this paper.

• Journal of the Korean Solar Energy Society
• /
• v.27 no.3
• /
• pp.37-44
• /
• 2007

For the past few years, the concern for clean energy has been greatly increased. Ocean Thermal Energy Conversion(OTEC) power plants are studied as a viable option for the supply of clean energy. In this paper, the thermodynamic performance of OTEC cycle was examined. Computer simulation programs were developed under the same condition and various working fluids for closed Rankine cycle, regeneration cycle, Kalina cycle, open cycle and hybrid cycle. The results show that the regeneration cycle using R125 showed a 0.17 to 1.56% increase in energy efficiency, and simple Rankine cycle can generate electricity when the difference in warm and cold sea water inlet temperatures are greater than $15^{\circ}C$. Also, the cycle efficiency of OTEC power plant using the condenser effluent from nuclear power plant instead of the surface water increased about 2%.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.5
• /
• pp.825-833
• /
• 2020

Currently, the K-Water uses various valves that can be remotely controlled for optimal water management. Valve system fault can be classified into rotor defects, stator defects, bearing defects, and gear defects of induction motors. If the valve cannot be operated due to a gear fault, the water management operation can be greatly affected. For effective water management, there is an urgent need for preemptive repairs to determine whether gear is damaged through failure prediction diagnosis.. Recently, deep learning algorithms are being applied for valve failure diagnosis. However, the method currently applied has a disadvantage of attaching a vibration sensor to the valve. In this paper, propose a new algorithm to determine whether a fault exists using a convolutional neural network (CNN) based on the voltage and current information of the valve without additional sensor mounting. In particular, a normalized Lisasjous diagram was used to maximize the fault classification performance in the CNN-based diagnostic system.

• Journal of Advanced Marine Engineering and Technology
• /
• v.36 no.8
• /
• pp.1151-1157
• /
• 2012

Ocean Thermal Energy Conversion(OTEC) technology is one of the new and renewable energy that utilizes the natural temperature gradient that exists in the tropical ocean between warm surface water and the deep cold water, to generate electricity. The selection of working fluid and the OTEC cycle greatly influence the effect on the system operation, and it's energy efficiency and impacts on the environment. Working fluids of the OTEC are ammonia, R22, R407C, and R410A. In this paper, we compared boiling pressure to optimize OTEC system at $25^{\circ}C$. Also, this paper showed net-power and efficiency according to working fluids for closed cycle and regeneration cycle.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.16 no.6
• /
• pp.825-831
• /
• 2016

In this paper, a transimpedance amplifier based on a new DC offset cancellation (DCOC) method is proposed for WCDMA/LTE applications. The proposed method applies a sample and hold mechanism to the conventional DCOC method with a DC feedback loop. It prevents the removal of information around the DC, so it avoids signal-to-noise ratio degradation. It also reduces area and power consumption. It was designed in a $0.13{\mu}m$ deep n-well CMOS technology and drew a maximum current of 1.58 mA from a 1.2 V supply voltage. It showed a transimpedance gain of $80dB{\Omega}$, an input-referred noise current lower than 0.9 pA/${\surd}$Hz, an out-of-band input-referred 3rd-order intercept point more than 9.5 dBm, and an output DC offset lower than 10 mV. Its area is $0.46mm{\times}0.48mm$.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.24 no.4
• /
• pp.392-400
• /
• 2016

In this paper, a study based on engine operating conditions versus hybrid excavator engines was conducted about the engine performance and fuel consumption via the 1-D engine simulation model. First of all, engine operating points with performance and emission were determined by driving patterns. The 1-D HFEM(High Frequency Engine Model) was developed for deep insight into engine combustion and the energy conversion phenomena. In accordance with changing operating points, especially High Idle and Rated output conditions, engine parameters and systems such as turbocharger(Waste Gate Turbocharger and Variable Geometry Turbocharger) injection strategies and EGR(Exhaust Gas Recirculation) should be considered. Therefore, various configurations and parametric analysis with optimization methods in hybrid excavator were simulated and optimized by NLPQL(Non-linear Programming by Quadratic Lagrangian algorithm) in 1-D HFEM. As a result, the fuel consumption with the developed hybrid electric excavator engine could be significantly decreased and bsfc(Brake Specific Fuel Consumption) was also reduced about 5 % to 7 % without any performance degradation.

• Bulletin of the Korean Chemical Society
• /
• v.34 no.1
• /
• pp.225-230
• /
• 2013

A new heteropolyanion-based ionic ($[Hmim]_5PMo_{10}V_2O_{40}$) was synthesized by the reaction of molybdovanadophosphoric acid ($H_5PMo_{10}O_{40}$) with N-methylimidazole. [$[Hmim]_5PMo_{10}V_2O_{40}$ showed a high catalytic activity in the oxidative desulfurization of sulfur-containing compounds in 1-methylimidazolium tetrafluoroborate ($[Hmim]BF_4$) ionic liquid using 30% aqueous $H_2O_2$ as the oxidant. The catalytic system was of high activity, simplified workup and flexible recyclability. The catalytic oxidation reactivity of sulfur-containing compounds decreased in the order dibenzothiophene (DBT) > 4,6-dimethyldibenzothiophene (4,6-DMDBT) > benzothiophene (BT). The influences of various parameters including reaction time (t) and temperature (T), catalyst dosage, and oxidant to sulfur molar ratio n(O)/n(S) on the desulfurization of model oil were investigated in details. 99.1% of DBT conversion in the model oil was achieved at atmospheric pressure under the optimal conditions: n(O)/n(S) = 4:1, $60^{\circ}C$, 100 min and molar ratio of catalyst to sulfur of 0.062. The ionic liquid can be recycled six times without significant decrease in activity.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.5
• /
• pp.1135-1140
• /
• 2011

This paper proposes a DTMB deinterleaver structure to reduce the SDRAM power consumption with buffer conversion and the single pointer-register structure. The DTMB deinterleaver with deep interleaving for higher performance consists of long delay buffers allocated on SDRAM. The conventional structure activates a new SDRAM row almost everytime when it reads and writes a datum. In the proposed structure, long buffers are transformed into several short buffers so that the number of row activations is reduced. The single pointer-register structure solves the problem of many pointer-registers. The experimental results show that the SDRAM power consumption can be reduced to around 37% with slight logic area reduction.

• Tunnel and Underground Space
• /
• v.30 no.1
• /
• pp.1-14
• /
• 2020

Microseismic monitoring is utilized for the performance verification and safety management of the structure by detecting fine levels of damage. In order to construct a highly reliable microseismic monitoring system, the role of signal conditioner is critical. The signal conditioner helps with accurate data collection and precision control of the device, and performs additional functions such as signal conversion, linearization, and amplification. In this technical report, noise reduction signal conditioner suitable for mining sites was developed and reviewed for the purpose of implementing more precise monitoring by supplementing the previously developed microseismic monitoring system.

• International Journal of Advanced Culture Technology
• /
• v.6 no.4
• /
• pp.12-19
• /
• 2018

This article aims to discuss the transition of the Kazakh writing system from Cyrillic to Latin. First, the study investigates the relationship between the Kazakh Cyrillic alphabet and phonology, in order to linguistically evaluate the efficiency of the writing system. Second, the process of determining the Kazakh Latin alphabet is discussed in terms of the Kazakh phonological system. Third, the factors that determined the Latin alphabet of Kazakh language are analyzed. In Kazakh, the phonemic system is subject to controversy among linguists, but it can be said that the phonological system basically follows the one-to-one correspondence to the Russian and Kazakh phonemes. As for the depth of orthographies, Kazakh Cyrillic writing system is not based on the shallow orthographies, so it incorporates morphophonemic information to make skilled readers understand easier. The political and social aspects are considered as a cause of the alphabet change. Although there are studies suggesting the conversion of the writing system is caused by the extrinsic factors rather than the intrinsic factors, the five criteria of Smalley (1964), which compromise the intrinsic and extrinsic factors, are also persuasive. The five factors are 1) Maximum motivation for the learner, 2) Maximum representation of speech, 3) Maximum ease of learning, 4) Maximum transfer, 5) Maximum ease of reproduction.