• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.1
• /
• pp.59-70
• /
• 2022

In this paper, we proposed an active fault tolerant control (AFTC) method for the position control of a quadrotor with complete loss of effectiveness of one motor. We obtained the dynamics of a quadrotor using Lagrangian equation without small angle assumption. For detecting the fault on a motor, we designed a fault detection module, which consists of the fault detection and diagnosis (FDD) module and the fault detection and isolation (FDI) module. For the FDD module, we designed a nonlinear observer that observes the states of a quadrotor based on the obtained dynamics. Using the observed states of a quadrotor, we designed residual signals and set the appropriate threshold values of residual signals to detect the fault. Also, we designed an FDI module to identify the fault location using the designed additional conditions. To make a quadrotor track the desired path after detecting the fault of a motor, we designed a fault tolerant controller based on the multiple sliding surface control (MSSC) technique. Finally, through simulations, we verified the effectiveness of the proposed AFTC method for a quadrotor with complete loss of effectiveness of one motor.

• Analytical Science and Technology
• /
• v.20 no.4
• /
• pp.347-354
• /
• 2007

A simple and rapid high-performance liquid chromatography assay for the determination of residual novobiocin levels in bovine, porcine, chicken, flatfish and japanese eel muscle has been developed and validated. The separation condition for HPLC/UV was optimized with phenyl hexyl ($4.6{\times}150mm$, $5{\mu}m$) column with 10 mM monobasic sodium phosphate buffer (pH 2.5)/acetonitrile (50/50, v/v) as the mobile phase at a flow rate of 1.0 mL/min and detection wavelength was set at 254 nm. Residues were extracted from tissue by blending with methanol and lipid materials were removed with n-hexane. Then, the methanol extract was evaporated to dryness under a nitrogen stream, reconstituted in the mobile phase. Aliquot of the organic extract was decanted and filtered through $0.45{\mu}m$ syringe filter. The $20{\mu}L$ of the resulting solution was injected into the HPLC system. The calibration ranges were $0.5{\sim}5{\mu}g/g$ and calibration curves were linear with coefficients of correlation better than 0.95. The limits of quantification were $0.5{\mu}g/g$ for all muscles. The recoveries of bovine, porcine, chicken, flatfish and japaneseel muscles were 99.8%, 102.4%, 91.0%, 104.0% and 93.0%, respectively. The procedures were validated according to the CODEX guideline, determining specificity, linearity, accuracy, precision, quantitation limit and recovery.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.536-543
• /
• 2023

Speaker diarization, which labels for "who spoken when?" in speech with multiple speakers, has been studied on a deep neural network-based end-to-end method for labeling on speech overlap and optimization of speaker diarization models. Most deep neural network-based end-to-end speaker diarization systems perform multi-label classification problem that predicts the labels of all speakers spoken in each frame of speech. However, the performance of the multi-label-based model varies greatly depending on what the threshold is set to. In this paper, it is studied a speaker diarization system using single-label classification so that speaker diarization can be performed without thresholds. The proposed model estimate labels from the output of the model by converting speaker labels into a single label. To consider speaker label permutations in the training, the proposed model is used a combination of Permutation Invariant Training (PIT) loss and cross-entropy loss. In addition, how to add the residual connection structures to model is studied for effective learning of speaker diarization models with deep structures. The experiment used the Librispech database to generate and use simulated noise data for two speakers. When compared with the proposed method and baseline model using the Diarization Error Rate (DER) performance the proposed method can be labeling without threshold, and it has improved performance by about 20.7 %.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.32 no.1
• /
• pp.98-104
• /
• 1999

To investigate the relationships between ocean environmental characteristics, the time-series data of temperature and salinity observed at a station near at Hanlim set net in 1995 and 1996 are analyzed, and the results are as follow ; 1. In hanlim set net, the diurnal range of temperature and salinity variation in summer is very large and the amplitude of short-period fluctuation of temperature and salinity is very large. That is, not only the water of the middle and bottom layers (low temperature and high salinity) but also the coalstal water (high temperature and low salinity) appears alternatively depending on the current direction 2. from the result of mooring for 22 days in Hanlim set net, the mean speed and direction of tidal current in neap tide were 9.1 cm/sec and south westward in ebb time, and 11.6 cm/sec and north or northeastward in flood time, respectively. The highest speed of the current was 15cm/sec in ebb time, and 22.6 cm/sec in flood time. The mean speed and direction of tidal current in spring tide were 10.4 cm/sec, and southwestward in ebb time, and 12.3 cm/sec, and north or northestward in flood time, respectively. The highest speed of the current was 19.4 cm/sec in ebb time, and 20 cm/sec in flood time respectively. The mean speed of the current in flood time was larger than that in ebb time. The velocity vector along the major axis of semidiurnal tide ($M_2$) component was 1.5 times larger than that of diurnal tide ($K_1$), The major directions of two compornants were northwestward and east-southeastward and residiual current were 3.25 cm/sec and northwestward-directed. Result of TGPS Buoy tracer for 3 days between Biyang-Do and Chgui-Do showed that the mean speed was 1.6 knot in ebb time and 1.3 knot in flood time. Direction of tidal was southwestward in ebb time and northeastward in flood time respectively. The maximum current speed was 4.8 knot in ebb time and 3.7 knot in flood time respectively. The mean speed and direction of tidal in of offshore were 1.7 knot and northwestward in flood time. The residual current appeared 0.3 knot northeastward.

• Journal of Korean Society for Atmospheric Environment
• /
• v.31 no.1
• /
• pp.54-62
• /
• 2015

Small scale paint overspray booths are being operated nationwidely, for repair of passenger car body parts. paint aerosols are emitted from the paint overspray booth in operations. In paint overspray booth operations without ventilation system and air pollutants collection unit, it may land on nearby equipment. In this study a removal of sticky paint aerosol for application of the small-scale overspray paint booth. it's cause the surface of filter bag from generated sticky paint aerosol. To remove adhesion of paint aerosol the agglomerating agents are injected and mixed with sticky paint aerosols prior to reach the filter bag. The paint spray rate was set as $10{\pm}5g/min$ from air-atomized spray guns in the spray booth, injection rate of agglomerating was $10{\pm}5g/min$ in the mixing chamber. The filtration velocity including air pollutants varied from 0.2 m/min to 0.4 m/min. Bag cleaning air pressure was set as $5.0kg_f/min$ for detaching dust cake from surface of filter bag. Bag cleaning interval at the filtration velocity of 0.2 m/min was around 3 times longer than that of the 0.4 m/min. The residual pressure drop maintained highest value at the highest filtration velocity. Fractional efficiency of 99.952%~99.971% was possible to maintain for the particle size of 2.5 microns. Total collection efficiency at the filtration velocity of 0.2 m/min was 99.42%. During this study we could confirm high collection efficiency and long cleaning intervals for the test with filtration velocity of 0.2 m/min indicating an optimal value for the given dimensions of the test unit and test operating conditions.

• Korean Journal of Environmental Agriculture
• /
• v.28 no.3
• /
• pp.310-325
• /
• 2009

Veterinary drugs have been used to prevent livestock diseases for many years. In spite of having advantages, sometimes indiscriminate application, overdose and abuse may cause risk for human. Therefore, management and risk assessment of veterinary drugs become essential to ensure food safety. So the National Veterinary Research & Quarantine Service (NVRQS) impose on registration for use of veterinary drugs also Korea Food and Drug Administration (KFDA) fixed the maximum residue limits (MRLs) for veterinary drugs in food. Basically, KFDA set MRLs based on the veterinary drugs residual data. Moreover, KFDA estimates the theoretical maximum daily intake (TMDI) with food consumption data and population mean body weight. Recently, 116 veterinary drug MRLs including banned veterinary drugs have been established and other 32 veterinary drugs MRLs will be established within 2009. In 2010, MRLs among antibiotics and synthetic antibacterial agents in livestock fishery products (including milk and eggs) and honey (including royal jelly and propolis) which have not been set in Korea Food Code and Codex Alimentarius Commission (CAC) will be regulated as uniform limit (0.03 mg/kg). In future, veterinary drugs will be controlled strictly to strengthen public health by improving analytical method.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.45 no.5
• /
• pp.13-20
• /
• 2008

Although digital hearing aid algorithms have been developed to compensate hearing loss and to help hearing impaired people to communicate with others, digital hearing aid user still complain about difficulty of hearing the speech. The reason could be the quality of speech through digital hearing aid is insufficient to understand the speech caused by feedback, residual noise and etc. And another thing is masking effect among formants that makes sound quality low. In this study, we measured the masking characteristics of normal listeners and hearing impaired listeners having presbyacusis to confirm masking effect in speech itself. The experiment is composed of 5 tests; pure tone test, speech reception threshold (SRT) test, word recognition score (WRS) test, puretone masking test and speech masking test. In speech masking test, there are 25 speeches in each speech set. And log likelihood ratio (LLR) is introduced to evaluate the distortion of each speech objectively. As a result, the speech perception became lower by increasing the quantity of formant enhancement. And each enhanced speech in a speech set has statistically similar LLR, however speech perception is not. It means that acoustic masking effect rather than distortion influences speech perception. In actuality, according to the result of frequency analysis of the speech that people can not answer correctly, level difference between first formant and second formant is about 35dB, and it is similar to result of pure tone masking test(normal hearing subject:36.36dB, hearing impaired subject:32.86dB). Characteristics of masking effect is not similar between normal listeners and hearing impaired listeners. So it is required to check the characteristics of masking effect before wearing a hearing aid and to apply this characteristics to fitting.

• Journal of Internet Computing and Services
• /
• v.17 no.6
• /
• pp.17-23
• /
• 2016

In the future network such as Internet of Things (IoT), the number of computing devices are expected to grow exponentially, and each of the things communicates with the others and acquires information by itself. Due to the growing interest in IoT applications, the broadcasting in Opportunistic ad-hoc networks such as Machine-to-Machine (M2M) is very important transmission strategy which allows fast data dissemination. In distributed networks for IoT, the energy efficiency of the nodes is a key factor in the network performance. In this paper, we propose a fuzzy logic based probabilistic multi-hop broadcast (FPMCAST) algorithm which statistically disseminates data accordingly to the remaining energy rate, the replication density rate of sending node, and the distance rate between sending and receiving nodes. In proposed FPMCAST, the inference engine is based the fuzzy rule base which is consists of 27 if-then rules. It maps input and output parameters to membership functions of input and output. The output of fuzzy system defines the fuzzy sets for rebroadcasting probability, and defuzzification is used to extract a numeric result from the fuzzy set. Here Center of Gravity (COG) method is used to defuzzify the fuzzy set. Then, the performance of FPMCAST is evaluated through a simulation study. From the simulation, we demonstrate that the proposed FPMCAST algorithm significantly outperforms flooding and gossiping algorithms. Specially, the FPMCAST algorithm has longer network lifetime because the residual energy of each node consumes evenly.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.185-191
• /
• 2022

In this study, we propose a novel super-resolution neural network based on U-Net, residual neural network, and sub-pixel convolution. To prevent the loss of detailed information due to the max pooling of U-Net, we propose down-sampling and connection using sub-pixel convolution. This uses all pixels in the filter, unlike the max pooling that creates a new feature map with only the max value in the filter. As a 2×2 size filter passes, it creates a feature map consisting only of pixels in the upper left, upper right, lower left, and lower right. This makes it half the size and quadruple the number of feature maps. And we propose two methods to reduce the computation. The first uses sub-pixel convolution, which has no computation, and has better performance, instead of up-convolution. The second uses a layer that adds two feature maps instead of the connection layer of the U-Net. Experiments with a banchmark dataset show better PSNR values on all scale and benchmark datasets except for set5 data on scale 2, and well represent local area patterns.

• Structural Engineering and Mechanics
• /
• v.30 no.6
• /
• pp.713-732
• /
• 2008

Serviceability and durability of the concrete members can be seriously affected by the corrosion of steel rebar. Carbonation front and or chloride ingress can destroy the passive film on rebar and may set the corrosion (oxidation process). Depending on the level of oxidation (expansive corrosion products/rust) damage to the cover concrete takes place in the form of expansion, cracking and spalling or delamination. This makes the concrete unable to develop forces through bond and also become unprotected against further degradation from corrosion; and thus marks the end of service life for corrosion-affected structures. This paper presents an analytical model that predicts the weight loss of steel rebar and the corresponding time from onset of corrosion for the known corrosion rate and thus can be used for the determination of time to cover cracking in corrosion affected RC member. This model uses fully the thick-walled cylinder approach. The gradual crack propagation in radial directions (from inside) is considered when the circumferential tensile stresses at the inner surface of intact concrete have reached the tensile strength of concrete. The analysis is done separately with and without considering the stiffness of reinforcing steel and rust combine along with the assumption of zero residual strength of cracked concrete. The model accounts for the time required for corrosion products to fill a porous zone before they start inducing expansive pressure on the concrete surrounding the steel rebar. The capability of the model to produce the experimental trends is demonstrated by comparing the model's predictions with the results of experimental data published in the literature. The effect of considering the corroded reinforcing steel bar stiffness is demonstrated. A sensitivity analysis has also been carried out to show the influence of the various parameters. It has been found that material properties and their inter-relations significantly influence weight loss of rebar. Time to cover cracking from onset of corrosion for the same weight loss is influenced by corrosion rate and state of oxidation of corrosion product formed. Time to cover cracking from onset of corrosion is useful in making certain decisions pertaining to inspection, repair, rehabilitation, replacement and demolition of RC member/structure in corrosive environment.