• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.7
• /
• pp.3543-3557
• /
• 2017

Compression is a very important technique for remotely sensed hyperspectral images. The lossless compression based on the recursive least square (RLS), which eliminates hyperspectral images' redundancy using both spatial and spectral correlations, is an extremely powerful tool for this purpose, but the relatively high computational complexity limits its application to time-critical scenarios. In order to improve the computational efficiency of the algorithm, we optimize its serial version and develop a new parallel implementation on graphics processing units (GPUs). Namely, an optimized recursive least square based on optimal number of prediction bands is introduced firstly. Then we use this approach as a case study to illustrate the advantages and potential challenges of applying GPU parallel optimization principles to the considered problem. The proposed parallel method properly exploits the low-level architecture of GPUs and has been carried out using the compute unified device architecture (CUDA). The GPU parallel implementation is compared with the serial implementation on CPU. Experimental results indicate remarkable acceleration factors and real-time performance, while retaining exactly the same bit rate with regard to the serial version of the compressor.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.4
• /
• pp.36-45
• /
• 2014

Model based control methods are widely used to improve the control performance of diesel engine air systems because the control results of the air system significantly affect the emission level and drivability. However, the model based control algorithm requires a lot of unmeasurable states which are hard to be measured in a mass production engine. In this study, an air system model of the diesel engine is proposed to estimate 11 unmeasurable states using only sensors equipped in a mass production engine. In order to improve the estimation performance in the transient condition, dynamic characteristics of the air system are analyzed and implemented as discrete filters. Turbine and compressor efficiency models are also proposed to overcome a limitation of the constant or look-up table based efficiency values. The proposed air system model was validated in steady state and transient conditions by real-time engine experiments. The maximum error of the estimation for 11 physical states was 11.7%.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.35 no.3
• /
• pp.238-247
• /
• 2007

Genetic Algorithms(GA) which searches optimum solution using natural selection and the law of heredity has been applied to learning algorithms in order to estimate performance deterioration of the aircraft gas turbine engine. The compressor, gas generator turbine and power turbine are considered for engine performance deterioration and estimation for performance deterioration of a single component at design point was conducted. As a result of that, defect diagnostics has been conducted. The input criteria for the genetic algorithm to guarantee the high stability and reliability was discussed as increasing learning data sets. As a result, the accuracy of defect estimation and diagnostics were verified with its RMS error within 3%.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.45 no.3
• /
• pp.352-357
• /
• 1996

Recently, Permanent Magnet Synchronous Motor(PMSM) drives are widely used for industrial applications due to its high efficiency and high power factor control strategy. PMSM generally have two classifications such as the SPMSM(Surface Permanent Magnet Synchronous Motors) and IPMSM(Inter Permanent Magnet Synchronous Motors). IPMSA has economical merits over SPMSM in higher speed range, mechanical robustness, and higher power rate by the geometric difference. The maximum torque operation in IPMSM is realized by the current angle control which is to utilize additional reluctance torque due to a rotor saliency. In traction, spindle and compressor drives, constant power operation with higher speed range are desirable. This is simply achieved in the DC motor drives by the reduction of the field current as the speed is increased. However, in the PMSM, direct control of the magnet flux is not available. The airgap flux can be weakened by the appropriate current angle control to demagnetize. In this paper, the control method of optimal current vector in IPMSM is described in order to obtain the maximum torque or maximum output with the speed and load variations. The applied algorithm is realized by the proto system with torque and speed control Experimental results show this approach is satisfied for the high performance servo applications. (author). 6 refs., 9 figs., 1 tab.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2007.11a
• /
• pp.350-353
• /
• 2007

In this study, the genetic algorithm has been used for the real-time defect diagnosis on the operation of the aircraft gas-turbine engine. The component elements of the gas-turbine engine for consideriation of the performance deterioration is consist of the compressor, the gas generation turbine and the power turbine, repectively. Compared to the on-design point on the sea-level condition, the learning data has been increased 200 times in case of the off-design conditions for the altitude, the flight mach number and the fuel consumption. Therefore, enormous learning time has been required for the satisfied convergence. The optimum division has been proposed to decrease learning time as well as to obtain high accuracy. As results, the RMS errors of the defect diagnosis using the genetic algorithm have been estimated under 5 %.

• Journal of Internet Computing and Services
• /
• v.8 no.3
• /
• pp.131-143
• /
• 2007

The current trends in information technology and intelligent systems use data mining techniques to discover patterns and extract rules from distributed databases. In distributed environment, the extracted rules from data mining techniques can be used in dynamic replications, adaptive load balancing and other schemes. However, transmission of large data through the system can cause errors and unreliable results. This paper proposes the intelligent distributed platform based on dynamic group binding using mobile agents which addresses the use of intelligence in distributed environment. The proposed grouping service implements classification scheme of objects. Data compressor agent and data miner agent extracts rules and compresses data, respectively, from the service node databases. The proposed algorithm performs preprocessing where it merges the less frequent dataset using neuro-fuzzy classifier before sending the data. Object group classification, data mining the service node database, data compression method, and rule extraction were simulated. Result of experiments in efficient data compression and reliable rule extraction shows that the proposed algorithm has better performance compared to other methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.551-553
• /
• 2016

Since air compressor, as an essential equipment used in the factory and plant operations, accounts for around 20% of the total domestic electricity consumption, a real time sensor data monitoring based analysis for electricity consumption reduction is important. In particular, flow rates and pressures of these monitored variables has a direct correlation with the power consumption. This paper proposes a method to identify if the measurement error of the flow rate sensor comes from the sensor measurement limit through bivariate classification analysis of the flow rate and power using the EM (Expectation and Maximization) Algorithm and show how to enable more accurate analysis by the correlation between the flow rate and power on the right-censored data.

• Smart Structures and Systems
• /
• v.29 no.6
• /
• pp.757-766
• /
• 2022

To train deep learning algorithms, a sufficient number of data are required. However, in most engineering systems, the acquisition of fault data is difficult or sometimes not feasible, while normal data are secured. The dearth of data is one of the major challenges to developing deep learning models, and fault diagnosis in particular cannot be made in the absence of fault data. With this context, this paper proposes an anomaly detection methodology for rotating machines using only normal data with self-labeling. Since only normal data are used for anomaly detection, a self-labeling method is used to generate a new labeled dataset. The overall procedure includes the following three steps: (1) transformation of normal data to self-labeled data based on a pretext task, (2) training the convolutional neural networks (CNN), and (3) anomaly detection using defined anomaly score based on the softmax output of the trained CNN. The softmax value of the abnormal sample shows different behavior from the normal softmax values. To verify the proposed method, four case studies were conducted, on the Case Western Reserve University (CWRU) bearing dataset, IEEE PHM 2012 data challenge dataset, PHMAP 2021 data challenge dataset, and laboratory bearing testbed; and the results were compared to those of existing machine learning and deep learning methods. The results showed that the proposed algorithm could detect faults in the bearing testbed and compressor with over 99.7% accuracy. In particular, it was possible to detect not only bearing faults but also structural faults such as unbalance and belt looseness with very high accuracy. Compared with the existing GAN, the autoencoder-based anomaly detection algorithm, the proposed method showed high anomaly detection performance.

• Journal of Biomedical Engineering Research
• /
• v.28 no.4
• /
• pp.569-576
• /
• 2007

Ventricular Assist Device(VAD) has switched its goal from a short-tenn use for bridge-to-transplantation to a long-tenn use for destination therapy, With this goal, the importance of long-tenn reliability gets more interests and importances, H-VAD is an portable extracorporeal biventricular assist device, and adopts an electro-pneumatic driving mechanism. The pneumatic pressure to pump out blood is generated with compression of bellows, and is transmitted in a closed pneumatic circuit through a pneumatic line. The existing pneumatic VAD adopts a air compressor which can generate stable pressures but has defects such as a noise and a size problem. Thus, it is not suitable for being used as a portable device, These problems are covered with adopting a closed pneumatic circuit mechanism with a bellows which has a small size and small noise generation, but it has defects that improper pneumatic setting causes a failure of adequate flow generation. In this study, the pneumatic pressure regulation system is developed to cover these defects of a bellows-type pneumatic VAD. The optimal pneumatic pressure conditions according to various afterload conditions for an optimal flow rate were investigated and the afterload estimation algorithm was developed, The final pneumatic regulation system estimates a current afterload and regulate the pneumatic pressure to the optimal point at a given afterload condition. The afterload estimation algorithm showed a sufficient performance that the standard deviation of error is 8.8 mmHg, The pneumatic pressure regulation system showed a sufficient performance that the flow rate was stably governed to various afterload conditions. In a further study, if a additional sensor such as ultrasonic sensor is developed to monitor the direct movement of diaphragm in a blood pump part, the reliability would be greatly increased. Moreover, if the afterload estimation algorithm gets more accuracy, it would be also helpful to monitor the hemodynamic condition of patients.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.41 no.9
• /
• pp.115-124
• /
• 2004

This paper proposes a high efficient and low power FIR filter chip for partial-response maximum likelihood (PRML) disk drive read channels; it is a 6-bit, 8-tap digital FIR filter. The proposed filter employs a parallel processing architecture and consists of 4 pipeline stages. It uses the modified Booth algorithm for multiplication and compressor logic for addition. CMOS pass-transistor logic is used for low power consumption and single-rail logic is used to reduce the chip area. The proposed filter is actually implemented and the chip dissipates 120mV at 100MHz, uses a 3.3V power supply and occupies 1.88 ${\times}$ 1.38 $\textrm{mm}^2$. The implemented filter requires approximately 11.7% less power compared with the existing architectures that use the similar technology.