• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.15 no.3
• /
• pp.159-165
• /
• 2003

The fault detection and diagnosis technology may be applied in order to decrease the energy consumption and the maintenance cost of the air-conditioning system. To detect partial faults of the air-conditioning system, a neural network algorithm may be used. In this study, the neural network algorithm using normalized input data by the standard deviation was applied. And the [7$\times$10$\times$10$\times$1] neural network structure was selected. Test results showed that the neural network algorithm using normalized input data was very effective to detect the condenser fouling and the evaporator fan fault of an air-conditioning system.

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.231-232
• /
• 2007

In this paper, a fault diagnosis and control integrated system (FDCIS) was developed to control the thermoelectric (TE) power in the SP-100 space reactor. The objectives of the proposed model predictive control were to minimize both the difference between the predicted TE power and the desired power, and the variation of control drum angle that adjusts the control reactivity. Also, the objectives were subject to maximum and minimum control drum angle and maximum drum angle variation speed. A genetic algorithm was used to optimize the model predictive controller. The model predictive controller was integrated with a fault detection and diagnostics algorithm so that the controller can work properly even under input and output measurement faults. With the presence of faults, the control law was reconfigured using online estimates of the measurements. Simulation results of the proposed controller showed that the TE generator power level controlled by the proposed controller could track the target power level effectively even under measurement faults, satisfying all control constraints.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.5
• /
• pp.73-81
• /
• 2015

Most of the software fault prediction studies focused on the binary classification model that predicts whether an input entity has faults or not. However the ability to predict entity fault-proneness in various severity categories is more useful because not all faults have the same severity. In this paper, we propose fault prediction models at different severity levels of faults using traditional size and complexity metrics. They are ternary classification models and use four machine learning algorithms for their training. Empirical analysis is performed using two NASA public data sets and a performance measure, accuracy. The evaluation results show that backpropagation neural network model outperforms other models on both data sets, with about 81% and 88% in terms of accuracy score respectively.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1853-1856
• /
• 2003

This paper considers the problem of longitudinal control of a platoon of automotive vehicles on a straight lane of a highway and proposes control laws in the event of loss of communication between the lead vehicle and the other vehicles in the platoon. Since safety plays a key role in the development of an Automated Highway System, fault-tolerant control is vital. In this paper, we develop a control algorithm in vehicle platooning and prove that this control algorithm is stable for certain class of faults such as parameter uncertainties. The performance of the controller is demonstrated through a series of simulations incorporating various vehicles and AHS faults. Results of simulation shows that the vehicles have good performance in spite of simple automotive and AHS failure, such as actuator failure,that is to say, engine input failure, communication failure between lead vehicle and the another vehicles.

• Journal of Power Electronics
• /
• v.14 no.6
• /
• pp.1157-1165
• /
• 2014

The T-type topology is a three-level topology that has an advantage in terms of its number of switching device and its efficiency when compared to the neutral-point clamped (NPC)-type topology. With the recent increase in the usage of the T-type topology, the interest in its reliability has also increased. Therefore, a tolerance control for a T-type rectifier is necessary to improve the reliability of applications when an open-switch fault occurs. NPC-type rectifiers cannot eliminate input current distortion completely. However, the T-type rectifier is able to restore distorted current. In this paper, a tolerance control for the $S_{x2}$ and $S_{x3}$ open-switch faults of a T-type rectifier is proposed where it is advantageous in terms of efficiency when compared with other tolerance controls. The performance of the proposed tolerance control is verified through simulation and experimental results.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.7 no.3
• /
• pp.135-142
• /
• 2012

The Quality of embedded system depends on the embedded software. As the complexity and the size of embedded software have been increasing, it is more likely that the software may include faults, and the reliability and stability issues are getting more important. In this paper, we propose a remote test tool for software based on VxWorks by using fault injection method. The test tool consists of test server and test client for testing on the cross development environment. The test server operates in the host system and user can not only test but also monitor the software by using it. The test client operates in the target system and it controls kernel objects and sends the input data into the software when receiving the control and data from the test server. We developed the prototype software and demonstrated the ability of testing software robustness by injecting faults.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.569-572
• /
• 1992

This paper describes a new method of pseudorandom testing of a digital circuit by use of correlation method and a neural network. The authors have recently proposed a new method of fault diagnosis of logical circuit by applying a pseudorandom M-sequence to the circuit under test, calculating the crosscorrelation function between the input and the output, and comparing the crosscorrelation functions with the references. This method, called MSEC method, is further extended by using a neural network in order to not only detect the existence of faults but also find the place or location of the faults. An experiment by using a simple digital circuit shows enough applicability of this method to industrial testing of circuit board.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10b
• /
• pp.1216-1221
• /
• 1990

An algorithm for multiple fault diagnosis of linear dynamic systems is proposed. The algorithm is constructed by using of the geometric approach based on observation that, when the number of faulty units of the system is known, the set of faulty units can be differentiated from other sets by checking linear varieties in the measurement data space. It is further shown that the system with t number of faults can be diagnosed within (t+1) sample-time units if the input-output measurements are rich and that the algorithm can be used for diagnosis even when the number of faults is not known in advance.

• Proceedings of the KIEE Conference
• /
• 2002.11b
• /
• pp.50-52
• /
• 2002

Distance relays need accurate current and voltage for determining the trip ignition. Therefore we must analyze nonlinear phenomena which cause distortion in signals first of all. This paper presents the effect of some distortion(arcing fault and current transformer saturation) in power system. The saturation of a current transformer distorts input current of a distance relay and arcing faults make current and voltage to be changed. This paper describes modeling methods of a current transformer and arcing faults, and describes the simulation result of two distance relay algorithms (discrete fourier transform and modified differential equation methods)

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.359-362
• /
• 2000

In this paper, we propose a test pattern generation algorithm on the basis of the identification of undetectable faults for Speed-Independent(SI) asynchronous control circuits. The proposed methodology generates tests from the specification of a target circuit, which describes the behavior of the circuit in the form of Signal Transition Graph (STG). The proposed identification method uses only topological information of a target circuit and reachability information of a fault-free circuit, which is generated in the form of Binary Decision Diagram(BDD) during pre-processing. Experimental results show that high fault coverage over single input stuck-at fault model is obtained for several synthesized SI circuits and the use of the identification process as a preprocessing decreases execution time of the proposed test generation with negligible costs.