• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.6
• /
• pp.503-510
• /
• 2002

This paper presents a fault diagnosis method using neural network-based multi-fault models and statistical method to detect and isolate faults in nonlinear systems. In the proposed method, faults are detected when the errors between the system output and the neural network nominal system output cross a predetermined threshold. Once a fault in the system is detected, the fault classifier statistically isolates the fault by using the error between each neural network-based fault model output and the system output. From the computer simulation results, it is verified that the proposed fault diagonal method can be performed successfully to detect and isolate faults in a nonlinear system.

• Journal of Applied Reliability
• /
• v.12 no.2
• /
• pp.105-119
• /
• 2012

In this paper, we have proposed a methodology which can make effective test for system integration of complex guided missile system. System integration test play a significant role in the development of weapon system, providing the means to detect and isolate faults on first linkage between sub-systems. Integration tests for domestic weapon system has executed not a technology-intensive method based on tool but labor-intensive method based on experience. Higher cost, longer period, and more resource are required to execute system integration test for complex guided missile system comparing with past weapon systems, because recently weapon systems have more complex and more networked functions. Because the proposed design method for system integration test decreases number of test case, it lead to a decrease of cost, period, and resource for integration test of weapon system. The proposed configuration for system integration test will ensure reliability through detection and isolation of fault on linkage between sub-systems.

• Smart Structures and Systems
• /
• v.22 no.2
• /
• pp.167-174
• /
• 2018

Digital input and output modules are widely used to connect digital sensors and actuators to automation systems. Digital I/O modules provide flexible connectivity extension to numerous sensors and actuators and protect systems from high voltages and currents by isolation. Components in digital I/O modules are inevitably affected by operating and environmental conditions, such as high voltage, high current, high temperature, and temperature cycling. Because digital I/O modules transfer signals or isolate the systems from unexpected voltage and current transients, their failures may result in signal transmission failures and damages to sensitive circuitry leading to system malfunction and system shutdown. In this study, the lifetime of optocouplers, one of the critical components in digital I/O modules, was predicted using Bayesian tracking approaches. Accelerated degradation tests were conducted for collecting the critical performance parameter of optocouplers, current transfer ratio (CTR), during their lifetime. Bayesian tracking approaches, including extended Kalman filter and particle filter, were applied to predict the failure. The performance of each prognostic algorithm was then compared using accuracy and robustness-based performance metrics.

• International Journal of Aerospace System Engineering
• /
• v.9 no.1
• /
• pp.1-15
• /
• 2022

Pyrotechnic initiators provide a source of pyrotechnic energy used to initiate a variety of space mechanisms. Pyrotechnic systems build in electromagnetic environment that may lead to critical or catastrophic hazards. Special precautions are need to prevent a pulse large enough to trigger the initiator from appearing in the pyrotechnic firing circuits at any but the desired time. The EMC verification shall be shown by analysis or test that the pyrotechnic systems meets the requirements of inadvertent activation. The MIL-STD-1576 and two range safeties, AFSPC and CSG, require the safety margin for electromagnetic potential hazards to pyrotechnic systems to a level at least 20 dB below the maximum no-fire power of the EED. The PC23 is equivalent to NASA standard initiator and the 1EPWH100 squib is ESA standard initiator. This paper verifies the two safety margins for electromagnetic potential hazards. The first is verified by analyzing against a RF power. The second is verified by testing against a DC current. The EMC safety margin requirement against RF power has been demonstrated through the electric field coupling analysis in differential mode with 21 dB both PC23 and 1EPWH100, and in common mode with 58 dB for PC23 and 48 dB for 1EPWH100 against the maximum no-fire power of the EED. Also, the EMC safety margin requirement against DC current has been demonstrated through the electrical isolation test for the pyrotechnic firing circuits with greater than 20 dB below the maximum no-fire current of the EED.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.603-622
• /
• 2023

Human error (HE) is an important concern in safety-critical systems such as nuclear power plants (NPPs). HE has played a role in many accidents and outage incidents in NPPs. Despite the increased automation in NPPs, HE remains unavoidable. Hence, the need for HE detection is as important as HE prevention efforts. In NPPs, HE is rather rare. Hence, anomaly detection, a widely used machine learning technique for detecting rare anomalous instances, can be repurposed to detect potential HE. In this study, we develop an unsupervised anomaly detection technique based on generative adversarial networks (GANs) to detect anomalies in manually collected surveillance data in NPPs. More specifically, our GAN is trained to detect mismatches between automatically recorded sensor data and manually collected surveillance data, and hence, identify anomalous instances that can be attributed to HE. We test our GAN on both a real-world dataset and an external dataset obtained from a testbed, and we benchmark our results against state-of-the-art unsupervised anomaly detection algorithms, including one-class support vector machine and isolation forest. Our results show that the proposed GAN provides improved anomaly detection performance. Our study is promising for the future development of artificial intelligence based HE detection systems.

• Transactions on Control, Automation and Systems Engineering
• /
• v.4 no.2
• /
• pp.114-123
• /
• 2002

Called VISTA (Variable-stability In-flight Simulator Test Aircraft), the one-of-a-kind NF-l6D has a simulation system that can mimic several aircraft. Though housed in an F-l6 Fighting Falcon airframe, VISTA can also act like the F-15 Eagle or the Navy's F-14 Tomcat. More importantly, such flexibility allows for improved training and consolidation of some sorties. Consequently USAF Test Pilot School students will have an opportunity to learn how to test future integrated cockpits. In this paper we will use the multiple model adaptive estimation (MMAE) and the multiple model adaptive controller (MMAC) techniques to model the aircraft's flight control system containing the longitudinal and lateral-directional axes. Single and dual actuator and sensor failures will also be included in the simulation. White Gaussian noise will be included to simulate the effects of atmospheric disturbances.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.3
• /
• pp.308-314
• /
• 2003

This paper presents a serial communication based fault diagnosis scheme for a brushless DC (BLDC) motor using parameter estimation and Bayes classifier. The presented scheme consists of a smart network board, and a fault detection and isolation (FDI) master. The smart network board is installed near the BLDC motor drive system to acquire motor data and transmit motor data to the FDI-master via serial communication channel. The FDI-master estimates BLDC motor resistance to detect symptom of faults, and assign symptom to fault type using Bayes classifier. In this scheme, since communication time delay has a serious effect on performance, periodic and fixed communication protocol is designed. Hence, the delay time is priory known. By experiment result, presented scheme was verified.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.7
• /
• pp.669-672
• /
• 2012

This paper investigates an algorithm for robust fault diagnosis in robot manipulators. The TOSM (Third Order Sliding Mode observer) provides both theoretically exact observation and unknown fault identification without filtration. The EOI (Equivalent Output Injections) of the TOSM observers can be used as residuals for the problem of fault diagnosis and to identify the unknown faults. The obtained fault information can be used for fault detection, isolation as well as fault accommodation to the self-correcting failure system. The computer simulation results for a PUMA 560 robot are shown to verify the effectiveness of the proposed strategy.

• Proceedings of the IEEK Conference
• /
• 2003.07c
• /
• pp.2775-2778
• /
• 2003

Neural network-based fault diagnosis algorithm to detect and isolate faults in the nonlinear systems is proposed. In the proposed method, the fault is detected when the errors between the system output and the neural network nominal system output cross a predetermined threshold. Once a fault in the system is detected, the system outputs are transferred to the fault classifier by ART2 NN (adaptive resonance theory 2 neural network) for fault isolation. From the computer simulation results, it is verified that the proposed fault diagonal method can be performed successfully to detect and isolate faults in a nonlinear system.

• Annual Conference of KIPS
• /
• 2006.05a
• /
• pp.605-608
• /
• 2006

본 논문에서는 SLA(Service Level Agreement)와 같이 차별화 서비스를 지원하는 웹서버 시스템의 가동성 척도를 향상시키기 위해 기존의 동적 성능 분리 기법에 퍼지 기법을 접목하였다. 특히, 클러스터 기반 웹서버 시스템의 부하량에 대한 판단기준 혹은 사용자 요청률 및 동적요청 비율 변화시에 발생하는 애매모호한 상황을 효과적으로 반영하기 위해, 퍼지제어 기법에 기초한 부하분배 메커니즘을 제안하였다. 이를 통해, 기존의 퍼지 기법을 활용하지 않은 성능분리 기법과 퍼지기법을 활용한 경우에 대해 응답시간(95-percentile of response time) 성능 비교 평가를 통해 퍼지기반의 성능분리 기법이 차별화 서비스 시스템의 성능을 더욱 강건하고 효율적으로 향상시킬 수 있다는 점을 검증하였다.