• Journal of Advanced Navigation Technology
• /
• v.7 no.2
• /
• pp.171-179
• /
• 2003

In this paper a neural network-based fault tolerant control system for aircraft sensor and actuator failures is considered. By exploiting flight dynamic relations a set of neural networks is constructed to detect sensor failure and give alternative signal for the faulty sensor. For actuator failures another set of neural networks is designed to perform fault detection, identification, and accomodation which returns the aircraft to a new stable trim. Integrated system is simulated to show the performance of the system with sensor and control surface failures.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.1 no.1
• /
• pp.82-91
• /
• 1998

This paper presents a neural networks based approach for the problem of sensor failure detection and accommodation for ship without physical redundancy in the sensors. The designed model consists of two neural networks. The first neural network is responsible for the failure detection and the second neural network is responsible for the failure identification and accommodation. On the yaw rate sensor of ship, simulation results indicates that the proposed method can be useful as failure detector and sensor estimator.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.32 no.4
• /
• pp.313-319
• /
• 2019

Induction motors connected with a three-phase AC system may malfunction due to reverse phase or open phase faults. Conventional overcurrent relays and overheating relays are used to prevent such accidents; however, their drawbacks include a low response speed and false operation. Therefore, in this study, a digital relay for the reverse-open phase was designed and fabricated. This relay can detect the reverse phase and open phase faults and send a trigger signal to the control circuit. The proposed relay was developed based on a microcontroller. The detection times of the reverse phase and open phase were verified as 320ms and 80ms, respectively. Compared with conventional relays that only protect the motor from one type of fault, the proposed relay can detect both, reverse phase and open phase faults. In addition, the fault detection, identification criterion, and trigger signal patterns can be modified by programming according to the requirements of users.

• Journal of Advanced Navigation Technology
• /
• v.10 no.4
• /
• pp.299-305
• /
• 2006

DURUMI-II is developed into test bed airplane for the multi-purpose flight test. It satisfied the civil aeronautics law. DURUMI-II is equipped with Airborne System for acquiring of flight test data and can fly by oneself. In this paper, the redundancy of DURUMI-II control system is operated sequentially is explained. The divided control surface and the requiring program method for flight test are described. Also, it is described that the exact control input is applied using the new method. Finally, the results of flight test for new method are analyzed.

• Journal of Advanced Research in Ocean Engineering
• /
• v.3 no.3
• /
• pp.125-135
• /
• 2017

The paper considers the practical application of the FMEA(Failure Mode and Effect Analysis) method to assess the operational reliability of the LNG(Liquefied Natural Gas) transfer system, which is a potential problem for the connection between the LNG FPSO and LNG carrier. Hazard Identification (HAZID) and Hazard operability (HAZOP) are applied to identify the risks and hazards during the operation of LNG transfer system. The approach is performed for the FMEA to assess the reliability based on the detection of defects typical to LNG transfer system. FTA and FMEA associated with a probabilistic risk database to the operation scenarios are applied to assess the risk. After providing an outline of the safety assessment procedure for the operational problems of system, safety assessment example is presented, providing details on the fault tree of operational accident, safety assessment, and risk measures.

• Journal of Electrical Engineering and Technology
• /
• v.3 no.3
• /
• pp.331-336
• /
• 2008

Electric load components have different characteristics according to the variation of voltage and frequency. This paper presents the load modeling of an electric locomotive by the parameter identification method. The proposed method for load modeling is very simple and easy for application. The proposed load model of the electric locomotive is represented by the combination of the loads that have static and dynamic characteristics. This load modeling is applied to the KTX in Korea to verify the effectiveness of the proposed method. The results of proposed load modeling by the parameter identification follow the field measurements very exactly.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.7 no.4
• /
• pp.42-49
• /
• 1998

This wear debris can be harvested from the lubricants of operating machinery and its morphology is directly related to the damage to the interacting surface from which the particles originated. The morphological identification of wear debris can therefore provide very early detection of a fault and can also often facilitate a diagnosis. The purpose of this study is to attempt the developement of intelligent system for moving condition diagnosis of the machine driving system. The four shape parameter(50% volumetric diameter, aspect, roundness and reflectivity) of war debris are used as inputs to the neural network and learned the moving condition of five values(material3, applied load 1, sliding distance 1). It is shown that identification results depend on the ranges of these shape parameter learned. The three kinds of the wear debris had a different pattern characteristics and recognized the moving condition and materials very well by neural network.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.97.2-97
• /
• 2002

A method for the non-destructive detection of damage using parameterized partial differential equations and Galerkin approximation techniques is presented. This method provides the theoretical and experimental verification of a nondestructive time domain approach to examine structural damage in smart structure. The time histories of the vibration response of structure were used to identify the presence of damage. Damage in a structure causes changes in the physical coefficients of mass density, elastic modulus and damping coefficient. This paper examines the beam-like structures with PVDF sensor and PZT actuator to perform identification of those physical parameters and to detect the...

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.2 s.95
• /
• pp.211-217
• /
• 1999

감지기의 고장 발견, 확인, 보완은 복잡한 항공 시스템의 중요한 문제로 부각되어 왔으며, 그동안 칼만 필터를 이용한 기존 추정기술 혹은 온라인 학습 인공지능 알고리듬 등이 이 같은 문제를 해결하기 위해 제시되어 왔다. 본 연구에서는 여분의 감지기가 없는 항공제어계에 대해 온라인 학습 신경망을 이용한 감지기의 고장 발견, 확인, 그리고 보완에 관해 초점을 둔다. 이 내고장성 항공제어계는 주 신경조직망과 n개의 국소 신경조직망으로 이루어지는데, 포괄적인 감지기의 고장을 발견하는 능력을 가진다. 어떤 경우에서는 기존의 감지기 고장 발견 방법의 성능을 향상시키기 위해 수정된 감지방법이 소개되고 그 보완된 감지방법을 이용하여 기존의 방법과 성능비교가 이루어졌다.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.181-193
• /
• 2022

Data-driven structural health monitoring (SHM) of civil infrastructure can be used to continuously assess the state of a structure, allowing preemptive safety measures to be carried out. Long-term monitoring of large-scale civil infrastructure often involves data-collection using a network of numerous sensors of various types. Malfunctioning sensors in the network are common, which can disrupt the condition assessment and even lead to false-negative indications of damage. The overwhelming size of the data collected renders manual approaches to ensure data quality intractable. The task of detecting and classifying an anomaly in the raw data is non-trivial. We propose an approach to automate this task, improving upon the previously developed technique of image-based pre-processing on one-dimensional (1D) data by enriching the features of the neural network input data with multiple channels. In particular, feature engineering is employed to convert the measured time histories into a 3-channel image comprised of (i) the time history, (ii) the spectrogram, and (iii) the probability density function representation of the signal. To demonstrate this approach, a CNN model is designed and trained on a dataset consisting of acceleration records of sensors installed on a long-span bridge, with the goal of fault detection and classification. The effect of imbalance in anomaly patterns observed is studied to better account for unseen test cases. The proposed framework achieves high overall accuracy and recall even when tested on an unseen dataset that is much larger than the samples used for training, offering a viable solution for implementation on full-scale structures where limited labeled-training data is available.