• 센서학회지
• /
• 제14권6호
• /
• pp.374-380
• /
• 2005

Vibration sensors have a wide scope of applications in the field of monitoring systems that needs to perceive an undesirable physical vibration before a critical failure occurs in a system, and then costly unplanned repairs can be avoided. The conventional vibration sensors developed so far have many disadvantages, such as complex manufacturing process, bulkiness, high cost, less reliability and so on. This paper reports a simple-structured vibration sensor, which has been developed using a commercialized conductive ball and silicon bulk-micromachining technology. The sensor consists of a conductive ball placed in $600{\mu}m$-deep micromachined silicon groove, in which Au thin film has been patterned using a shadow mask technique. Prior to the formation of the Au thin film, the sharp convex corner was rounded for smooth meatl deposition on the non-planar surface at the edge of the groove. The measurement results of the fabricated vibration sensor demonstrate a stable response characteristic to low-frequency vibration range ($1{\sim}30{\;}Hz$).

• 한국자동차공학회논문집
• /
• 제18권4호
• /
• pp.8-17
• /
• 2010

In this paper, a fault tolerant control against sensor faults of electric parking brake (EPB) is proposed. Fault tolerant control method of EPB system is strongly demanded since sensor faults can endanger a driver's safety. In this paper, a clamp force estimation method is presented using motor's armature current and angular velocity. Clamp force estimation method is applied for fault detection method with parity equations. The goal of the detection method is to detect and identify faults in encoder, current sensor, force sensor, and parking cable. And a switching logic for fault tolerant control against the three sensor faults is suggested. Experimental results show that the proposed force estimation method satisfies the specifications of EPB system. The effectiveness of the fault detection method is validated with experimental results. Although a single sensor fault happens, EPB system with the proposed fault detection method does not develop into a failure on subsystem or system level.

• 한국정밀공학회지
• /
• 제28권12호
• /
• pp.1372-1378
• /
• 2011

The smart air condition system is superior to conventional air condition system in the aspect of control accuracy, environmental preservation and it is foundation for intelligent vehicle such as electric vehicle, fuel cell vehicle. In this paper, failure analyses of smart air condition system will be performed and then sensor fusion technique will be proposed for fail safety of smart air condition system. A sensor fusion logic of air condition system by using CO sensor, $CO_2$ sensor and VOC, $NO_x$ sensor will be developed and simulated by fault injection simulation. The fusion technology of smart air condition system is generated in an experiment and a performance analysis is conducted with fusion algorithms. The proposed algorithm adds the error characteristic of each sensor as a conditional probability value, and ensures greater accuracy by performing the track fusion with the sensors with the most reliable performance.

• 한국지능시스템학회논문지
• /
• 제15권5호
• /
• pp.615-620
• /
• 2005

Sensor networks are usually composed of tens or thousands of tiny devices with limited resources. Because of their limited resources, there will often be some faulty nodes within the network. As nodes in some certain regions rely on each other to route the information gathered by different sensors to a base station (sink), the network should be able to detect a non-operational node in order to determine new paths for routing the information. Failure detection, which identifies the faulty nodes, is rather necessary in sensor networks and a very important research issue. The detection of a non-operational node can be performed using Consensus Algorithm with the purpose of achieving agreement about a node which is supposed to be faulty (non-operational). In this work, we discuss the application of a Consensus Algorithm to sensor node called 'mote'. Our experimental results show that it works efficiently for identifying faulty nodes in sensor networks.

• Journal of information and communication convergence engineering
• /
• 제9권6호
• /
• pp.775-779
• /
• 2011

As the technology of multimedia sensor networks is desired in large numbers of applications nowadays, real-time service becomes one of the most important research challenges. Even though lots of related works have been conducted to meet this requirement in several ways, the specific traffic model for real-time has not been taken yet. Thus, it causes lack of adaptability of those approaches in real deployment. To solve this problem, in this paper, we model the application via (m,k)-firm streams which have weakly hard real-time property. And then, a novel forwarding scheme based on modified DBP (Distance-Based Priority) is proposed by considering local-DBP and stream DBP together. Local-DBP can contribute to identify the detailed causes of unsatisfied quality, that is, network congestion or wireless link failure. Simulation results reveal that (m,k)-firm is a good traffic model for multimedia sensor networks and the proposed scheme can contribute to guarantee real-time requirement well.

• ETRI Journal
• /
• 제31권5호
• /
• pp.525-533
• /
• 2009

This paper proposes a distributed borrowing addressing (DIBA) scheme to solve problems of failure in address assignments resulting from limited tree depth and width when the distributed address assignment mechanism is used in a ZigBee/IEEE 802.15.4 wireless sensor network. DIBA is a method of borrowing addresses from neighbor nodes for newly entering nodes and assigning the borrowed addresses. Its network or sensing coverage can increase with almost the same overhead as the existing method. DIBA is a simple and lightweight means of addressing and routing, making it suitable for wireless sensor networks. Simulations showed that DIBA is a distributed addressing scheme with consistently excellent performance.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2003년도 춘계학술대회
• /
• pp.370-375
• /
• 2003

Optical fiber vibrations sensors (OFVSs) and extrinsic Fabry-Perot interferometer (EFPI) were used in damage monitoring of fiber-metal laminates(FML). The optical fiber vibration sensor and EFPI were applied in order to detect and evaluate the strain, damage and failure of FML. Damages in composites, such as matrix cracks, delamination and fiber breakage may occur as a result of excessive load, fatigue and low-velocity impacts. Tensile and indentation test was performed with the measurement of optical signal and acoustic emission (AE). The signals of the optical fiber vibration sensor due to damages were quantitatively evaluated by wavelet transform. It was found that damage information of comparable in quality to acoustic emission data could be obtained from the optical fiber vibration sensor signals.

• 한국전기전자재료학회논문지
• /
• 제30권4호
• /
• pp.258-262
• /
• 2017

Recently energy consumption has been increasing because of advances in the industry, and electrical fires have accounted for 31.9% of all fire accidents. An electrical fire is caused by a short circuit, power surge, or poor contact. Safety devices for short circuits or power surges are currently mandatory and can actually detect problems, but arcing caused by contact failure is difficult to detect in advance. This study used an IR sensor to detect the heat concentration caused by the arc. The data from the low-resolution sensor was amplified as much as four times by interpolation to find the exact location of the heat source and were then investigated.

• International Journal of Precision Engineering and Manufacturing
• /
• 제5권2호
• /
• pp.46-52
• /
• 2004

Numerous connectors are used in automobiles for transmission of electrical signals across various electro-mechanical components. The connectors must operate with high reliability in order to minimize failures due to signal degradation. In this work, the effects of contamination at the contact interface of connectors used fur automobile crankshaft position sensor on the impedance change were investigated. An experimental set-up was built to simulate the electrical signal transmitted from the sensor to the engine control unit through a connector. Output from the connector was investigated using connectors contaminated with engine block residues and water droplets. It was found that slight contamination of the connectors could lead to significant signal degradation which can lead to engine failure. Also, the effect of water in the connector altered the signal severely. However, the signal gradually regained the original state as the water evaporated from the interface.

• 한국정밀공학회지
• /
• 제14권2호
• /
• pp.41-47
• /
• 1997

Fault diagnosis problem is currently a subject of extensive research in the control field. Although there are several works on the fault detection and isolation observers and the residual generators, those are con- cerned with only the detection of actuator failures or sensor failures. So, the perfect detection and isolation for the actuator and sensor failures is strongly required in the field of the practical applications. In this paper, a strategy of fault diagnosis using multiple proportional integral (PI) observers including the magnitude of actuator failures is provided. It is shown that actuator failures are detected and isolated perfectly by monitoring the integrated error between actual output and estimated output by a PI observer. Also in presence of complex actuator and sensor failures, these failures are detected and isolated by multiple PI observers.