• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.20 no.9
• /
• pp.28-34
• /
• 2021

The Fourth Industrial Revolution has led to the development of drones for commercial and private applications. Therefore, the malfunction of drones has become a prominent problem. Failure mode and effect analysis was used in this study to analyze the primary cause of drone failure, and blade breakage was observed to have the highest frequency of failure. This was tested using a vibration sensor placed on drones along the breakage length of the blades. The data exhibited a significant increase in vibration within the drone body for blade fracture length. Principal component analysis was used to reduce the data dimension and classify the state with machine learning algorithms such as support vector machine, k-nearest neighbor, Gaussian naive Bayes, and random forest. The performance of machine learning was higher than 0.95 for the four algorithms in terms of accuracy, precision, recall, and f1-score. A follow-up study on failure prediction will be conducted based on the results of fault diagnosis.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.14 no.9 s.90
• /
• pp.819-826
• /
• 2004

A multi-step optimum strategy for the selection of the locations and directions of strain gauges is proposed in this paper to capture at best the modal response of blade in a series of modes on fan blades. It is consist of three steps including two pass reduction step, genetic algorithm and fine optimization to find the locations-directions of strain gauges. The optimization is based upon the maximum signal-to-noise ratio(SNR) of measured strain values with respect to the inherent system measurement noise, the mispositioning of the gauge in location and gauge failure. Optimal gauge positions for a fan blade is analyzed to prove the effectiveness of the multi-step optimum methodology and to investigate the effects of the considering parameters such as the mispositioning level, the probability of gauge failure, and the number of gauges on the optimal strain gauge position.

• Ocean Systems Engineering
• /
• v.3 no.4
• /
• pp.295-307
• /
• 2013

More FOWTs (floating offshore wind turbines) will be installed as relevant regulations and technological hurdles are removed in the coming years. In the present study, a numerical prediction tool has been developed for the fully coupled dynamic analysis of FOWTs in time domain including aero-loading, tower elasticity, blade-rotor dynamics and control, mooring dynamics, and platform motions so that the influence of rotor-control dynamics on the hull-mooring performance and vice versa can be assessed. The developed coupled analysis program is applied to Hywind spar design with 5 MW turbine. In case of spar-type floaters, the control strategy significantly influences the hull and mooring dynamics. If one of the control systems fails, the entire dynamic responses of FOWT can be significantly different. Therefore, it is important to maintain various control systems in a good operational condition. In this regard, the effects of failed blade pitch control system on FOWT performance including structural and dynamic responses of blades, tower, and floater are systematically investigated. Through this study, it is seen that the failure of one of the blade pitch control system can induce significant dynamic loadings on the other blades and the entire FOWT system. The developed technology and numerical tool are readily applicable to any types of floating wind farms in any combinations of irregular waves, dynamic winds, and steady currents.

• Proceedings of the KSME Conference
• /
• 2001.11a
• /
• pp.258-265
• /
• 2001

Turbine blade is subject to force of three type ; torsional force by torsion-mount, centrifugal force by rotation of rotor and cyclic bending force by steam pressure. Cyclic bending force of them is main factor on fatigue fracture. In the X-ray diffraction method, the change in the values related to plastic deformation and residual stress near the fracture surface mat be determined, and information of internal structure of material can be obtained. Therefore, to find a fracture mechanism of torsion-mounted blade in nuclear plant, based on the information from the fracture surface obtained by fatigue test, the correlation of X-ray parameter and fracture mechanics parameter was determined, and then the load applied to actual broken turbine blade parts was predicted. Failure analysis is performed by finite element method and Goodman diagram on torsion-mounted blade.

• Journal of Biosystems Engineering
• /
• v.18 no.1
• /
• pp.15-20
• /
• 1993

Wearness has been a major failure criterion in Korean-made rotary blade. However, few studies have been conducted to improve it. In this study, the fundamental data obtained from the measurement of wearness and failure of rotary blade were analyzed to provide a guideline for the design of rotary blades. For the straight part(about 20-23 em from bolt hole) from the bolt hole to bending point of rotary blade, modifications were proposed for improvements, however, for the portion from bending point to tip was made no design recommendations because the failure behavior of that portion was difficult to analyze with the experimental data. The results are summarized as follows. 1. The current V-shape section has to be moved about 5 em toward the bending point of rotary blade. 2. The section modulus at the portion about 5-7 em distant from bolt hole has to be increased about 15-20 %. 3. The V-shape section has to be changed into U-shape to reduce the on account of recieving initial stress in blades. 4. The radius of curvature of the neck(the portion about 5-7 cm apart from bolt hole) has to be made larger to decrease the stress concentration.

• Journal of Welding and Joining
• /
• v.19 no.3
• /
• pp.298-304
• /
• 2001

Turbine blade is subject to force of three types ; the torsional force by torsional mount, the centrifugal force by the rotation of rotor and the cyclic bending force by steam pressure. The cyclic bending force was a main factor on fatigue strength. SEM fractography in root of turbine blade showed micro-clack width was not dependent on stress intensity factor range. Especially, fatigue did not exist on SEM photograph in root of turbine blade. To clear out the fracture mechanism of turbine blade, nanofractography was needed on 3-dimensional crack initiation and crack growth with high magnification. Fatigue striation partially existed on AFM photograph in root of turbine blade. Therefore, to find a fracture mechanism of the torsion-mounted blade in nuclear power plant, the relation between stress intensity factor range and surface roughness measured by AFM was estimated, and then the load amplitude ΔP applied to turbine blade was predicted exactly by root mean square roughness.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.25 no.11
• /
• pp.1705-1712
• /
• 2001

Mechanical component has striation with constant width and SEM can estimate fracture type and loading condition. SEM has benefit to fatigue fracture analysis but striation can be observed according to the kind of material and range of crack growth rate and can't. In this case, it needs AFM that can measure 3-dimensional surface profile with resolution of atomic size. In this study. to find fracture reason of torsion-mounted blade in nuclear plant, we estimate the relation between stress intensity factor range and root mean square roughness in 12% Cr steel by AFM and predict in-service loading condition of turbine blade. failure analysis is performed by finite element method and Goodman diagram on torsion-mounted blade.

• 한국신재생에너지학회:학술대회논문집
• /
• 2009.11a
• /
• pp.448-448
• /
• 2009

A new blade has been developed to apply to Doosan 3MW offshore wind turbine named as WinDS3000TM. The 3MW blade has been designed by the concept of slim external shape and optimized structure. High-performance glass fiber reinforced epoxy composites were used as the main material of the blade. The blade was manufactured using vacuum infusion process in order to increase the fiber volume fraction and to reduce micro-porosities. The blade has successfully passed the full-scale blade static test for certification. During the test, micro-failure signal and strain change of the blade were measured using acoustic emission sensors and strain gages. The blade has robust structure and weighs lighter compared to conventional blade since the new blade was designed by optimization process. The 3MW blade will be commercially applied to WinDS$3000^{TM}$ in 2010.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.9 no.6
• /
• pp.170-177
• /
• 2001

Turbine blade is subject to torsional load by torsion-mount, centrifugal load by rotation of rotor and repeated bending load by steam pressure. Turbine with partially cracked blade has normal working condition at initial repair time but vibratory working condition at middle repair time due to crack growth. Finite element analysis on turbine blade indicates that repeated bending load out of all loads is the most important factor on fatigue strength of turbine blade. Therefore, this study shows root mean square roughness has linear relation with stress intensity factor range in 12% Cr steel and can predict loading condition of fractured turbine blade.

• Proceedings of the KSME Conference
• /
• 2003.04a
• /
• pp.752-757
• /
• 2003

The steady stress, modal analysis for the damaged blade was carried out to evaluate the integrity of LP 4 blade row. As a result, 4 dangerous modes for LP blade row were found in the interference diagram and it was confirmed that the nozzle passing frequency has nothing to do with the blade failure. And then the dynamic stress are analysed for the 4 dangerous modes. There are some points far out of maximum allowable stress in the cover and tenon. Therefore the blade is not safe according to the Goodman judgement. So the manufacturer have modified the design of cover and tenon. Until now, the power plant is being operated without special problems.