• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.870-881
• /
• 2020

Because SCADA monitoring data of wind turbines are large and fast changing, the unbalanced proportion of data in various working conditions makes it difficult to process fault feature data. The existing methods mainly introduce new and non-repeating instances by interpolating adjacent minority samples. In order to overcome the shortcomings of these methods which does not consider boundary conditions in balancing data, an improved over-sampling balancing algorithm SC-SMOTE (safe circle synthetic minority oversampling technology) is proposed to optimize data sets. Then, for the balanced data sets, a fault diagnosis method based on improved k-nearest neighbors (kNN) classification for wind turbine blade icing is adopted. Compared with the SMOTE algorithm, the experimental results show that the method is effective in the diagnosis of fan blade icing fault and improves the accuracy of diagnosis.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.26 no.5
• /
• pp.551-560
• /
• 2020

Recently, the offshore wind power generator market is expected to grow significantly because of increased energy demand, reduced dependence on fossil fuel-based power generation, and environmental regulations. Consequently, wind power generation is increasing worldwide, and several attempts have been made to utilize offshore wind power. Norway's Petroleum Safety Authority (PSA) requires a leg-structure design with a collision energy of 35 MJ owing to the event of a collision under operation conditions. In this study, the results of the numerical analysis of a wind turbine installation vessel subjected to ship collision were set such that the maximum collision energy that the leg could sustain was calculated and compared with the PSA requirements. The current leg design plan does not satisfy the required value of 35 MJ, and it is necessary to increase the section modulus by more than 200 % to satisfy the regulations, which is unfeasible in realistic leg design. Therefore, a collision energy standard based on a reasonable collision scenario should be established.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2001.04a
• /
• pp.91-94
• /
• 2001

A structural test of the wind turbine rotor blade must be required to evaluate the uncertainty in design assessment due to use of material, design concepts, production processes and so on, and the possible impact on the structural integrity. In the full-scale static strength test, the measuring parameters are strain, displacements, loads, weight and the center of gravity. There are test equipments, measuring sensors, a test rig and fixtures to obtain measuring parameters. In order to simulate the aerodynamics load, the three-point loading method instead of the one-point loading method is applied. There is slightly some difference between the measured results and the predicted results with the reference fiber volume fraction of 60%. However, the agreement between the measured results and the predicted results with the actual fiber volume fraction of 52.5% is good. Even though a slightly non-linearity from 80% loading to 100% loading, a linear static solution is sufficient for the design purpose as the amount of the non-linearity is relatively small. Comparison between measured and predicted strain results at the maximum thickness positions of the blade profile for 0.236R(5.56m), 0.493R(11.59m) and 0.574R(13.43m), under 20%, 40%, 60%, 80% and 100% loadings for the upper part of the blade. The predicted values are in good agreement with the measured values.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.5 no.1
• /
• pp.76-81
• /
• 2001

A structural test of the wind turbine rotor blade is to evaluate the uncertainty of design due to selection of material, design concepts, production processes and so on, and their possible impacts on the structural integrity. In the full-scale static strength test, the measuring parameters are strain and displacements vs. loads, weight and the center of gravity. In order to simulate the aerodynamics load, the three-point loading method is applied. There is slight difference between the measured results and the predicted results for the reference fiber volume fraction of 60% . However, the agreement between the measured results and the predicted results with the actual fiber volume fraction of 52.5% is good. Even though a slightly non-linearity from 80% loading to 100% loading exists, a linear static solution is sufficient for the design purpose due to te small amount of non-linearity. Comparison between measured and predicted strain results at the maximum thickness positions of the blade profile for 0.236R(5.56m), 0.493R(11.59m) and 0.574R(13.43m), under 20%, 40%, 60%, 80% and 100% loadings for the upper part of the blade. The predicted values are in good agreement with the measured values.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.30 no.3
• /
• pp.236-243
• /
• 2010

Fiber Bragg grating(FBG) sensor arrays can be used to monitor the mechanical behavior of the large composite structures such as wind turbine rotor blades and aircrafts. However, brittle FBG sensors, especially multiplexed FBG sensors are easily damaged when they are installed in the flexible structures. As a protection of brittle FBG sensors, epoxy packaged FBG sensors have been presented in this paper. Finite element analysis and experiments were performed to evaluate the effects of adhesives, packaging materials and the bonding layer thickness on the strain transmission. Two types of epoxy were used for packaging FBG sensors and the sensor probes were attached with various bonding layer thickness. It was observed that thin bonding layer with high elastic modulus ratio of the adhesive to packaging provided good strain transmission. However, the strain transmission was significantly decreased when elastic modulus of the adhesive was much lower than the packaged FBG sensor probe's one.

• Journal of Ocean Engineering and Technology
• /
• v.30 no.4
• /
• pp.260-267
• /
• 2016

Considering the effect of dynamic response amplification, a reliability analysis of an offshore wind turbine support structure under an earthquake is presented. A reliability analysis based on the dynamic response requires a large amount of time when using not only a level 3 approach but also level 2 such as a first order reliability method (FORM). Moreover, if a limit state is defined by using the maximum stress at a structural joint where stress concentration occurs, a three-dimensional element should be used in the finite element analysis. This makes the computational load much heavier. To deal with this kind of problem, two techniques are suggested in this paper. One is the application of a quasi-static structural analysis that takes the dynamic amplification effect into account. The other is the use of a stress concentration factor to estimate the maximum local stress. The proposed reliability analysis is performed using a level 2 FORM and verified using a level 3 simulation approach.

• Journal of the Institute of Convergence Signal Processing
• /
• v.25 no.2
• /
• pp.94-99
• /
• 2024

Recently, wind power has been gaining attention as a highly efficient renewable energy source, leading to various technological developments worldwide. Typically, wind power is operated in the form of large wind farms with many wind turbines installed in areas rich in wind resources. However, in developing countries or regions isolated from the power grid, off-grid small wind power systems are emerging as an efficient solution. To efficiently operate and expand off-grid small-scale power systems, the development of real-time monitoring systems is required. For the efficient operation of small wind power systems, it is essential to develop real-time monitoring systems that can actively respond to excessive wind speeds and various environmental factors, as well as ensure the stable supply of produced power to small areas or facilities through an Energy Storage System (ESS). The implemented system monitors turbine RPM, power generation, brake operation, and more to create an optimal operating environment. The developed small wind power system can be utilized in remote road lighting, marine leisure facilities, mobile communication base stations, and other applications, contributing to the development of the RE100 industry ecosystem.

• Journal of Ocean Engineering and Technology
• /
• v.31 no.3
• /
• pp.202-207
• /
• 2017

This paper presents the results of an experimental study on the motions and mooring characteristics of a floating vertical axis wind turbine system. Based on a comparison of regular wave experiment results, the motions of structures with different types of mooring are almost the same. Based on the tension response results of a regular wave experiment with a catenary mooring system, the mooring lines in front of the structure have a larger tension effect than the back of the structure by the drifted offset of the structure. The dynamic response spectrum of the structure in the irregular wave experiments showed no significant differences in response to differences in the mooring system. As a result of the comparison of the tension response spectra, the mooring lines have a larger value with a drifted offset for the structure, as shown in the previous regular wave experiment. The results of the dynamic response of the structure under irregular wave and wind conditions showed that the heave motion response is influenced by the coupled effect with the mooring lines of the surge and pitch motion due to the drifted offset and steady heeling. In addition, the mooring lines in front of the structure have a very large tension force compared to the mooring lines in back of the structure as a result of the drifted offset of the structure.

• Journal of Wind Energy
• /
• v.13 no.4
• /
• pp.80-89
• /
• 2022

In this paper, a study was conducted to evaluate the safety of pitch reducers for 8 MW class wind turbines. The housing and carrier of the pitch reducer were subjected to structural analysis for the ultimate load by load duration distribution (LDD). As a result of the finite element analysis of the housing parts, the part with the highest stress was the output housing, and the equivalent stress was 522.4 MPa and the safety factor was 1.14. As a result of finite analysis of the carrier, the highest stress occurred at 80.5 MPa in the first carrier, and the safety factor was 10.3. In addition, extreme strength and life analysis by LDD load were performed for gears and bearings included in each stage. The strength analysis of the planetary gear train was conducted based on ISO 6336, and the stability evaluation of the bearings through life analysis based on ISO 281 found all to be safe.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.40 no.8
• /
• pp.543-556
• /
• 2016

In this work, an insulated coupling test machine for a 5-MW-class wind turbine was designed and developed, along with the public performance testing of a 3-MW-class wind turbine. The results of the device design, development requirements, functional considerations, structural vibration analysis, and the evaluation of the insulated coupling test machine are presented in this study. For the coupling models, thick fiberglass composite pipe insulation, fabricated by filament winding, was considered. Results of three-dimensional finite element analysis conducted using both solid element and shell element modeling were analyzed and compared, considering the effect of thickness. In addition, results from the nonlinear finite element analysis of multiple leaf springs of the laminated disk pack structure were verified and compared with experimental data.