研究目的
To develop and evaluate thermography techniques, including active and passive methods, for monitoring the integrity and detecting subsurface defects in wind turbine blades, with a focus on improving image quality through processing algorithms.
研究成果
Thermography techniques, particularly when enhanced with image processing, are effective for detecting subsurface defects in wind turbine blades. Active thermography with step heating and SPAT provided the best results, with a minimum detectable defect diameter-to-depth ratio of 1.33. Passive thermography was most effective in the morning and noon, but image processing improved quality and eliminated false indications. The methods significantly increased SNR, demonstrating their potential for non-destructive testing in wind energy applications.
研究不足
The study was limited to laboratory and outdoor conditions with specific samples; it did not include overnight tests or real-time operational wind turbines. The Matched Filters method requires manual selection of sound areas, which is time-consuming and may affect results. Small defects (e.g., 4mm diameter) were not detectable regardless of depth.
1：Experimental Design and Method Selection:
The study employed both active (pulsed and step heating) and passive thermography techniques. Active thermography used flash and halogen lamps for heating, while passive thermography utilized solar radiation. Image processing algorithms such as Matched Filters (MF) and Step Phase and Amplitude Thermography (SPAT) were applied to enhance thermal contrasts.
2：Sample Selection and Data Sources:
Samples included a 3m long damaged wind turbine blade section and a defect plate (170mm x 195mm x 8mm) cut from the blade laminate, with flat-bottomed holes of varying diameters (4mm to 20mm) and depths (0.5mm to 3mm) drilled from the rear to simulate defects.
3：5mm to 3mm) drilled from the rear to simulate defects.
List of Experimental Equipment and Materials:
3. List of Experimental Equipment and Materials: Equipment included a FLIR T1030Sc IR camera with a 21.2mm lens, a 2400W flash lamp, two 500W halogen lamps, and ResearchIR software. Materials were fiberglass composite samples from a wind turbine blade.
4：2mm lens, a 2400W flash lamp, two 500W halogen lamps, and ResearchIR software. Materials were fiberglass composite samples from a wind turbine blade.
Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: For passive thermography, the blade was monitored outdoors on a sunny day from 9:00 a.m. to 7:30 p.m., with IR camera at 4m distance. For active thermography, the defect plate was heated with flash or halogen lamps, and thermal images were recorded at 15Hz. Image processing was applied to raw data to improve defect visibility.
5：Data Analysis Methods:
Signal-to-noise ratio (SNR) was calculated for quantitative evaluation. Image processing involved algorithms like MF (SAM, ACE, t-statistic, F-statistic) and transform-based techniques (FFT for phase and amplitude analysis).
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
