The invention discloses a quick positioning device and method for a
stress concentration area in a static test of a
wind power blade. According to the device, strain data of key parts of a blade are collected in real time through a multi-type
sensor array (strain gauges,
optical fiber sensors and piezoelectric films), and a
stress concentration point is accurately positioned in combination with an intelligent
algorithm (finite
element analysis,
machine learning and
stress gradient calculation). The method comprises the steps of graded loading testing, data closed-loop feedback and three-dimensional stress nephogram generation, the positioning precision error is smaller than or equal to 2%, and the time is shortened to 1 / 10 of that of a traditional method. The technology is suitable for blade design
verification, operation and maintenance detection and
material fatigue analysis, the detection efficiency is greatly improved, and the labor cost is reduced. The
sensor array is composed of a
surface strain gauge (the grid spacing is smaller than or equal to 5 mm), an embedded
optical fiber sensor (the sensitivity is larger than or equal to 1 mu epsilon) and a piezoelectric film sensor (the
response frequency is larger than or equal to 1 kHz) and covers the
front edge, the rear edge, the main beam, the root and other key parts of the blade. The intelligent
algorithm module comprises a finite
element model library (presetting different types of blade structures), a
machine learning model (TensorFlow / PyTorch framework training) and a
stress gradient calculation unit, and material attribute errors are corrected through historical data; the testing method supports bending, torsion and combined load working conditions, the positioning time is only 0.5-1 hour, and the testing method is compatible with 1.5-10MW blade models.