One of the essential parts of a wind power generator that captures wind energy is the wind turbine blade. The safety of the blades rapidly declines as a wind turbine's operating period grows. For real-time monitoring, a chip-type pre-stressed fiber Bragg grating (FBG) strain sensor was fabricated. The sensor's structure was improved using simulation analysis along with optimization. It was discovered through calibration trials that the pre-stressing method expanded the sensor's range of measurement, guaranteed overall linearity, and prevented the potential hysteresis phenomena during compression. The sensor's final sensitivity was calculated to be 1.970 pm/με, and its linear fitting coefficient was 0.999. Finally, the sensor was used to monitor the wind turbine blades and the strain change curve of the root of a normally functioning blade is found to be a sine curve, which provides a certain reference value for judging whether the blade is damaged in the future.