In this paper, we present a dynamic calibration method for FBG sensor temperature measurement, utilizing the online sequential extreme learning machine (OS-ELM). During the measurement process, the calibration model is continuously updated instead of retrained, which can reduce tedious calculations. In particular, fiber Bragg grating (FBG) sensors are excellent candidates for sensing various physical quantities, including temperature and strain, owing to their remarkable properties like small size, high accuracy, and low energy consumption. An FBG which is used for a wide temperature range needs an expensive calibration curve measured for this particular FBG to enable the.