At present

the transformer winding strain monitoring is mainly divided into off-line detection and on-linedetection. Due to the interference of the complex electromagnetic environment

on-line detection has not been widelyused. Although the off-line detection is more mature

it can not accurately judge the winding strain form. Based onthe above problems

this paper presents a strain gage strain detection method based on distributed fiber optic sens-ing

and proposes a winding strain identification method based on S transform and extreme learning machine（ELM）.First

the deformation of the winding in the process of transformer operation is simulated

and the corresponding Bril-louin frequency shift is collected. Then

the time-frequency analysis of the strain signal is carried out by S-trans-form

and the transformed time-frequency feature is extracted as the input sample of the neural network. Extremelearning machine（ELM）for training identification. Experimental results show that the method can effectively identi-fy the common winding deformation form

the recognition effect is better and the accuracy is high.