Considering the low identification accuracy of single-phase-to-ground(SLG) fault in distribution systems with the neural point small-current grounded, breaking through the traditional fault identification method based on voltage or current, this paper focuses on mining the features of electric field variation when SLG fault occurs. Firstly, based on the charge simulation method, the electric field at the same height under three-phase distribution lines and that on the line surface during SLG fault are respectively calculated. Accordingly, the variation law of electric field in small-current grounded distribution systems during SLG fault is obtained theoretically, and the corresponding relationship between the variation mode of electric field and the occurrence of SLG fault is established. Besides, MEMS sensor is innovatively adopted for electric field measurement in distribution systems by taking advantage of its small size, low power consumption and high accuracy, and its integration scheme with fault indicator is formed. On this basis, a SLG fault identification method and corresponding devices based on the variation of electric field are proposed, and the effectiveness is verified by experiments.