Abstract: In this paper, the geo-electricity model of coal fires were established, and the forward modelings of Wenner α, dipole-dipole, Wenner-Schlumberger, pole-pole and pole-dipole were carried out by using the finite difference method. After adding noise interference into the forward data, the inversion was implemented to get the sectional drawing of apparent resistivity by using Gauss-Newton iteration method. The inversion results show that all the burning center, burned cave and loose zone present abnormalities in the sectional drawings of apparent resistivity of the five array types, while Wenner-Schlumberger array has the best effect in distinguishing the burning center, and the Wenner α array has the best effect in distinguishing the burned cave and loose zone. Moreover, the two array types mentioned above were applied to carry out high-density electrical detection in one fire zone, and the burning center, burned cave and loose zone were distinguished accurately. The research results provide theoretical basis and reference for the accurate detection of coal fires.

Key words: coal fires；detection；high-density electrical method；wenner &alpha, array type；wenner- schlumberger array type