Abstract: In recent years, line faults of coalmine power net has happened frequently, causing a great damage to coalmine production and affecting the supply of coal resources. So it is necessary to make early warning for line fault of coalmine power net. At present, papers about early warning for line fault of coalmine power net focus on system structure construction and online measuring of circuit element. This paper proposes a hybrid reasoning method to make intelligent analysis and early warning for coalmine power net. In this method, the RBR module mainly uses the method combining AHP and evidence theory to accomplish the early warning for line fault of coalmine power net, and the RBR module uses the case adaptation method based on combined similar algorithm and RBR to complete the adaptation of early warning cases