Abstract: In view of problems of poor ability of automation cutting, bad stability and difficult real-time identification of dynamic load of cutting of extra-heavy type rock roadheader, a kind of dynamic load identification device for cutting mechanism of rock roadheader based on RBF neural network was designed according to characteristics of mining roadway, and designs of hardware and software of the identification device were introduced in details. To take industrial computer as analysis center, the device can real-timely monitor signals of cutting motor current and cutting head vibration, and uses multi-sensor fusion technology and RBF neural network to realize signal processing and intelligent analysis, so as to realize feature extraction and recognition of dynamic load of cutting rock and achieve effective distinction of rocks with different hardness.