Abstract:
Intelligent coal mining is a new stage in the development of coal mining technology in China and intelligent perception is one of the core technologies of intelligent coal mining. Aiming at the problems of fast deformation speed,large deformation and serious floor heave of deep roadway,which make it difficult for technicians to plan a unified guiding bolt support scheme,a FBG sensor and a surrounding rock state perception monitoring system capable of sensing the axial force of bolt body are proposed. Based on the basic principle of fiber grating sensing,the key technologies of FBG dynamometry bolts are studied,including the structural design of dynamometry bolts,the working principle and strain transfer mechanism of dynamometry bolts. The strain transfer laws of the FBG sensor with surface mounted package and with grooved package is analyzed. The laws and influencing factors of strain transfer for the FBG sensor is obtained. The effect of the elastic modulus of the adhesive,the shape of the groove and the bonding length of the fiber grating on the strain transfer efficiency is studied by the orthogonal test method. Nine test schemes and three sets of bolt tension test are designed for testing simultaneously. The extreme value calculation results show that in order of the effect on strain transfer efficiency,the type of binder has the greatest impact on strain transfer efficiency,followed by groove shape and fiber grating bonding length. The strain transfer efficiency is the highest when AB epoxy resin adhesive is used,the groove shape is semicircular and the bonding length is 100 mm. The variance calculation results show that the type of adhesive has a significant effect on the strain transfer efficiency,followed by the groove shape and then the effect of the fiber grating bonding length on the strain transfer efficiency is not significant. Finally,a surrounding rock state perception monitoring system was carried out in the return airway 15106 in Sijiazhuang coal mine,and the application effect was remarkable.