There are various uncertainties in the belt conveyor system. Therefore, it is necessary to describe accurately and compensate for them for reducing system errors and control costs. This paper proposes a method by applying fuzzy set theory to establish a fuzzy dynamic model, including the belt and PMSMs. Then, an exponential robust controller with an adaptive tension force is designed to achieve the fuzzy belt-PMSMs system uniform boundedness and uniform ultimate boundedness. Finally, a series of simulation experiments are performed to verify the effectiveness of the proposed method. The simulation experiment implies the advantages of our method in reducing system errors and saving control energy.