Abstract: Due to the difficulty of knowledge accessing in traditional gear-box intelligent diagnosis system,a diagnosis technology based on CBR(Case-based Reasoning) and RBR(Rule-based Reasoning) hybrid method was proposed in this paper.On the basis of analyzing the advantages and disadvantages of the two methods,the combination of the reasoning technologies was utilized in fault diagnosis of gearbox of rolling mills.As a result,the accuracy and efficiency of fault diagnosis were improved greatly.Due to the shortage in similarity algorithm in traditional retrieval,a new case retrieval algorithm was proposed.Therefore,the problem of inaccuracy in traditional similarity algorithm was solved effectively.