Prediction of assembly height of tapered roller bearing based on WOA-BP neural network

Abstract To solve the problem of low matching efficiency of inner and outer rings during tapered roller bearing assembly, a BP neural network based on the Whale optimization algorithm (WOA) was constructed. Firstly, the influencing factors of the assembly height of tapered roller bearings are identified and taken as input parameters, and the assembly height is taken as output. Then, the BP neural network is optimized by the WOA algorithm, and the high prediction model of the tapered roller bearing assembly is established. Finally, the real results are compared to the predicted data. We can obtain that the determination coefficient of the assembly height prediction model is 0.9985, and the overall error is less than 0.0051%, which can accurately predict the assembly height of the tapered roller bearing.