基于改进BP神经网络的烟草收获机械故障诊断研究OA北大核心

Tobacco harvesting machinery is an important technical support in tobacco production and an important guaran-tee to improve the efficiency of tobacco harvesting,but due to the complex internal structure of tobacco harvesting machinery,it is very easy to cause mechanical operation failure in the process of use.With the rapid development of big data and sensor technology,the prediction and diagnosis of mechanical failure based on artificial neural network model is an important technology to improve the efficient operation of tobacco harvesting machinery.This study proposed an im-proved BP neural network model,and constructed a gear fault diagnosis model based on GA-BP neural network model for tobacco harvesting machinery,and conducted experimental verification by selecting signals of gear wear,gluing,cracking,broken teeth and normal gears.The results showed that the improved BP neural network model has MAPE of on-ly 0.87%,RMSE of 1.12,MAE of 0.92,MSE of 1.19,this fault diagnosis accuracy meets the actual needs of tobacco harvesting production and has been greatly improved in terms of model algorithm and computation speed.