Research on printing path planning based on AM-PSO-BP neural network
In order to improve the welding effect of arc welding,an arc welding printing path planning method based on AM-PSO-BP neural network is proposed.Firstly,adaptive mutation(AM)based on adaptive variance was adopted to solve the problem of low iterationefficiency of particle swarm optimization(PSO)in the later stage.Then,AM-PSO algorithm was utilized to optimize the weight and threshold of BP(back propagation)neural network,so as to optimize the parameters of BP neural network.Finally,AM-PSO-BP neural network algorithm was used to predict the parameters of arc welding printing process,thereby obtaining more accurate parameters of arc welding printing process.The simulation results show that the proposed method can accurately predict the parameters of arc welding printing process.Under the process parameters,the error between the predicted value and the measured value of the hexagonal cylinder,cylinder and cube printed by arc welding is small,and the error is within the allowable range,which has higher accuracy.It shows that the above method can provide a basis for accurate arc welding printing.
万方数据
维普
