Improved GABP Algorithm Of Abrasive Water Jet Cutting Quality Intelligent Control Research

The cutting operation of oil and gas transmission pipeline usually in oil and gas environment.If there is a open flame or high temperature during cutting in such situation,combustion or explosion could occurred and safety accidents could be caused.Therefore,a safe cutting method is urgently needed.Abrasive water jet cutting is a safe and cold cutting method developed on the basis of high pressure water jet technology.Because of its unique cold cutting characteristics,fire or high temperature does not produce during cutting process,it is more suitable for cutting oil and gas pipeline than traditional cutting method.Because abrasive water jet is a kind of "soft cutting tool",it causes regular arc stripes on the cutting section of workpiece during actual cutting process and rough surface.Oil and gas pipeline has strict requirements of cutting quality and fire prevention control.The establishment of cutting quality model can guide the precision cutting of abrasive water jet and obtain the required cutting quality.Based on the study of abrasive water jet cutting quality model,the research topic of the quality intelligent control of abrasive water jet cutting was put forward.Abrasive water jet cutting experiments were carried out on Q235 steel,which commonly used in oil and gas pipeline.In view of the relationship between the cutting quality and the process parameters of abrasive water jet is very complicated,the traditional mathematical method can not be used for modeling.The influence of cutting speed,standoff distance,abrasive flow and abrasive grain size on cutting quality respectively were studied.The comprehensive test and range analysis were investigated.The quality control model of abrasive water jet cutting based on BP neural network is established.The genetic algorithm(abbreviated GA)is applied to the cutting quality control algorithm of the model.The quality control model of abrasive water jet cutting is established by improved GABP algorithm to avoid the model falling into local minimum in the learning process.The results show that the quality control model of abrasive water jet cutting with improved GABP algorithm has better prediction accuracy and can be used to predict the cutting speed effectively.The effective control of the cutting quality of the material by water jet can be realized.On the basis of improving the quality control model of abrasive water jet cutting in GABP algorithm,combined with VC++ programming languagethe,Matlab neural network toolbox,genetic algorithm toolbox.the intelligent control module of abrasive water jet cutting quality is developed.The intelligent selection and visualization operation of the process parameters can be realized in the intelligent control module.