Font Size: a A A

High Precise Forecast Of Temperature In Electric Heating Incremental Forming And Optimization Of Process Parameters

Posted on:2020-07-30Degree:MasterType:Thesis
Country:ChinaCandidate:Z X MaoFull Text:PDF
GTID:2370330590493835Subject:Engineering
Abstract/Summary:PDF Full Text Request
TC4 titanium alloy has the advantages of corrosion resistance,heat resistance,high specific strength and fatigue resistance.It is the main lightweight structural material in aerospace manufacturing industry.The plasticity of TC4 titanium material is poor at low temperature,and the traditional furnace heating process for forming titanium alloy has the problems of high die cost and long production cycle,so it is urgent to carry out another forming technology research of titanium alloy material.Electric assisted heating incremental forming(EAHIF)has the characteristics of rapid die-less forming,rapid heating and simple device,which can meet the requirements of high temperature forming of TC4 titanium alloy and personalized,small batch and rapid manufacturing of aviation parts.Previous research on EAHIF focuses on single-point electric heating,and this method will produce such problems as large temperature gradient and tool arc drawing.The integral EAHIF,although there is little research information,can overcome the above shortcomings of single-point electric heating.In this paper,the integral EAHIF process is adopted,and the work of this paper is as follows:(1)EAHIF is analyzed theoretically,and based on the principle of resistance,current,temperature field and electric heating,the temperature condition during incremental forming is studied.According to principle analysis,the design scheme of integral electric heating device,which is different from single point electric heating,is put forward,including the structure design of tool,integral fixture and tool grip handle,and the selection of power supply,cable and infrared thermometer.(2)The integral EAHIF single factor test scheme is designed and tested.The effects of tool diameter,feed speed,vertical feed rate and current on the temperature of forming area are studied respectively.The relationship between current and vertical step and forming temperature is positive correlation,while the relationship between tool diameter and feed rate and forming temperature is negative correlation.The influence is mainly caused by the Joule heat effect of the resistance in the forming area and the heat transfer of the temperature field.(3)The Box-Behnken test scheme of response surface methodology is designed by using Design-expert software.The significant level of forming temperature affected by each process parameter of EAHIF is studied by variance analysis,and the contour and response surface diagrams between the interactive items are drawn.Among them,current is the most important factor affecting forming temperature.At the same time,the mathematical model between process parameters and target variables is established.The error is 4.09% after verification by experiment.(4)Based on the experimental data,the BP neural network prediction model of tool diameter,feed speed,vertical step,current and forming temperature is established by using MATLAB software,and the error is 4.97%.Then the weight and threshold in the topological structure are optimized by particle swarm optimization.The error of the prediction model is reduced to 2.04%.Compared with the response surface model and the unoptimized neural network model,the prediction result is more accurate.Thereby,high precision prediction of the temperature in EAHIF is realized.(5)The restriction ranges of tool diameter,feed speed,vertical step and current are set up.Based on the high-precision prediction model in the previous chapter,the process parameters are searched and optimized in the whole restriction range,then the joint optimization of three PSO-BP-GA algorithms is realized,and a better combination scheme of process parameters outside the discrete test points is found.
Keywords/Search Tags:integral electric heating, incremental forming, response surface methodology, neural network, particle swarm optimization, genetic algorithms
PDF Full Text Request
Related items