Study On Multiscale Analysis And Control Method Of Process-Induced Distortion For 3D Woven Composites

3D woven composites has become one of the most promising solutions to overcome the bottleneck problems of unidirectional and 2D fabric laminated composites,such as low impact resistance,poor interlaminar performance and high manufacturing cost,thanks to the continuous and multi-directional connection of fibers,and without the use of expensive molding equipment such as the autoclave.However,because of the influence of material properties and external forming conditions,residual stresses and process-induced distortion will inevitably appear,which will seriously affect the service performance of the parts,reduce the forming accuracy,and even lead to the scrapping of the parts.Although a lot of study on the processinduced distortion of composites has been carried out,most of them are limited to the conventional laminated composites.For 3D woven composites,the complex microstructure determines the uniqueness of the residual stress generation mechanism.The existing research results are difficult to accurately predict the macroscopic and microscopic residual stresses and distortion of 3D woven composites,and there is a lack of effective distortion control methods.Therefore,aiming at the problem that it is difficult to predict and control the process-induced distortion of 3D woven composites,the monitoring method of the 3D woven composite curing process was proposed in this dissertation,and the evolutions of internal temperature and strain during the curing process were obtained.Then,a multiscale analysis method of residual stresses of the 3D woven composite was built up,the macroscopic and microscopic residual stresses were predicted accurately.Subsequently,the multi-objective optimization methodology for the cure cycle of the 3D woven composite was established,which reduced the microscopic residual stresses of 3D woven composites and shortened the processing time from the mesoscale.Finally,the tool-part interaction modeling method of the 3D woven composite was established,and an improved distortion compensation scheme was proposed,which improved the forming precision of the parts from the macroscale.The main contents of the dissertation are as follows:(1)To study the thermo-chemical behavior of thermosetting resin in the curing process,the curing kinetics equation of LY 1564 epoxy resin was established based on non-isothermal differential scanning calorimetry.At the same time,the temperature sensitivity coefficient of the fiber Bragg grating(FBG)sensor was calibrated.It was found that the temperature sensitivity coefficient of the FBG sensor changed bilinearly with the temperature taking 120?as the cut-off point.Furthermore,a method for monitoring the curing process of the 3D woven composites was proposed.By mixing the FBG sensors with the weft yarns,and then weaving it into the designated position of the preform synchronously by 3D weaving technique,the survival rate of FBG sensors was effectively improved,and the internal temperature and strain evolutions of 3D woven composites were obtained.(2)According to the multiscale characteristics of 3D woven composites,the representative volume elements of fiber scale and yarn scale were established,and the degree of cure stepping method was proposed.Combined with the Cure Hardening Instantaneous Linear Elastic(CHILE)model,the modulus evolution model of the 3D woven composites during the curing process was established.At the same time,the equivalent temperature load method(ETLM)was proposed to establish the curing shrinkage strain model of the 3D woven composites.On this basis,the multiscale-multiphysics coupling analysis of the curing process of 3D woven composites was carried out,and the accuracy of the model was validated by comparing with the experimental results.The analysis results of show that the maximum microscopic residual stresses in the 3D woven composite parts is much higher than the maximum macroscopic residual stresses.(3)The yam scale curing process analysis model of the 3D woven composite was established,and the Elliptic Basis Function(EBF)of the curing process simulation was established by three sequential sampling methods based on spatial information,error information and result information,respectively.Based on the EBF surrogate model,the global sensitivity analysis was carried out,and the influence of the process parameters on the microscopic residual stresses and process time was revealed.Based on the fast and elitist Nondominated Sorting Genetic Algorithm(NSGA-II)method,the multi-objective optimization of the cure cycle was carried out,and the Pareto front of the curing system was obtained.The results show that the optimized process cycle can be shortened by 80%at most,and the average residual stress can be reduced by 14%at most.(4)The analysis model of tool-part interaction for the 3D woven composites RTM process was established,and the accuracy of the model was validated by comparing with the experimental results.Based on the model,the relationship between the distortion and compensation of 3D woven composites plate and L-shaped parts was studied,it is found that the process-induced distortion is related to the geometry.Then an improved distortion compensation scheme was proposed and a typical L-shaped part was comprehensively compensated.The results show that 99.2%of the flange warpage and 98.8%of the angle springin distortion are effectively compensated by the proposed deformation compensation scheme.