Optimization And Analysis Of GDI Engine Spray Calibration Process

This thesis focus on the optimization and anslysis of GDI engine spray calibration process.It will cost much to match the spray experimental data,if numerical model is adjusted manually.This thesis tries to use the self-built optimization platform to perform spray calibration automatically.The thesis has built spray bomb experiment platform,and perform experiments under four conditions,the spray image are recorded by camera.Then the thesis use Matlab to process the data to get penetration and spray cone angle.The thesis also use semi-empirical formula to get the SMD information,the sampling area is 40 mm under the injector tip.The thesis discussed how injection pressure and background pressure influence the penetration,cone angle and SMD data as well.The thesis uses STARCD to build 3-D CFD spray bomb model,and use Rosin-Rammler function to model the initial droplet diameter distribution.The thesis uses KHRT breakup model to simulate the secondary break-up process.The thesis uses user subroutine dropro.f to model the liquid properties,and use posdat.f to enhance the post-processing capbility.The thesis gets penetration,SMD information through CFD calculation.The thesis uses HEEDS to build the optimization platform,and set up eight variables.They are X and Q in Rosin-Rammer,khb0,khb1,rtc3 and rtct in KHRT break-up model,spray parcel numbers and nozzle diameter.The thesis set up two targets,one is the SMD and the other is penetration.The thesis use Python to enhance optimization post-processing.The thesis use the 2 level full factorial DOE method to analyse the variables sensitivity.The thesis use the SHERPA optimization algorithm to run for 250 cases continuously.The thesis finished experiment and CFD calculation for a spray bomb,clarify the whole process,and tell how variables influence the spray results.The practice could be applied to engineering calibration.