Research On Parameters Optimization And Process Database Of Fiber Laser Sharpening For Bronze-bonded Diamond Grinding Wheel

Bronze-bond diamond grinding wheels are characteristic of high hardness and abrasionresistance. In order to keep good grinding performance, bronze-bond diamond grinding wheelrequires to be regularly sharpened because it has poor self-sharpening. However, traditionalsharpening has big tool loss, low efficiency and precision and also is unfriendly to theenvironment, which hampers the wide range of applications of bronze diamond wheel. Basedon the fiber laser, parameters optimization for laser sharpening bronze-bond diamondgrinding wheels is done and process database of fiber laser sharpening for bronze-bondeddiamond grinding wheel is researched as well.Firstly, the fiber laser sharpening system for superabrasive grinding wheel is brieflydescribed, and the ablative morphology of bronze-bond wheels under different power isanalyzed by ultra-digital microscope. Then based on the experimental result of laser ablationabove, laser sharpening bronze-bond wheel experiments are conducted, and also ultra-digitalmicroscope is used to observe the sharpening morphology between before laser sharpeningand after laser sharpening under different repetition frequency.Secondly, neural network prediction model between the process parameters and thespecimen surface roughness value is established by BP neural networks with the powerfulnonlinearity approach capability, five groups of experiment data except the training sampledata are selected to test the model, and the relative error between the predicted value from BPnetwork and observed value was lower than5%, this illustrates that good mapping relationbetween process parameters and the specimen surface roughness value is formed and themodel is of high approximation accuracy.Thirdly, the basic process of process parameters optimization based on particle swarmoptimization is analyzed, then optimization mathematical model is established and the fitnessfunction is constructed. Combined the BP neural network prediction model with particleswarm optimization algorithm, the process parameters of laser sharpening grinding wheel areoptimized. Five groups of optimized process parameters show that this optimization method ishelpful to obtain the expected quality.Lastly, the function requirements, design principles and implementation platform of thelaser sharpening grinding wheel process parameters database are represented. Then, generalframework of the database is designed. Finally the database tables and their relationships oflaser sharpening grinding wheel database are elaborated, the various business implementationmethod of the business logic layer and the concrete implementation process ofhuman-computer interaction interface are described as well.