Application of Random Forest Algorithm in the Prediction of Electrohydrodynamic Inkjet Printing Droplet Diameter
Electrohydrodynamic inkjet printing technology is a new type of micro-nano-scale additive manufacturing method.Due to the low cost,simple process,high efficiency and high precision of Electrohydrodynamic inkjet printing,it has great poten-tial application in fabrication of microfluidic chips,printed-flexible electronic devices,and various optical devices,etc.Therefore,taking the droplet diameter prediction of electrohydrodynamic inkjet printing as the research object,a model based on the random forest regression algorithm is developed,in order to predict the printed droplet diameter under the condition of several key parame-ters(base voltage,high voltage,pulse frequency,distance between the print head and the substrate,material conductivity and vis-cosity).The prediction result shows that the prediction accuracy of the model is significantly higher than that of the theoretical modeling and regression analysis.Among several common algorithm prediction models,random forest regression model has the highest prediction accuracy when the training samples are limited.The droplet diameter prediction model based on random forest regression algorithm provides an economic and effective method for droplet diameter prediction and control of electrohydrodynam-ic inkjet printing.
NETL
NSTL
万方数据
