Study On Fine Microstructure Analysis And Macroscopic Response Mechanism Of Straw-Potato Residue Mixture Briquetting

To analyze the micro-forming mechanism and improve the forming quality of straw-potato residue mixture briquettes,this study was conducted with the straw-potato residue mixture briquette as the research object,based on the existing molding press and its related mechanical properties parameters,as well as macro-forming quality parameters.Observations were made on the surface microscopic morphology of the briquette using a stereomicroscope and other equipment,and diagrams were collected.And based on Open CV diagram processing and Py Torch machine learning algorithms,extracted and analyzed the mesoscopic and microscopic feature information of the surface diagrams of straw-potato residue mixture briquettes collected,and established a correlation with macro-forming quality and mechanical properties parameters of the briquettes.The main research contents are as follows:(1)The surface crack and the solid bridge of the microscopic image are taken as the target features,and the data set is made after data enhancement.The pre-processing method of the fine-scale diagrams was determined,which involved sequentially processing the diagrams using Open CV-based diagram processing algorithms such as grayscale,Prewitt operator edge detection,binarization,expansion,and diagram filtering,to ensure that the processed diagrams meet the requirements for extracting fine-scale feature parameters.(2)The surface cracks of the fine-scale diagrams and solid bridges of the micro-scale diagrams were labeled using labelme to generate JSON files,and the dataset was created by data augmentation.The Vgg16-Unet and ResNet-Unet deep learning models were selected for comparison.By optimizing the parameters of the deep learning model,the ResNet-Unet deep learning model was finally selected for target detection,which was used to extract the surface morphology characteristic parameters of the briquette.The accuracy of crack recognition on the surface of the briquette can reach98%,and MIo U can reach 81.52%.The accuracy of the model for solid bridge recognition can reach 92%,and MIo U can reach 79.21%.Statistical analysis of the extracted feature parameters revealed that a higher temperature,lower compression rate,greater pressure and feeding amount,and a higher proportion of potato residues in the mixed particles resulted in a lower number of surface cracks and more solid bridges in the straw-potato residue mixture briquettes during compression.(3)Drop tests were conducted on straw-potato residue mixture briquettes produced at different forming conditions to obtain the crushing strength,which is a measure of the briquette quality.The correlation and significance tests were performed among the macroscopic mechanical parameters,forming quality parameters,and surface micro-fine feature parameters of the straw-potato residue mixture briquettes,in combination with the existing forming conditions and compression process mechanical characteristic parameters.The relationships between the number of solid bridges,the number of cracks and forming conditions,forming quality parameters,and macroscopic mechanical parameters were obtained.Based on the results,the microstructure morphology and macroscopic response conditions of the straw-potato residue mixture briquettes were analyzed to provide a basis for optimizing the forming conditions.Furthermore,the forming mechanism of the straw-potato residue mixture briquettes was analyzed from the perspectives of bonding mechanism and plastic deformation.The binding mechanism of the briquettes was analyzed from a micro-scale perspective.