Font Size: a A A

Research On Sorting Technology Of Non-ferrous Metal Rushing Materials Of Retired Vehicles Based On Machine Vision

Posted on:2019-01-25Degree:MasterType:Thesis
Country:ChinaCandidate:Z M XuFull Text:PDF
GTID:2381330596465622Subject:Vehicle Engineering
Abstract/Summary:PDF Full Text Request
There are abundant recyclable metal resources which come from End-of-life vehicles after dismantling and crushing.Copper and aluminum are the main non-ferrous metals which have high recycling rate and market value.However,at present,there is still a lack of relevant equipments for the separation and recycling of non-ferrous metal crushing materials.This paper developed a sorting machine for non-ferrous metal based on machine vision,and focused on the key technologies and sorting rules.Based on machine vision,the sorting technological process for non-ferrous metal was designed.The overall design of the sorting system was studied,including the design and selection of visual inspection systems,transmission control mechanisms,sorting actuators and air source systems.The control flow of the electronic control system and the layout connection of the circuit elements were completed.Machine vision sorter prototype was developed.An identification scheme for non-ferrous metal crushing material was designed.The image acquisition and preprocessing techniques were specifically studied.According to the classification and recognition research of crushing color images,R,G,B,H,S,I,R-G and R-B in RGB and HSI color models were extracted as color features,CON?COR?ASM?IDM?ENT of grayscale co-occurrence matrix features serve as texture features.Random forests are used to build classifiers and study the importance of different features.The best classification feature subsets are R-G,H,S1,and the best single item classification feature is the color calculation feature R-G,which recognition accuracy is greater than 90%.According to the actual equipment conditions,a 1:1 sorting airflow field model was established,and the appropriate sorting operation parameters were selected.The numerical simulation of the sorting airflow field and the material motion trajectory was conducted to explore the influence of the number of nozzles and material factors on the trajectory and separation distance.Simulation results show that the nozzle opening number satisfies the sorting requirements well.The effect of the material parameters on the separation distance was obtained:The larger the size,the smaller the separation distance;The heavier the quality,the smaller the separation distance;Different shapes of materials and sliding attitude have a significant influence on separation distance.Five operating parameters affecting sorting performance and their respective adjustment ranges were analyzed,namely the conveying speed?v?,the main air line pressure?p?,the nozzle horizontal position?N1?,the nozzle vertical position?N2?,and the nozzle tilt angle???.A verification experiment for separation distance was conducted.The difference between the measured separation distance and the simulated data does not exceed 0.1m,verifying the accuracy of numerical simulation and material factor laws.By designing and conducting orthogonal test L16?45?,the accuracy and purity of sorting under different parameter combinations were obtained.The importance of various adjustable factors and the optimum under the experimental parameters were analyzed through the range difference and main effect diagrams.Optimum is:N1=150mm200mm;N2=-10mm;v=1.5m/s;p=0.3MPa;?=45°.This paper focused on the key technologies and sorting rules for sorting equipment,providing theoretical methods and technical support for sorting and recycling non-ferrous metals.
Keywords/Search Tags:Non-ferrous metal crushing material, Sorting machines, Machine vision, Random forests, Sorting rules
PDF Full Text Request
Related items