Research On The Air Suction Disc Seed Metering Device’s Intelligent Design And Optimization System

Air suction disc seed metering device had been suitable for high-speed operation,had the advantages of no damage to seeds and strong adaptability to seeds,and was widely used in large-scale high-speed precision seeders.However,its traditional design method had some problems,such as large calculation,low design efficiency,long research and development cycle and high test and manufacturing costs.Therefore,the intelligent design and optimization of air suction disc seed metering device were deeply studied in this paper.The specific research contents are as follows:(1)By collecting the related doctoral papers,patent monographs,national standards and other resources,the design knowledge of air-suction disc seed metering device was summarized.Because of the heterogeneity of agricultural machinery knowledge,the design knowledge of air suction disc seed metering device was divided into agronomic knowledge,design rules and material parameters,and stored in SQLite database according to their characteristics for users to query.The knowledge representation method was studied on this basis.For regular knowledge,it was represented by production method,and for example knowledge,it was represented by ontology-based method,the Asmd ontology model of air suction disc seed metering device was developed.In the end,a knowledge management system was developed using Visual Studio as the platform,VB.NET as the development language,and SQLite as the knowledge storage tool to realize the functions of adding,deleting,modifying and querying the design knowledge of air suction disc seed metering device.(2)Based on the analysis of the design process and the characteristics of design knowledge,a hybrid reasoning process combining rule-based reasoning(RBR)and case-based reasoning(CBR)was adopted.In order to solve the problem of strong subjectivity in the assignment of attribute weights in case retrieval,the weighted similarity of cases was calculated by combining subjective weighting based on Analytic Hierarchy Process(AHP)and objective weighting based on coefficient of variation.After sorting the weighted similarity,the highest similarity example was taken as the reuse model.(3)Using Visual Studio programming software,based on VB.NET development language,integrating ADO.NET technology,Solid Works secondary development technology and parametric modeling technology,an intelligent design and optimization system of air suction disc seed metering device was built.The main functions of the system included the design of air suction disc metering device,virtual simulation test,multi-objective optimization and query of material parameter knowledge.(4)The intelligent design and optimization system of air suction disc seed metering device was used to design and optimize the soybean seed metering device.According to the seed name input by the user,the system retrieves the knowledge base to obtain the triaxial size,agronomic parameters,and related material parameters of the seed.Subsequently,the inference engine deduces the three-dimensional model of the soybean seed metering device based on the related design knowledge and outputs it to the user.A virtual simulation experiment was conducted for this model by using DEM-CFD coupling simulation technology.The test factors were the vacuum degree of suction chamber,the diameter of suction hole and the numberof suction hole,the test indexes are qualified index,reseedting index and missed-seeding index.A threefactor and five-level quadratic regression orthogonal rotation combination experiment was designed.Test results show that the qualified index of soybean seed metering device is above 88.5%.Based on the test results,taking the test factors as the optimization variables and the test indicators as the evaluation indicators,the mathematical model between the optimization variables and the evaluation indicators was constructed.The NSGA-Ⅱ algorithm was adopted and compiled in Matlab,so as to complete the multi-objective optimization of key design parameters of soybean seed metering device.Based on the optimization results,the optimized soybean seed metering device model was regenerated and stored in the system instance model base.A virtual simulation experiment based on DEM-CFD coupling technology was conducted for purpose of verifying the reliability of the optimization results.The test results showed that the qualified index is 95.83%,the reseeding index is 3.67% and the missed-seeding index is 0.5%.It’s not difficult to see the optimization results were reliable.