Research And Development Of On-Machine Verification Planning System Based On Process And Feature

Manufacturing is the main body of the national economy,and it is the foundation of industry,the path to strengthen the country,and the base of prospering the country.Flexible Manufacturing System(FMS)can be widely used in multi-variety and mass production,which greatly promotes the development of manufacturing industry.The core objective of manufacturing development is to ensure the quality of parts,and inspection is an effective means of quality assurance.Traditional offline inspection requires secondary clamping of the parts,and the results are also susceptible to human factors.In order to realize automation and real-time on machine verification,high-precision probe is mounted on CNC system,which can not only make up for the shortcomings of off-line inspection,but also directly guide the processing results to ensure the product quality.In pursuit of quality,time and cost are also important factors to be considered in production.In the multi-process manufacturing system,the processing quality of the parts will be improved with the inspection of all the processes,but the following will be the increase of cost and the decrease of manufacturing efficiency.In view of this problem,the characteristics of on machine verification are combined to investigate the inspection planning of multi-process manufacturing,so as to balance the relationship between the total production cost and the quality of parts.The main research contents are as follows:(1)The process inspection schemes are researched.The error composition of multiprocess manufacturing process is analyzed,and the object and aim of on machine verification are defined.Taking the process capability index as the quality index of parts,the quality of parts is simulated and analyzed under the two conditions that the parts are inspected or not,and the feature critical index is constructed to explain the restrictive effect of part quality on machine verification.The composition of total cost in multiprocess manufacturing process is analyzed.The total cost of parts manufacturing process is modeled in three cases: full inspection,no inspection,and sampling inspection.Particle swarm optimization is used to solve the problem,and the process and characteristics to be inspected under the conditions of satisfying the quality requirements and the low total cost of parts are planned.(2)The sampling strategy and inspection path planning of parts are researched.The NURBS surface is used to express the elementary analytic surface and the free surface in a unified way.A sampling strategy based on the nodal partitioning of the surface and the Gaussian curvature of the partitioned surface is proposed and simulated.The advantages and disadvantages of the proposed sampling strategy and the commonly used sampling strategy as well as the applicable conditions are compared.In order to ensure the safety of on-machine inspection,a point selection algorithm based on inverse calculation of surface parameters is proposed for trimmed surfaces,which can eliminate the points outside and inside the outer boundary.The surface measurement point vector is used to measure the approach direction,and the measurement path is analyzed.The ant colony algorithm is used to optimize the measurement path with the shortest total path as the target,and the path optimization effect is compared through simulation analysis.(3)Combining the process inspection schemes,the measurement points and path planning,the on machine verification planning software is developed.By using the development tools such as QT and OpenCASCADE,the software can realize the interactive display of 3D models of parts,the interactive establishment of part information library,the automatic generation of process inspection plan,the generation of measuring points and detection paths.