Research And Application Of Closed-Loop Variation Analysis System Based On Dimensional Engineering

The dimensional accuracy of the product is an important factor affecting the appearance quality and functional characteristics.In the multi-station assembly process,all kinds of variation sources will use the product as a carrier to produce coupling,transmission and accumulation to form a comprehensive variation,which will affect the dimensional accuracy of the product and the manufacturing process.At present,most researches on dimensional accuracy control are focused on tolerance design,diagnosing variation sources and optimizing strategies formulation under design tolerance conditions,which increases the degree of mismatch with the actual assembly results.Therefore,a closed-loop variation analysis system based on dimensional engineering is established in this paper,which strives to make new progress in product dimensional accuracy control,providing new methods and means for product quality improvement.The main research work of this paper is as follows:(1)A closed-loop variation analysis system based on dimensional engineering was established.Based on the relevant theoretical basis of the closed-loop control and variation analysis,from the perspective of dimensional engineering,this paper established a closed-loop variation analysis system with the core content that was the actual measurement data associated with the variation analysis model of parts to conduct assembly simulation and calculation.In the system,by comparing the analysis results under the design input and the plant data,the diagnosis of the real variation source and the formulation of the optimization strategy were carried out,and the prediction effect of the variation model under the plant data was verified by using the measurement data.The system can truly reflect the assembly process.It has been verified by the example of the engine compartment of the automobile to determine the real variation source and predict the effectiveness of the assembly results.(2)A closed-loop variation analysis of the automobile engine compartment was carried out.Based on the analysis of the engine compartment assembly level,DTS requirements,locating strategy and tolerance definition,the variation model of engine compartment was completed.furthermore,the variation analysis model based on plant data was created by combining the collected single piece measurement data and the data set associated with the 3DCS.The results show that under the design input,if the out of specification was 5% as the baseline,the engine compartment variation model dimension verification was passed.Under the plant data,the results of the engine compartment variation analysis were consistent with the assembly quality problems reported by the plant,indicating that the variation analysis model based on plant data can effectively characterize the actual assembly process,and the validity of the variation model prediction result has also been verified by the actual measurement data.By comparing the HLM results and Geo Factor values based on design input and plant data,the true variation sources were determined,and three effective optimization strategies were proposed to solve the quality problem that the X direction of headlight mounting holes of the engine compartment were out of tolerance.(3)Locating strategy optimization based on Particle Swarm Optimization algorithm was proposed.Aiming at the experience and contingency in the optimization strategy formulation process,a locating strategy optimization method based on Particle Swarm Optimization was further proposed.It was based on the principle of parts location,the grid node set of parts was feasible region,the minimum sum of squares of the variation value of the measurement point was the objective function,the 3DCS was used to conduct the variation modeling and calculation,and the positioning point layout was optimized with the Particle Swarm Optimization.The validity and feasibility of the method was verified with the example of automobile engine compartment right front sill three-stage assembly.The results show that,compared with the initial design,the layout of the positioning point after optimization conformed to the principle of maximum area,and the objective function value,the variation mean and maximum value of the measurement point were reduced by 40.69%,21.55% and 12.83% respectively,which effectively solved the problem of low assembly accuracy of parts due to unreasonable locating layout.