基于组合预测模型的车身装配工艺优化控制

Optimization Control of Auto Body Assembly Based on Combined Prediction Model

阐述了现有车身质量控制方法并总结其存在的问题,提出基于组合预测模型的车身装配工艺优化控制方法。通过加权将数据驱动模型、机理模型相融合建立组合模型,在此模型基础之上建立了一种新的预测控制方法,实现了对装配工艺参数的优化,将该方法应用于车身车门装配的质量控制案例。结果表明,所建立的组合模型的平均绝对误差预测相比单一模型有显著降低。利用所提出预测控制方法对建立的模型进行求解得到优化后的工艺参数,实现了对装配质量的预测性控制。

This paper describes the existing methods of quality control of auto body.The shortcomings of the existing auto body quality control methods are analyzed.And a new control method of auto body assembly process based on combinatorial prediction model iss proposed.Data-driven model and mechanism model are combined to build the combined model by weighting.Meanwhile,based on this model,a new predictive control method is proposed to optimize the assembly process parameter.Applying this method to the quality control case of car door assembly,the results show that the average prediction error of the combined model is significantly lower than that of other single models.When the qualified rate can’t meet the requirements,the optimized process parameters can meet the requirements by using the predictive control based on the combined model,so as to realize predictive control of the assembly quality.