面向多品种小批量机加工车间的工序质量控制方法研究

Research on Process Quality Control for Multi-varieties and Small-batch Machining Workshop

针对某电梯零部件制造企业对产品制造过程质量可控性的要求,结合电梯零部件生产企业多品种、小批量的生产特点,提出基于SPC(统计过程控制)和神经网络的电梯零部件工序质量控制方法。将加工件的质量相关数据采集到MES系统中,再根据质量数据绘制出加工件的x-R控制图(均差-极值),然后建立用于控制图模式识别的神经网络模型,并使用遗传算法对神经网络的参数进行优化,使网络误差达到最小。在正确的识别了控制图模式后,可实时的监控生产过程中存在的问题并予以处理。最后,通过实例验证了基于SPC和神经网络的多品种小批量机加工车间工序质量控制方法的可行性。

Concerning the process quality control requirements for elevator component manufacturing enterprise,combined with the many varieties and small batch production characteristics in elevator component manufacturing enterprises,this paper presents a method of quality control for elevator component based on SPC and neural network. Collected the quality data of the processing parts to the MES,according to the quality data to drawthe x-R control chart,then,establish the neural network model for control pattern recognition and improve the parameters of the neural network with the method of genetic algorithms to minimize the network error. After identifying the control chart pattern correctly,the production process problems can be monitored and solved in time. Finally,an example is given to demonstrate the feasibility of the method of quality control for multi-varieties and small-batch Machining Workshop based on SPC and neural network.