Abstract: The discrete assembly workshop of complex products for multi-variety variable batch production is a typical complex manufacturing system.Its operation process presents the characteristics of high complexity,strong dynamics and many uncertain factors.To realize the dynamic scheduling of the discrete assembly workshop of complex products,a digital twin-based dynamic scheduling method for discrete assembly workshops of complex products was proposed,which realized more precise dynamic scheduling through data interaction between physical space and virtual space.A framework and a scheduling problem model for the discrete assembly workshop of complex products based on digital twins were constructed.On this basis,the online prediction of man-hours based on real-time data and neural networks,and the dynamic scheduling of digital twin workshops for discrete assembly of aerospace products were elaborated.The key technology was realized,and finally a complex product discrete assembly shop scheduling system was built.The scheduling process of an aerospace product assembly shop was taken as an example to illustrate the effectiveness and feasibility of the proposed method.

Key words: digital twin, dynamic scheduling, discrete assembly shop-floor, complex product, working hours forecast

摘要： 面向多品种变批量生产的复杂产品离散装配车间是典型的复杂制造系统,其运行过程呈现高复杂性、强动态性和不确定性因素多等特征,为实现复杂产品离散装配车间的动态调度,提出一种基于数字孪生的复杂产品离散装配车间动态调度方法,通过物理空间和虚拟空间之间的数据交互进行更精确的动态调度。构建了基于数字孪生的复杂产品离散装配车间调度框架和调度问题模型,详细阐述了基于实时数据和神经网络的工时在线预测、面向复杂产品离散装配的数字孪生车间动态调度等关键实现技术;构建了复杂产品离散装配车间调度系统,并以某航天产品装配车间的调度过程为例,验证了该方法的有效性和可行性。

关键词: 数字孪生, 动态调度, 离散装配车间, 复杂产品, 工时预测