注：国家重点研发计划项目（No. 2020YFB1708200）
作者：刘炜，恒庆海，李庆奎
单位：北京信息科技大学自动化学院，北京 100101
中图分类号：TP273          
文献标识码：A         
文章编号：1006-883X(2022)01-0019-07
收稿日期：2021-11-29

摘要：针对动态供应链系统正常运行过程中生产环节发生变更的情况，提出了一种基于数据驱动的自适应预测控制算法。首先，利用数据驱动的方法建立供应链系统的子空间预估模型，将子空间预估器参数与预测控制策略相结合，直接设计自适应预测控制器；其次，求得多级供应链系统在运作过程中发生内部节点变更状况下的库存控制策略；最后，以三节点生产—库存供应链系统为例，验证所提出的算法。在仿真部分验证了基于数据建模的准确性，分析了供应链系统在自适应预测控制的生产策略下，其在内部节点发生变更前后库存水平的波动状况。仿真结果表明该方法的鲁棒性和有效性。
关键词：供应链；系统变更；数据驱动；预测控制

Data Driven Adaptive Predictive Control of Supply Chain System Subject to Design Change
LIU Wei, HENG Qinghai, LI Qingkui
(College of Automation, Beijing Information Science and Technology University, Beijing 100101, China)
Abstract: Aiming at the change of production links in the operation of dynamic supply chain system, a data-driven adaptive predictive control algorithm is proposed. Firstly, the subspace prediction model of supply chain system is established by using the data-driven method, and the adaptive predictive controller is directly designed by combining the subspace predictor parameters with the predictive control strategy. Then the inventory control strategy under the condition of internal node change in the operation of multi-level supply chain system is obtained. Finally, a three nodes production inventory supply chain system is taken as an example to verify the proposed algorithm. In the simulation part, the accuracy of data modeling is verified, and the fluctuation of inventory level before and after the change of internal nodes in the supply chain system under the production strategy of adaptive predictive control is analyzed. The simulation results show the robustness and effectiveness of the method.
Key words: supply chain; system change; data-driven; predictive control

阅读全文

备注：2022年 第28卷 第01期