端边云协同的PID整定智能系统

本文在分析智能制造对PID整定的新需求及PID整定面临的挑战难题的基础上, 将自动化的建模、控制与优化和人工智能的深度学习与强化学习深度融合与协同, 提出了自适应与自主的PID整定的智能优化方法, 包括端边云协同的PID控制过程数字孪生模型和强化学习与数字孪生模型相结合的PID整定算法. 将工业互联网的端边云协同技术与PLC控制系统相结合, 研制了PID整定智能系统, 并在重大耗能设备 — 电熔镁炉成功应用. 该系统安全、可靠与优化运行, 取得显著的节能减排效果. 最后, 提出了控制系统智能化研究方向需要进一步深入研究的内容.

PID Tuning Intelligent System Based on End-edge-cloud Collaboration

Based on the analysis of the new requirements of intelligent manufacturing for PID tuning and the challenges and difficulties faced by PID tuning, this paper proposes an adaptive and autonomous PID tuning intelligent optimization method by deeply integrates and coordinates the modeling, control and optimization in automation and deep learning and reinforcement learning in artificial intelligence. The proposed method contains the digital twin model of the PID control process based on end-edge-cloud collaboration and the PID tuning algorithm combining reinforcement learning and digital twin model. Furthermore, the PID tuning intelligent system is developed by combining the end-edge-cloud collaboration technology of Industrial Internet with the PLC control system, and has been successfully applied to the energy intensive equipment — Fused magnesium furnace. This system operates safely, reliably and optimally, achieving remarkable effects in energy conservation and emission reduction. Finally, the further research content in the intelligent research direction of control system is proposed.