Optimization Strategy of Rolling Mill Hydraulic Roll Gap Control System Based on Improved Particle Swarm PID Algorithm

• Published in: Biomimetics (Basel, Switzerland) Vol. 8; no. 2; p. 143
• Main Authors: Yu, Ying, Zeng, Ruifeng, Xue, Yuezhao, Zhao, Xiaoguo
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 31.03.2023; MDPI
• Subjects: Accuracy; Aerospace materials; Algorithms; Controllers; Hydraulics; improved particle swarm algorithm; medium and thick plates; Neural networks; Optimization algorithms; PLC; Residual stress; roll gap control; semi-physical simulation; Simulation; Steel pipes; China; roll gap control; improved particle swarm algorithm; semi-physical simulation; PLC; medium and thick plates
• ISSN: 2313-7673; 2313-7673
• DOI: 10.3390/biomimetics8020143

Medium and heavy plates are important strategic materials, which are widely used in many fields, such as large ships, weapons and armor, large bridges, and super high-rise buildings. However, the traditional control technology cannot meet the high-precision control requirements of the roll gap of the thick plate mill, resulting in errors in the thickness of the medium and heavy plate, thereby reducing the quality of the product. In response to this problem, this paper takes the 5500 mm thick plate production line as the research background, and establishes the model of the rolling mill plate thickness automatic control system, using the Ziegler-Nichol response curve method (Z-N), particle swarm optimization (PSO) algorithm and linear weight particle swarm optimization (LWPSO) algorithm, respectively, optimizes the parameter setting of the PID controller of the system, and uses OPC UA communication technology to realize the online semi-physical simulation of Siemens S7-1500 series PLC (Siemens, Munich, Germany) and MATLAB R2018b (The MathWorks, Natick, Massachusetts, United States). Comparative studies show that when the same roll gap displacement step signal is given, the overshoot of the system response using the LWPSO algorithm is reduced by 14.26% and 10.18% compared with the Z-N algorithm and the PSO algorithm, and the peak time is advanced by 0.31 s and 0.05 s. The stabilization time is reduced by 3.71 s and 4.31 s, which effectively improves the control accuracy and speed of the system and has stronger anti-interference ability. It has certain engineering reference and application value.