基于ADHDP方法的生产线喷涂烘干过程温度控制

ADHDP for the Temperature Control in the Spraying and Drying Process of Production Lines

针对自动化生产线喷涂烘干过程中电加热炉温度具有纯滞后、大惯性等非线性控制特点,提出一种基于控制依赖启发式动态规划的自适应控制方法,实现对电加热炉温度的优化控制。它不需要精确的数学模型,也不需要精确定义最优控制理论中的系统性能指标函数,可以在线学习控制。仿真试验表明,这种方法相比传统的PID控制在稳态误差、超调量、响应速度,抗干扰能力和鲁棒性等方面具有更好的控制效果。

in the spraying and drying process of automatic production lines, the electric heating furnace tem- perature has the typical characteristics of time-delay, large inertia, and non-linearity. It is difficult to control the temperature very well in industrial fields. Action-Dependent Heuristic Dynamic Programming （ADHDP） algorithm, a class of adaptive control method, is used to achieve optimal control of the electric furnace temperature. It does not need accurate mathematical model and it also does not need define a precise performance indicators function （cost- to-go） of optimal control theory. ADHDP can be self-learning online in order to get more system performance. The simulation results indicate that ADHDP has a better control effect compared with the traditional PID algorithm in steady-state error, overshoot, response speed, anti-jamming capability and robustness, etc.