摘要
目的提出一种能够提高退火炉温度控制系统的性能和精度的具体方案,增强控制系统的鲁棒性.方法针对退火炉温度控制系统具有多变量,非线性和不确定性的特点,将T-S模糊神经网络与预测控制相结合,在线建立被控对象的数学模型,并用BP神经网络控制器对所得到的信息在线修正,进而控制退火炉炉温.并通过仿真与传统的模糊PID控制方案进行对比分析.结果 T-S模糊神经网络预测控制方案具有较强的控制精度和动态性能,预测精度高、容错性好、收敛速度快,基本无超调等特点.结论 T-S模糊神经网络预测控制能够提高产品退火质量、节能环保,可以应用于退火炉炉温的优化控制.
This paper aims to propose a new control scheme,in order to improve the temperature control sys- tem of annealing furnace control performance and control precision, increase the robustness of the system. Method for annealing furnace temperature control system is multivariable, nonlinear and uncertain character- istics, T-S fuzzy neural network predictive control combined with online, the mathematical model of the ob- ject, and using BP neural network controller to the information available on-line correction, and the annealing furnace temperature control. The results of simulation, the traditional fuzzy PID control and the control plan are compared and it can be clearly seen that this control scheme has the advantages in fast convergence speed, and has no overshoot and prediction precision. Conclusions achieve the basic with a large delay, strong coupling of the temperature control system of annealing furnace for precise control, while the other has the same characteristic industry object also has certain reference function.
出处
《沈阳建筑大学学报(自然科学版)》
CAS
北大核心
2014年第1期181-186,共6页
Journal of Shenyang Jianzhu University:Natural Science
基金
国家自然科学基金项目(61272253)
住房和城乡建设部基金项目(2011-k11-22)