摘要
从工程实际应用的角度出发,针对金属材料真空退火过程中系统的非线性、工艺参数的不确定性和对外部干扰的不灵敏性等特点,结合非线性不确定系统理论研究,采用神经网络建立系统的模型,利用自适应免疫遗传算法(AIGA)对滑模面和控制器的参数及神经网络的权值、阈值进行优化,得出了一种真空退火炉工件温度精确控制的智能变结构控制策略。实验表明,这种方法控制温差在±4℃,优于基于单一模型的模型参考自适应控制或PID策略的控制方式,具有较好的动态特性和耐久性。
In metal vacuum annealing process, the accurate annealing temperature control is a nonlinear, uncertain, insensitive to external interference system. The control method of this paper combines nonlinear uncertain system theory researching and the neural network is used to model the predictive part of the system. The adaptive immune genetic algorithm (AIGA) is utilized to optimize parameters of sliding mode surface and controller. Finally, an intelligent variable structure control strategy of accurately controlling work-pieces temperature in vacuum annealing furnace is given. The experimental results show that this control method is superior to single model reference adaptive control or to the PID strategy control, and it has better dynamic characteristic and stronger robustness.
出处
《材料热处理学报》
EI
CAS
CSCD
北大核心
2006年第1期124-129,共6页
Transactions of Materials and Heat Treatment
基金
国家科技攻关计划基金资助项目(2002BA901A28)
甘肃省省长基金资助项目(GS015A52012)
关键词
真空退火炉
智能变结构
自适应免疫遗传算法(AIGA)
控制
vacuum annealing furnace
intelligent variable structure
adaptive immune genetic algorithm(AIGA)
control
