木材干燥过程温湿度的T-S型模糊神经网络控制器设计  

Design of T-S fuzzy neural network controller for temperature and humidity in wood drying process

作　　者：姜滨[1,2] 孙丽萍[1] 曹军[1] 

机构地区：[1]东北林业大学机电工程学院,黑龙江哈尔滨150040 [2]哈尔滨电工仪表研究所,黑龙江哈尔滨150028

出　　处：《电机与控制学报》2016年第10期114-120,共7页Electric Machines and Control

基　　金：国家林业公益性行业科研专项(201304502)

摘　　要：木材干燥过程是一个强耦合、大滞后的非线性动力系统,很难准确建立被控对象的数学模型。为了准确控制木材干燥过程的温度和湿度,提高木材干燥质量,将智能控制引入木材干燥控制系统是必然的发展趋势。结合模糊控制和神经网络优点,设计了一种木材干燥窑内温湿度的Takagi-Sugeno(T-S)型模糊神经网络控制器。该控制器无需对象的精确数学模型,适应性强,利用模糊算法解除木材干燥窑内温度和湿度间的强耦合关系,采用神经网络的自学习和自适应能力来实现整个非线性过程的模糊逻辑推理。仿真和实验结果表明,T-S型模糊神经网络控制器有效解决了木材干燥过程的温湿度控制,控制器响应速度快、超调小、鲁棒性强、控制精确度高,可以满足木材干燥控制系统要求。Wood drying process presents normally the non-linear characteristics of strong coupling and large lagging, therefore, it is hardly to build the math model of controlled object. In order to control more precisely the temperature and humidity of the wood drying process so as to improve the drying quality, it is necessary to apply the intelligent controller in wood drying control system. Combining the merits of fuzzy control and neural control, a Takagi-Sugeno （T-S） fuzzy neural network controller is designed to control the inner temperature and humidity of wood drying kiln. This controller has strong adaptability and did not depend on the precise math model. With fuzzy algorithm, the coupling relationship was removed between inner temperature and humidity of wood drying kiln. The self-learning and adaptive ability of neural net- work was used to accomplish the fuzzy logic of the whole non-linear process. The simulation reveales that T-S fuzzy neural network controller solves the problem of low control precision of temperature and humidi- ty in wood drying process. And this controller has fast response speed, low overshoot, strong robustness and high control precision. It might fulfill the demand of wood drying control system.

关 键 词：木材干燥过程 T-S模型 模糊神经网络控制器 温湿度控制 神经网络 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置]