基于智能解耦补偿的绿茶烘焙改进模糊控制

A Improved Fuzzy Control Based on Intelligent Decoupling Compensator for Green Tea Baking

针对绿茶烘焙生产过程中,物质交换和热交换复杂,温、湿度变化耦合性强、设备性能易变的问题,提出了一种基于解耦补偿的改进模糊控制方法;首先采用模糊控制器对温、湿度独立控制,为了抑制参数漂移,利用遗传算法对模糊隶属度进行在线优化;同时引入神经元学习算法,实现对温度和湿度控制量解耦关系的学习,对控制量进行补偿,从而保证改进模糊算法计算的控制量相互独立;结果表明采用文章方法处理的3种绿茶样本,橙花叔醇成分平均提升15.7%,α-法呢烯成分平均提升21.4%,芳樟醇成分平均提升13.4%。

In the baking process of green tea, material exchange and heat exchange complex, temperature and humidity changes and strong coupling, equipment performance variable problem, we propose a compensation based on improved fuzzy decoupling control method. First, the fuzzy controller on the temperature and humidity control, in order to suppress parameter drift, genetic algorithm fuzzy membership for online optimization; while introducing neuron learning algorithm to achieve the decoupling of temperature and humidity control volume relationship of learning, the control amount of compensation. Thus ensuring improved fuzzy control algorithm to calculate the amount of mutually independent. The results showed that three kinds of Green Tea samples treated by this method, the average increase of 15.7% tertiary alcohol ingredient orange flower, alpha farnesene component increased an average of 21.4%, linalool increased an average of 13.4% components.