温室温度模糊控制参数在线自整定算法

Study on Greenhouse Environmental Temperature Fuzzy Controller with On-line Self-tuning Parameters

温室环境温度变化规律由于受诸多因素影响而表现出非线性和时变性。为了保证温室温度模糊控制器的性能,需要根据被控对象的变化不断调整其参数。温室温度响应的主要动态性能指标包括:超调量、系统发散程度、系统振荡程度、稳态误差、过渡过程时间等。通过分析模糊控制器比例因子对以上指标的影响,建立了以系统性能指标为输入的比例因子模糊整定算法。由于该算法程序简单、所需系统资源很少,因此可用于比例因子的在线整定。在温室冬季供暖情况下的试验结果显示,系统响应性能全面改善,ITAE指标下降了15%。

The principle of Greenhouse Environment Temperature, under the influence of many factors, has the characters of nonlinear and non-stationary. This makes it hard to tune the parameters of the fuzzy control system. The paper analyzed the relationship of the scale factors and several main performance indexes of the system＇s respond curve like over-regulation, divergence, surge-degree, steady-state error and transient time, and denoted these indexes by some fuzzy membership functions. The self-tuning algorithm that took the indexes as inputs was presented. It made use of fuzzy inference to decide the increment of each factor. The results of experiments carried out in last winter indicate that it improves the performance of the fuzzy controller entirely, the ITAE decreases over 15 percents, and it is feasible to use it on line since the algorithm is simple and need few system resources.