具有部分不可测前提变量的模糊系统的控制器设计

现有基于模糊系统模型的控制器设计方法大多假定模型的前提变量都是在线可测的。实际问题并非如此，如双槽系统，双轮驱动车系统等它们的模糊模型具有部分不可测前提变量。对于这类模糊系统，因为模糊规则权重同时依赖于可测和不可测前提变量，所以近年来的研究将规则权重完全视为未知，往往不能充分利用可测前提变量的信息获得较好的控制效果。本项目试图通过使用不可测前提变量的估计和可测前提变量共同作为控制器前提变量，来构造一类新的模糊控制方案，并基于该控制方案探求控制器设计条件。与现有的研究相比，本项目通过在控制器中直接使用可测前提变量，能更有效的利用系统可获取的信息来设计控制器。这一项目的难点和关键问题是如何探求新的技术充分利用可测与不可测前提变量及其估计值的隶属度函数给出更有效的设计条件。该项目的成功实施，将对具有部分不可测前提变量模糊系统的镇定控制和性能的改善产生积极影响，进一步推动模糊控制理论及应用的发展。

The premise variables of the fuzzy model are often assumed to be acquired on line in the existing literature. However, the real problems are not so, for example, tank systems, two-wheeled vehicle systems and so on, the fuzzy model of which are with measurable and unmeasurable premises. For the class of fuzzy systems, the rule weight of each fuzzy rule is dependent on both unmearable and mearable premises. In recent years, the rule weights are considered as completely unknown information for the class of fuzzy systems, so that effective control cannot be obtained.In this project,the measurable premise vairables and the estimate values of the unmeasurable premise variables are all chosen as the premise vaiables of the fuzzy controllers,based on the fuzzy control scheme, new condition for designing fuzzy controllers will be explored. In contrast to the existing approaches, the measurable premise variables are directly used in controllers, which takes full advantage of the available information on line for achieving better control effects. The key problems and difficulties of this project are how to apply the mearable premise variables,unmeasurable premise variables and their estimated values to give more effective condition for designing fuzzy controllers. The successful implementation of the project will have a positive impact on stabilization and improvment of system performances, and it will further promotes the development of the theory of fuzzy control.