基于着色Petri网模糊专家系统的研究

针对变电站无功控制模糊专家系统知识表示不确定性及规则数量多的特点,文章以模糊、着色Petri网为基础,提出了一种基于模糊着色Petri网的知识表示与规则获取方法。该方法利用Petri网的图形化环境特点,将模糊规则库的不同变量用不同的颜色加以区分,不同规则中的同一个变量用该变量的颜色集表示,构成一个模糊着色Petri网模型。充分利用着色Petri网的特点,对推理过程进行了仔细研究,并提出一种基于着色模糊Petri网的启发式搜索策略。将其用于变电站无功控制的模糊专家系统中,结果表明,基于着色Petri网的模糊知识表示和获取方法,对于大型、复杂变电站模糊专家控制系统是非常有效的。

Study on Modeling Methods for Fuzzy Expert Systems Based on Colored Petri Net Model

With the purpose of tackling problems such as uncertainty and mass rules of Knowledge in substation fuzzy expert systems,a effective method for knowledge denotation and rule acquisition is proposed by using colored FPN models.In this method,diverse variable can be distinguished with heterogeneous color and homogeneous variable in different rules is expressed with its colored sets by making use of its characteristics of graph.Its reasoning process is deeply considered and heuristic search method is proposed,and applied in fuzzy expert systems of substation reactive control for uncertainty knowledge acquisition as well as denotation.The results show this approach is very effective and available for large and intricate substation fuzzy expert systems.