基于数据表结构的FPN并行推理算法

通过对模糊产生式规则一般形式的研究，抽取出适用于FPN模糊推理的两类模糊产生式规则。定义并建立了FPN模型，将在图形结构上进行推理的传统Petri网模型映射到数据表结构上．在此基础上实现了对基于规则的正向推理和基于模糊Peari网的模糊推理算法。最后通过实例对该算法进行了检验．结果表明该算法不仅具有简单通用、容易实现和高效的特点，而且还适食于对比较复杂的模糊产生式知识库进行并行推理。

A Parallel Reasoning Algorithm for FPN Based on Table Structure

Two types of fuzzy production rules, which is suited to the fuzzy Petri net, is extracted by the research on the general rules first in this paper. A fuzzy Petri net model is defined to represent the fuzzy production rules of a rulebased system, and then mapping the fuzzy Petri net model to a data table. A fuzzy reasoning algorithm based on the table is implemented. Using a practical instance to test the algorithm at last, the result shows that the algorithm is simple, high-powered and universal. Specially, it has parallel reasoning ability and fits reasoning for the large-scale FPN model.