一种基于模糊Petri网的并行推理方法

把模糊Petri网模型转化为矩阵形式,在此基础上提出了一种并行推理算法。算法将推理过程转化为矩阵运算,不但考虑了前提条件的权值、变迁阈值和规则可信度等因素,还进一步将变迁触发条件严格化,有效的避免了一些变迁不必要的重复触发,降低了算法复杂度。通过实例说明,此推理算法易于实现并可以提高推理效率,尤其适合较大较复杂的模糊Petri网模型。

A Parallel Reasoning Algorithm Based on Fuzzy Petri Net

This paper maps the fuzzy Petri net into a matrix, and presents a parallel reasoning algorithm. The algorithm transforms the reasoning process to matrix operation. Many constraints of the rules, such as weights, thresholds of transitions, degree of true of rules, are considered in the algorithm. Beside this, according to the algorithm, the fETing condition of transitions becomes more strictly. It avoids unnecessary repetitions of fETing of transitions effectively, and reduces the complexity of the algorithm. At last, an example is provided to demonstrate that the reasoning algorithm is simple and can improve the efficiency of the reasoning process. It fits reasoning for the large-scale FPN model especially.