In this paper, we discuss an identification and realization method of discrete event systems modeled by the timed Petri nets using neural networks. First, we propose a strict neural network representation of the timed Petri nets. By using the proposed representation, the realization problem of the Petri nets can be formulated as a combinatorial optimization problem. We propose a method for solving the problem using the genetic algorithm. Numerical experiments are performed to demonstrate the performance of the proposed method.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Identification and Realization of a Class of Discrete Event Systems by Neural Networks–Timed Petri Nets

  • Yasuaki Kuroe

摘要

In this paper, we discuss an identification and realization method of discrete event systems modeled by the timed Petri nets using neural networks. First, we propose a strict neural network representation of the timed Petri nets. By using the proposed representation, the realization problem of the Petri nets can be formulated as a combinatorial optimization problem. We propose a method for solving the problem using the genetic algorithm. Numerical experiments are performed to demonstrate the performance of the proposed method.