This paper compared the difference between the traditional Petri nets and reasoning Petri nets(RPN),and presented a fuzzy reasoning Petri net(FRPN) model to represent the fuzzy production rules of a rule based system....This paper compared the difference between the traditional Petri nets and reasoning Petri nets(RPN),and presented a fuzzy reasoning Petri net(FRPN) model to represent the fuzzy production rules of a rule based system.Based on the FRPN model,a formal reasoning algorithm using the operators in max algebra was proposed to perform fuzzy reasoning automatically.The algorithm is consistent with the matrix equation expression method in the traditional Petri net.Its legitimacy and feasibility were testified through an example.展开更多
Intuitionistic fuzzy Petri net is an important class of Petri nets,which can be used to model the knowledge base system based on intuitionistic fuzzy production rules.In order to solve the problem of poor self-learnin...Intuitionistic fuzzy Petri net is an important class of Petri nets,which can be used to model the knowledge base system based on intuitionistic fuzzy production rules.In order to solve the problem of poor self-learning ability of intuitionistic fuzzy systems,a new Petri net modeling method is proposed by introducing BP(Error Back Propagation)algorithm in neural networks.By judging whether the transition is ignited by continuous function,the intuitionistic fuzziness of classical BP algorithm is extended to the parameter learning and training,which makes Petri network have stronger generalization ability and adaptive function,and the reasoning result is more accurate and credible,which is useful for information services.Finally,a typical example is given to verify the effectiveness and superiority of the parameter optimization method.展开更多
Though the dominance-based rough set approach has been applied to interval-valued information systems for knowledge discovery, the traditional dominance relation cannot be used to describe the degree of dominance prin...Though the dominance-based rough set approach has been applied to interval-valued information systems for knowledge discovery, the traditional dominance relation cannot be used to describe the degree of dominance principle in terms of pairs of objects. In this paper, a ranking method of interval-valued data is used to describe the degree of dominance in the interval-valued information system. Therefore, the fuzzy rough technique is employed to construct the rough approximations of upward and downward unions of decision classes, from which one can induce at least and at most decision rules with certainty factors from the interval-valued decision system. Some numerical examples are employed to substantiate the conceptual arguments.展开更多
模糊Petri网(Fuzzy Petri Net,FPN)是Petri网(Petri Net,PN)的模糊化描述的一种扩充.基于FPN模型的模糊推理规则表示和模糊推理已经得到了广泛的研究.传统的方法不能准确表示含有否定命题的产生式规则,并解决正确推理问题.本文讨论了模...模糊Petri网(Fuzzy Petri Net,FPN)是Petri网(Petri Net,PN)的模糊化描述的一种扩充.基于FPN模型的模糊推理规则表示和模糊推理已经得到了广泛的研究.传统的方法不能准确表示含有否定命题的产生式规则,并解决正确推理问题.本文讨论了模糊逻辑中否定的含义,将条件命题中的否定理解为其对推理规则的阻碍作用,结果命题中的否定理解为规则中的前提条件阻碍该命题的发生.在此基础上提出一种新的适合于含有否定命题逻辑规则的一致性FPN模型(Consistent Fuzzy Petri Net,CFPN)表示方法,同时在CFPN模型中引入域值的概念,并给出相应的形式化推理算法及相关证明.展开更多
针对模糊Petri网存在隶属度单一的问题,将直觉模糊集理论与Petri网理论相结合,构建直觉模糊Petri网(Intuitionistic Fuzzy Petri Nets,IFPN)模型,用于知识的表示和推理.首先构建了IFPN模型,并将其应用于知识的表示,通过在模型中引入抑...针对模糊Petri网存在隶属度单一的问题,将直觉模糊集理论与Petri网理论相结合,构建直觉模糊Petri网(Intuitionistic Fuzzy Petri Nets,IFPN)模型,用于知识的表示和推理.首先构建了IFPN模型,并将其应用于知识的表示,通过在模型中引入抑止转移弧,解决了否命题的表示问题.其次提出了基于矩阵运算的IFPN推理算法,通过修改变迁触发后token值的传递规则,解决了推理过程中的事实的保留问题;通过修改变迁的触发规则,抑制了变迁的重复触发.最后对推理算法进行了分析,并举例验证了提出的IFPN模型及其推理算法的可行性,结果表明IFPN是对FPN的有效扩充和发展,其对推理结果的描述更加细腻、全面.展开更多
针对车电系统总线故障现象和故障原因之间存在复杂模糊关系而无法实现快速故障诊断这一难题,提出采用模糊Petri网(fuzzy petri net,FPN)建立车电系统总线的故障诊断分析模型,并给出了相应的形式化推理算法;以某型车电系统总线为例,分析...针对车电系统总线故障现象和故障原因之间存在复杂模糊关系而无法实现快速故障诊断这一难题,提出采用模糊Petri网(fuzzy petri net,FPN)建立车电系统总线的故障诊断分析模型,并给出了相应的形式化推理算法;以某型车电系统总线为例,分析比较了各种故障原因的可信度;分析表明:所建模型有效地描述了车电系统总线故障现象和故障原因之间的模糊关系,能够实现故障的快速诊断,该模型可以作为车电系统总线快速故障诊断的理论支持工具。展开更多
文摘This paper compared the difference between the traditional Petri nets and reasoning Petri nets(RPN),and presented a fuzzy reasoning Petri net(FRPN) model to represent the fuzzy production rules of a rule based system.Based on the FRPN model,a formal reasoning algorithm using the operators in max algebra was proposed to perform fuzzy reasoning automatically.The algorithm is consistent with the matrix equation expression method in the traditional Petri net.Its legitimacy and feasibility were testified through an example.
文摘Intuitionistic fuzzy Petri net is an important class of Petri nets,which can be used to model the knowledge base system based on intuitionistic fuzzy production rules.In order to solve the problem of poor self-learning ability of intuitionistic fuzzy systems,a new Petri net modeling method is proposed by introducing BP(Error Back Propagation)algorithm in neural networks.By judging whether the transition is ignited by continuous function,the intuitionistic fuzziness of classical BP algorithm is extended to the parameter learning and training,which makes Petri network have stronger generalization ability and adaptive function,and the reasoning result is more accurate and credible,which is useful for information services.Finally,a typical example is given to verify the effectiveness and superiority of the parameter optimization method.
基金Acknowledgements This work was supported by the National Natural Science Foundation of China （Grant No. 60632050） and Postdoctoral Science Foundation of China （20100481149）.
文摘Though the dominance-based rough set approach has been applied to interval-valued information systems for knowledge discovery, the traditional dominance relation cannot be used to describe the degree of dominance principle in terms of pairs of objects. In this paper, a ranking method of interval-valued data is used to describe the degree of dominance in the interval-valued information system. Therefore, the fuzzy rough technique is employed to construct the rough approximations of upward and downward unions of decision classes, from which one can induce at least and at most decision rules with certainty factors from the interval-valued decision system. Some numerical examples are employed to substantiate the conceptual arguments.
文摘模糊Petri网(Fuzzy Petri Net,FPN)是Petri网(Petri Net,PN)的模糊化描述的一种扩充.基于FPN模型的模糊推理规则表示和模糊推理已经得到了广泛的研究.传统的方法不能准确表示含有否定命题的产生式规则,并解决正确推理问题.本文讨论了模糊逻辑中否定的含义,将条件命题中的否定理解为其对推理规则的阻碍作用,结果命题中的否定理解为规则中的前提条件阻碍该命题的发生.在此基础上提出一种新的适合于含有否定命题逻辑规则的一致性FPN模型(Consistent Fuzzy Petri Net,CFPN)表示方法,同时在CFPN模型中引入域值的概念,并给出相应的形式化推理算法及相关证明.
文摘针对模糊Petri网存在隶属度单一的问题,将直觉模糊集理论与Petri网理论相结合,构建直觉模糊Petri网(Intuitionistic Fuzzy Petri Nets,IFPN)模型,用于知识的表示和推理.首先构建了IFPN模型,并将其应用于知识的表示,通过在模型中引入抑止转移弧,解决了否命题的表示问题.其次提出了基于矩阵运算的IFPN推理算法,通过修改变迁触发后token值的传递规则,解决了推理过程中的事实的保留问题;通过修改变迁的触发规则,抑制了变迁的重复触发.最后对推理算法进行了分析,并举例验证了提出的IFPN模型及其推理算法的可行性,结果表明IFPN是对FPN的有效扩充和发展,其对推理结果的描述更加细腻、全面.
文摘针对车电系统总线故障现象和故障原因之间存在复杂模糊关系而无法实现快速故障诊断这一难题,提出采用模糊Petri网(fuzzy petri net,FPN)建立车电系统总线的故障诊断分析模型,并给出了相应的形式化推理算法;以某型车电系统总线为例,分析比较了各种故障原因的可信度;分析表明:所建模型有效地描述了车电系统总线故障现象和故障原因之间的模糊关系,能够实现故障的快速诊断,该模型可以作为车电系统总线快速故障诊断的理论支持工具。
