会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 2. 发明公开
    • Fuzzy rule acquisition method and apparatus for fuzzy inference system and fuzzy inference system using the apparatus
    • 获得模糊规则和装置模糊推理和模糊推理系统,用于该设备的方法。
    • EP0594168A2
    • 1994-04-27
    • EP93117073.2
    • 1993-10-21
    • HITACHI, LTD.
    • Ichimori, ToshihideMaeda, AkiraFunabashi, MotohisaNakano, Toshihiko
    • G05B13/02G06F9/44
    • G06N7/023G05B13/0285
    • A fuzzy rule acquisition method for a fuzzy inference system in which a fuzzy inference is achieved based on fuzzy knowledge including fuzzy rules each described in a format having an IF part and a THEN part, and membership functions defining meanings of respective propositions described in the IF part and THEN part, and in which a fuzzy inference is performed for at least one input value by using the fuzzy rules of the fuzzy knowledge to obtain a result of the execution of the fuzzy inference (Step 101), the result of the execution of the fuzzy inference is compared with a teaching value to obtain an inference error (Step 102), the errors of the fuzzy rules are obtained by using the inference error (Step 103), it is judged whether each of the fuzzy rules is contradictory, based on the errors of corresponding rules of the fuzzy rules (Step 105), and a fuzzy rule judged as contradictory is modified to dissolve the contradiction (Step 106).
    • 为模糊推理系统,其中,模糊推理是基于模糊知识,包括模糊规则的格式描述的每个具有在IF部分和THEN部分,以及隶属函数定义respectivement命题的含义在IF描述实现了模糊规则获取方法 部分和THEN部分,并且其中一个模糊推理是通过使用的模糊知识的模糊规则来获得模糊推理的执行结果(步骤101),执行的结果,执行对至少一个输入值 模糊推理与教学值相比,以获得在推理误差(步骤102),的模糊规则的误差通过使用推理误差(步骤103)获得,判断是否每个模糊规则是矛盾的,基于 上的相应的模糊规则(步骤105),并判断为矛盾模糊规则的规则的错误被修改以溶解矛盾(步骤106)。
    • 4. 发明公开
    • A method and system for adjusting fuzzy inference parameters
    • 一种调整模糊推理参数的方法和系统
    • EP0460642A2
    • 1991-12-11
    • EP91109196.5
    • 1991-06-05
    • HITACHI, LTD.
    • Maeda, AkiraSomeya, RyukoFunabashi, MotohisaMizuno, HirotakaNakano, Toshihiko
    • G06F15/18G06F7/60G05B13/02
    • G06N7/023G06N3/0436
    • A method and system is disclosed for fuzzy inference parameter adjustment in a fuzzy inference system for obtaining an output value by a node from at least one input value in a calculation network having a plurality of nodes for calculating the output value from at least one input value in predetermined procedures and at least link for connection between the nodes, and for obtaining a fuzzy inference output value with use of the calculation network by inputting an output value of a node to another node through the link for the node and by successively repeating similar input-output conversion as necessary.
      The system can trace back a cause of error of the inference output value from a teacher data on the calculation network to adjust each fuzzy inference parameter, thereby providing a highly accurate and highly efficient inference.
      The system learns input-output relationships between pluralities of sets of inputs and outputs expected in advance, and obtains a value by differentiating or finite differencing a difference of an output value to an input from an expected output value with respect to the parameter for each of the sets of the inputs and the outputs, thereby being capable of indicating high order statistical value of dispersion and the like for each pair of input and output and indicating distribution of differentiated or infinite difference calculated values.
    • 公开了一种用于模糊推理系统中的模糊推理参数调节的方法和系统,用于从具有多个节点的计算网络中的至少一个输入值获得节点的输出值,所述多个节点用于从至少一个输入值计算输出值 在预定的过程中,并且至少用于连接节点之间的连接,并且通过使用计算网络通过将节点的输出值通过节点的链接输入到另一个节点并且通过连续地重复相似的输入来获得模糊推理输出值 必要时输出转换。 系统可以从计算网络上的教师数据中追溯推理输出值的误差原因,以调整每个模糊推理参数,从而提供高度准确和高效的推断。 该系统学习预先预期的多组输入和输出之间的输入 - 输出关系,并且通过对输出值相对于期望输出值的输入的差异进行微分或有限差分,获得相对于每个输入和输出的参数的值 输入和输出的集合,从而能够为每对输入和输出指示色散等的高阶统计值,并且指示差分或无限差分计算值的分布。
    • 9. 发明公开
    • A method and system for adjusting fuzzy inference parameters
    • 一种用于调整FUZZY INFERENCE参数的方法和系统
    • EP0460642A3
    • 1993-05-19
    • EP91109196.5
    • 1991-06-05
    • HITACHI, LTD.
    • Maeda, AkiraSomeya, RyukoFunabashi, MotohisaMizuno, HirotakaNakano, Toshihiko
    • G06F15/18G06F7/60G05B13/02
    • G06N7/023G06N3/0436
    • A method and system is disclosed for fuzzy inference parameter adjustment in a fuzzy inference system for obtaining an output value by a node from at least one input value in a calculation network having a plurality of nodes for calculating the output value from at least one input value in predetermined procedures and at least link for connection between the nodes, and for obtaining a fuzzy inference output value with use of the calculation network by inputting an output value of a node to another node through the link for the node and by successively repeating similar input-output conversion as necessary. The system can trace back a cause of error of the inference output value from a teacher data on the calculation network to adjust each fuzzy inference parameter, thereby providing a highly accurate and highly efficient inference. The system learns input-output relationships between pluralities of sets of inputs and outputs expected in advance, and obtains a value by differentiating or finite differencing a difference of an output value to an input from an expected output value with respect to the parameter for each of the sets of the inputs and the outputs, thereby being capable of indicating high order statistical value of dispersion and the like for each pair of input and output and indicating distribution of differentiated or infinite difference calculated values.