知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

12 条记录 (耗时 0.011 秒)

    ACTION INFORMATION LEARNING DEVICE, ACTION INFORMATION OPTIMIZATION SYSTEM AND COMPUTER READABLE MEDIUM
    4.
    发明申请
    ACTION INFORMATION LEARNING DEVICE, ACTION INFORMATION OPTIMIZATION SYSTEM AND COMPUTER READABLE MEDIUM 审中-公开

    公开(公告)号：US20180210431A1

    公开(公告)日：2018-07-26

    申请号：US15857911

    申请日：2017-12-29

    申请人： FANUC CORPORATION

    发明人： Zheng TONG ,  Takuma NISHIMURA ,  Yuuzou INAGUCHI

    IPC分类号： G05B19/418 ,  G05B13/02 ,  G05B19/404

    CPC分类号： G05B19/41835 ,  G05B13/027 ,  G05B19/404 ,  G05B19/41855 ,  G05B19/41875 ,  G05B2219/33038

    摘要： To perform reinforcement learning that enables selecting action information for shortening a cycle time while also avoiding the occurrence of overheating. An action information learning device (300) includes: a state information acquisition means (310) for acquiring state information including an operation pattern of a spindle and a combination of parameters related to machining of a machine tool (100); an action information output means (320) for outputting action information including adjustment information for the operation pattern and the combination of parameters included in the state information; a reward calculation means (333) for acquiring judgment information which is information for temperature of the machine tool (100) and a machining time related to the machining of the machine tool (100), and calculating a value of a reward for reinforcement learning based on the judgment information thus acquired; and a value function update means (332) for updating a value function by performing the reinforcement learning based on the value of the reward, the state information and the action information.

    OPERATION TRAINING SYSTEM AND PLANT OPERATION SUPPORTING SYSTEM
    8.
    发明申请
    OPERATION TRAINING SYSTEM AND PLANT OPERATION SUPPORTING SYSTEM 审中-公开
    操作培训系统和工厂操作支持系统

    公开(公告)号：US20110183303A1

    公开(公告)日：2011-07-28

    申请号：US13122165

    申请日：2009-04-21

    申请人： Takaya Yamamoto ,  Kazuyuki Mori

    发明人： Takaya Yamamoto ,  Kazuyuki Mori

    IPC分类号： G09B19/00

    CPC分类号： G09B25/02 ,  G05B17/02 ,  G05B2219/33038

    摘要： The present invention includes an operation training system, including: an operation training simulator obtained by modeling a real plant, for simulating a plant state of any one of during a normal operation and in case of an accident; training operation input means for inputting an operation for the operation training simulator; simulated accident information input means for inputting information of a simulated accident for allowing the operation training simulator to simulate the accident; a training operation procedure database in which operation procedures to be input for the normal operation and for the simulated accident during operation training are registered; training mis-operation detecting means for comparing the operation procedures registered in the training operation procedure database and the operation input to the training operation input means with each other to detect whether the input operation is a mis-operation; and a recovery operation scenario database for registering the operation procedure input to the training operation input means as an operation procedure for recovery of the plant state after the mis-operation is detected by the training mis-operation detecting means.

    摘要翻译： 本发明包括一种操作训练系统，包括：通过对真实设备进行建模而获得的操作训练模拟器，用于模拟正常操作期间和事故情况下的任何一个的植物状态; 训练操作输入装置，用于输入操作训练模拟器的操作; 模拟事故信息输入装置，用于输入模拟事故的信息，以允许操作训练模拟器模拟事故; 训练操作程序数据库，其中记录在操作训练期间输入正常操作和模拟事故的操作程序; 训练误操作检测装置，用于将训练操作过程数据库中登记的操作过程和对训练操作输入装置的操作输入进行比较，以检测输入操作是否是错误操作; 以及恢复操作场景数据库，用于将输入到训练操作输入装置的操作过程登记为用于通过训练误操作检测装置检测到误操作之后恢复工厂状态的操作过程。

    CONTROL DEVICE AND CONTROL METHOD
    10.
    发明申请
    CONTROL DEVICE AND CONTROL METHOD 审中-公开

    公开(公告)号：US20180218262A1

    公开(公告)日：2018-08-02

    申请号：US15877288

    申请日：2018-01-22

    申请人： Panasonic Intellectual Property Corporation of America

    发明人： MASASHI OKADA

    IPC分类号： G06N3/08 ,  G05B13/02

    CPC分类号： G06N3/084 ,  G05B13/0285 ,  G05B2219/33038 ,  G05B2219/34066 ,  G06N3/0445 ,  G06N3/0454 ,  G06N7/005

    摘要： A control device for performing optimal control by path integral includes a neural network section including a machine-learned dynamics model and cost function, an input section that inputs a current state of a control target and an initial control sequence for the control target into the neural network section, and an output section that outputs a control sequence for controlling the control target, the control sequence being calculated by the neural network section by path integral from the current state and the initial control sequence by using the dynamics model and the cost function. Here, the neural network section includes a second recurrent neural network incorporating a first recurrent neural network including the dynamics model.