公开(公告)号：US20230168155A1

公开(公告)日：2023-06-01

申请号：US17562852

申请日：2021-12-27

Applicant: Industrial Technology Research Institute

Inventor： Yi-Jin LIN ,  Shuo-Peng LIANG ,  Chien-Chih LIAO ,  Tzuo-Liang LUO ,  Wan-Kun CHANG ,  Jen-Ji WANG

IPC: G01M99/00

CPC classification number: G01M99/005

Abstract: A process diagnosis system includes a digital twin calculation unit, a process diagnosis calculation unit, and a remote calculation analysis unit. The digital twin calculation unit obtains a vibration-related parameter and a cutting-related parameter of a processing device, and performs a simulation calculation for the vibration-related parameter, the cutting-related parameter and a three-dimensional model corresponding to the processing device to generate a three-dimensional calculation result. The process diagnosis calculation unit receives a three-dimensional calculation result and displays the three-dimensional calculation result. The remote calculation analysis unit receives the three-dimensional calculation result, and performs a simulation analysis for the three-dimensional calculation result to generate an analysis result.

公开(公告)号：US20170100810A1

公开(公告)日：2017-04-13

申请号：US14971438

申请日：2015-12-16

Applicant: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE

Inventor： Tsung-Ling HWANG ,  Chin-Te LIN ,  Shuo-Peng LIANG ,  Ta-Jen PENG ,  Jen-Ji WANG ,  Tzuo-Liang LUO

IPC: B23Q17/09 ,  G05B19/404

CPC classification number: B23Q17/0976 ,  G05B19/404 ,  G05B2219/31407 ,  G05B2219/41115 ,  G05B2219/41256 ,  G05B2219/49075

Abstract: A chatter avoidance method and device is provided, including steps of: providing a stable operating condition plot; partially removing a first layer of a workpiece with a predetermined first removal depth according to a safe removal depth of the stable operating condition plot and sensing a chatter caused by the removal operation; if no chatter is sensed, completing the removal operation, otherwise, continuing to partially remove the first layer with a second removal depth less than the predetermined first removal depth; and determining a minimum removal depth according to the removal operation, and removing a last layer of the workpiece with a last removal depth less than or equal to the minimum removal depth, allowing the workpiece to have a target thickness. The disclosure prevents a chatter from continuously occurring without requiring a shut-down and thereby maintains a desired production rate.

公开(公告)号：US20210150107A1

公开(公告)日：2021-05-20

申请号：US16721378

申请日：2019-12-19

Applicant: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE

Inventor： Jui-Ming CHANG ,  Shuo-Peng LIANG

IPC: G06F30/27

Abstract: A modeling method of a cutting force model is provided, including: using a cutting tool to cut a unidirectional fiber reinforced polymer along a circular path; using a measurement member to measure the cutting force on the cutting tool corresponding to the angle between the feeding direction of the cutting tool and the fiber direction of the unidirectional fiber reinforced polymer; and obtaining the functions of the cutting force coefficients in a formula according to the measurement result of the measurement member. With the modeling method, a mechanistic force model can be rapidly established to predict approximate cutting forces of the cutting tool in use.

公开(公告)号：US20200184720A1

公开(公告)日：2020-06-11

申请号：US16225931

申请日：2018-12-19

Applicant: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE

Inventor： Yang-Lun LIU ,  Yu-Lin TSAI ,  Yao-Yang TSAI ,  Cheng-Chieh WU ,  Shuo-Peng LIANG

IPC: G06T17/10 ,  G06K9/62 ,  G06F17/50 ,  G06N3/08

Abstract: A machining parameter automatic generation system includes a geometric data capturing module, a feature recognition learning network and a machining parameter learning network. The geometric data capturing module captures a geometric shape of a workpiece to generate a candidate feature list. The feature recognition learning network trains the candidate feature list according to a first neural network model to obtain an applicable feature list. The machining parameter learning network trains the applicable feature list and the candidate machining parameter according to a second neural network model to obtain an applicable machining parameter. The applicable machining parameter is used to generate a machining program, and the machining program is read by a machine tool for processing.