Shape control system
By setting a threshold distance in the data space, collecting new data samples at appropriate distances for relearning, and constructing a differential model, the problem of uncontrolled influence on relearning in the rolling system is solved, and accurate and stable updates of the learning model are achieved.
CN120662646APending Publication Date: 2025-09-19HITACHI LTD
1 Cites 0 Cited by
Patent Information
- Application Number
- CN202411207066.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-03-18
- Filing Date
- 2024-08-30
- Publication Date
- 2025-09-19
Smart Images
Figure CN120662646A_ABST
Abstract
The invention provides a shape control system. In a shape control system that controls the shape of an object using machine learning, unexpected effects on a learning model by re-learning of a learning device are suppressed, and a portion of the learning model that is intended to be updated by re-learning is appropriately updated. According to the invention, the shape control system corrects the controller model; when data sampled in a region separated from correction data by a threshold distance or more within a data space to which the correction data belongs is input to the controller model, the difference between an output from the controller model before correction and an output from the controller model after correction is made to face 0.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Patent Citations
Plant control device and method of controlling the same, rolling machine control apparatus and control method and program therefor
JP2019028824A