Damage prediction device and program

The damage prediction device analyzes spindle current and vibration data to predict rotary tool damage using regression equations, addressing inaccuracies in existing methods by accounting for varying operating states and operations.

JP2026095785APending Publication Date: 2026-06-12NACHI FUJIKOSHI CORP
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2024209227
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-12-02
Publication Date
2026-06-12

AI Technical Summary

Technical Problem

Existing damage prediction methods for rotary tools fail to accurately predict tool damage when the operating state changes, such as during operations other than machining or variations in machining intervals, due to power consumption fluctuations from movements and non-machining operations.

Method used

A damage prediction device that acquires and analyzes time-series data of spindle current and vibration measurements during forward rotation machining periods, calculates statistical values, and uses regression equations to predict tool damage by assessing deviations from established relationships, considering air-cut operations and non-processing periods.

Benefits of technology

Enables accurate prediction of rotary tool damage even when the operating state changes, improving the reliability of tool life assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026095785000001_ABST
    Figure 2026095785000001_ABST
Patent Text Reader

Abstract

The present invention provides a damage prediction device that can predict damage to a rotary tool even when the operating state of the rotary tool changes. [Solution] The damage prediction device comprises an acquisition unit, an extraction unit, a calculation unit, a storage unit, a regression equation calculation unit, and a prediction unit. The acquisition unit acquires measured values ​​from the measuring device of the machining machine. The extraction unit extracts the measured value of the forward rotation machining period of the rotary tool from the time-series data of the measured values. The calculation unit calculates at least one of the length of the forward rotation machining period and the maximum value of the measured value as a calculated value from the time-series data of the extracted measured values. The storage unit stores the calculated value as statistical data, associating it with the number of machining operations performed by the rotary tool. The regression equation calculation unit calculates a regression equation showing the relationship between the number of machining operations and the calculated value from the statistical data stored in the storage unit. The prediction unit calculates the degree of deviation of the calculated value from the regression equation calculated by the regression equation calculation unit, and predicts damage to the rotary tool according to the degree of deviation.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method and device for determining life of rotary tool

    JP6864394B1