CNC Tool Wear Monitoring for Sensorless RUL Estimation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current tool wear monitoring technologies in machining are not cost-effective or robust for industrial-scale use, often failing to accurately capture tool wear physics due to external noise interference and requiring large datasets specific to machining conditions, leading to inaccurate predictions and inefficient tool changes.

Innovation Solution

A processor-implemented method and system that derive the rate of volumetric wear loss per unit contact area of a tool using spindle power, radial and axial depths of cut, cutting velocity, and temperature wear coefficients, allowing for real-time estimation of Remaining Useful Life (RUL) without external sensors, seamlessly indicating when the RUL crosses predefined thresholds.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sensor-based signal processing techniques are used for tool wear monitoring, then measurement capability is improved, but device complexity and cost increase

Engineering Contradiction:
Improvetool wear detection accuracyVSAvoidsensor and signal processing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the essential wear information directly from CNC machine control data (spindle power, cutting parameters) without requiring external sensor systems. This removes the complex sensor-based measurement infrastructure while retaining the ability to monitor tool wear through processed control signals that already exist in the machining system.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system uses the CNC machine's own control data and parameters to monitor its own tool wear condition. By leveraging existing spindle power measurements and cutting parameters from the machine controller, the system enables self-diagnosis without external monitoring equipment, reducing device complexity while maintaining monitoring capability.

Inventive Principle:
Principle #25Self-service

2Reliability

If machine learning techniques are used for tool wear prediction, then prediction capability is improved, but data collection complexity and model adaptability requirements increase

Engineering Contradiction:
Improvetool wear prediction accuracyVSAvoidmodel adaptability to different machining conditions
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent transforms the approach by changing from collecting diverse sensory data to utilizing standardized CNC control parameters (spindle power, cutting velocity, depths of cut). By modifying the input data type to these controlled parameters that are consistently recorded during machining, the system achieves reliable predictions without requiring complex data collection infrastructure or frequent model retraining for different conditions.

Inventive Principle:
Principle #35Parameter changes

3Loss of information

If external sensors are mounted for data collection, then measurement data availability is improved, but system cost and implementation difficulty increase

Engineering Contradiction:
Improvemachining data availabilityVSAvoidsystem implementation ease
Core Design Contradiction:
Loss of informationVSEase of manufacture

Solution Approach 1:

The patent uses the CNC machine controller as an intermediary to access machining parameters. Instead of mounting sensors directly on the machine, the system queries the controller for spindle power, cutting parameters, and operational data that are already being tracked for machining control, thereby obtaining necessary information without additional hardware installation.

Inventive Principle:
Principle #24Intermediary (Mediator)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The solution provides a stable, reliable, and robust tool wear prediction model that is cost-effective and scalable, capturing tool wear physics effectively, reducing part inaccuracies and improving machining precision by enabling timely tool changes.

Implementation Method 1

The temperature wear coefficient Kw considers effect of temperature rise due to friction caused by current tool wear state of the tool during the machining operation

Methodology Applied
Scientific EffectFriction: Friction

Data Source

PatentEP3864475B1Method and system for monitoring tool wear to estimate RUL of tool in machining
Publication Date: 2024.01.03 TATA CONSULTANCY SERVICES LTD
  • EP3864475B1 patent drawingFigure 1
  • EP3864475B1 patent drawingFigure 2
  • EP3864475B1 patent drawingFigure 3

AI summary

Tool wear monitoring is critical for quality and precision of manufacturing of parts in the machining industry. Existing tool wear monitoring and prediction methods are sensor based, costly and pose challenge in ease of implementation. Embodiments herein provide method and system for monitoring tool wear to estimate Remaining Useful Life (RUL) of a tool in machining is disclosed. The method provides a tool wear model, which combines tool wear physics with data fitting, capture practical considerations of a machining system, which makes the tool wear prediction and estimated RUL more stable, reliable and robust. Further, provides cost effective and practical solution. The disclosed physics based tool wear model for RUL estimation captures privilege of physics of tool wear and easily accessible data from CNC machine to monitor and predict tool wear and RUL of the tool in real-time.