Tightening Class Estimation Using Context-Aware ML Models

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Determining the correctness of tightening operations by a tightening tool is challenging due to the difficulty in analyzing sensor data, and existing machine-learning models may perform poorly for specific tools or environments, leading to inaccurate estimations of tightening classes.

Innovation Solution

A method that involves acquiring torque and angle values, context data such as environmental conditions and tool settings, and training a machine-learning model to improve the accuracy of tightening class estimations, using neural networks or random forest-based classification, and providing alerts for incorrect operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If machine-learning models are used to analyze torque and angle values, then automation and productivity are improved, but measurement precision and reliability deteriorate due to poor model performance for specific tools or environments

Engineering Contradiction:

Improveautomation of tightening analysisVSAvoidtightening class estimation accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies local quality by training separate machine-learning models for different tightening tool types and environmental conditions. Each model is specialized to handle specific tool characteristics and operating environments, thereby maintaining high measurement precision while achieving automation across diverse scenarios

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent changes parameters by incorporating context data (environmental conditions, tool settings, fastener properties) as additional inputs to the machine-learning models. This allows the models to adapt to varying conditions and maintain accurate tightening class estimations across different environments and tool types

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If context data is acquired and integrated into the model, then measurement precision is improved, but device complexity increases

Engineering Contradiction:

Improvetightening class estimation accuracyVSAvoiddata acquisition and processing system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements universality by designing a multi-functional data acquisition system that collects various types of context data (environmental, tool, fastener) using integrated sensors and communication interfaces. This single system serves multiple purposes: characterizing the operating environment, identifying tool properties, and providing input features for the machine-learning model, thereby managing complexity while improving precision

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20240369431A1ML estimation of tightening classes

Publication Date: 2024.11.07 ATLAS COPCO IND TECHNIQUE AB INTELLECTUAL PROPERTY DEPARTMENT

AI summary

The present disclosure relates to a method of a device for determining a tightening class of a tightening operation performed by a tightening tool. The determination is based on training a machine-learning model with an acquired set of observed torque and angle values, at least one set of context data and one or more associated tightening classes. Subsequently, the trained model is utilized for determining from supplied torque and angle values, as well as context data, one or more tightening classes associated with the supplied values and context data.