Focus on First Open Call winners: Intelligent measurement of tool wear at the edge

WHAT ABOUT THE EXPERIMENT? 
 

The development of Monitoring Systems for autonomous decision-making plays a fundamental role in the era of Industry 4.0, aiming to automatically optimize production processes comprehensively.

In the machining industry, the most significant losses are due to downtime. It is in this domain that the most substantial improvement potential lies. Among these downtimes, those attributed to cutting tool failure rank among the most prominent. Moreover, both direct tool costs (purchase cost) and indirect costs (downtime due to tool changes, errors resulting from tool failure, wastage, etc.) collectively contribute to 3-12% of the total production cost.

Since the 1980s, effort has been put into Tool Condition Monitoring Systems (also known as Tool Condition Monitoring Systems or TCMS) to respond to this need. Most TCMS estimate the condition of the tool by monitoring indirect variables, such as power consumed, acoustic emission or vibrations. Given the number of tool types, manufacturing processes, types of materials, cutting parameters, etc, cases tend to infinity and trained AI models do not have the ability to adapt to new use cases (generalize). It's as if they were overtrained for specific cases.

The tool condition detection models used to date are severely lacking and are not useful tools for the large number of cases that occur in reality. As a general rule, they work well in large batch production (millions of parts in stable environments). The adaptability of these systems to new situations requires long training phases for the models.

With this in mind, EHiNA aims at developing and AI at the edge driven system to optimise the use of cutting tools, reducing direct and indirect costs associated with conservative tool changes. EHiNA aims to rely on an edge-cloud architecture that allows the various models deployed around the world to learn as if they were a single model.

​

WHAT IS THE EXPECTED IMPACT?  
 

Nowadays, in the sector of machine-tool, tool manufacturing companies have no access to the behaviour of their tools during the manufacturing process and have little opportunity to enhance their product. Moreover, there no system that can say how much cutting time the tool has left and recommends optimal cutting parameters. Therefore, manufacturing companies replace the tools before the end of their lifetime, to avoid defective pieces, but wasting tools.

The implementation of EHiNA represents a step forward to achieve a more sustainable manufacturing as tools will be replaced only when they reach the end of their lifetime, without any loss of quality. The advantages of the system are not only tool savings, but it will also make the process more efficient as the tool manufacturer will have access to process data and can provide better tools by optimizing the shape and the composition of the tool.