AutoBohr - System zur autonomen Prozessüberwachung von Bohrprozessen

Automated process monitoring is becoming increasingly important in single-part and small-batch production. Drilling processes pose particular challenges for companies. Various disruptive influences such as insufficient coolant supply, friction effects, and chip jams lead to unpredictable process signals. Existing systems are based on fixed thresholds or reference signal envelopes. These systems are sensitive to aforementioned fluctuations, which leads to false alarms. Tool breakages are only detected after they have already occurred. This causes scrap and machine downtime. A robust, autonomous, and predictive monitoring system for single-part and small-batch production that detects tool breakages before they occur and reliably detects wear is not yet available on the market.

Objectives

The intended process monitoring system should autonomously recognize drilling processes and segment them into individual process phases. AI is being developed to predict tool breakages and other anomalies. The system configures itself automatically, eliminating the need for manual configuration by the user. Another goal is transferability. The system should be able to adapt itself to different machines and process configurations.

Benefits

Advantages/potential benefits for practical application:

• Can be used in single-part and small-batch production
• No need for trained personnel for parameterization
• Easy integration into existing production
• Robust monitoring despite chip jams
• Reduction of scrap, downtime, and costs
• Increase in productivity

Approach

The project is being carried out in close cooperation with iba AG. The ibaDAQ and iba HD Store systems are used to record internal machine signals and combine them with acceleration signals from a sensor. Data is collected both at IFW and at a contract manufacturer in order to attain practical drilling processes. Based on this data, AI algorithms for autonomous segmentation and predictive detection of tool breakages and anomalies are then developed, tested, and validated.

Are you also interested in a cooperation project?

Contact Kirill Tkachuk via email at tkachuk@ifw.uni.hannover.de or by phone at +49 511 762 18382.