AI for retrofit Artificial intelligence helps in machine modernization

Related Vendors

FAULHABER Drive Systems

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

SIMTOS (Seoul International Machine Tool Show)

Researchers from Hannover (Germany) have investigated how process monitoring with artificial intelligence (AI) can assist in cost-effectively retrofitting aging machines.

Artificial intelligence in process monitoring can reduce scrap, improve component quality, and ease the burden on personnel. Expensive investments in new machines are not necessarily required for this. This has been demonstrated by a successfully completed research project called "AutoPress" by the IPH—Institute for Integrated Production Hannover and Jobotec GmbH. The experts have reportedly developed a system of sensors and AI that allows old machines to be retrofitted. This was developed using a spindle press as an example, but the principle could also be applied to other machines and systems.

Less waste and a more relaxed personnel situation

The process monitoring based on AI detects parameter deviations with a success rate of 95 to 98 percent. If errors are detected, the person operating the system receives feedback:

• "Warning, the tool is not installed correctly!"

• "Warning, the raw material is not centered!"

• "Warning, you have inserted the wrong material!"

Companies can thus not only detect errors early, avoid waste, and improve component quality, but also relieve their staff. With AI support, even less qualified employees can operate the machines correctly. In times of skilled labor shortage, this research success represents a real competitive advantage.

Manufacturing more sustainably through retrofit

To use the latest sensors and AI, it is not necessary to acquire expensive new machines, the researchers note. Instead, old machines can be brought up to date through retrofitting. A retrofit is generally also cheaper than a new purchase. It is also more sustainable, as it allows machines to remain in use for longer. Looking at the old press, the project partners have made sure to work with affordable components and existing industry standards. This approach enables small and medium-sized enterprises to retrofit their machines without much effort.

The spindle press was equipped with laser distance sensors, voltage measurement sensors, and temperature sensors. Various AI models evaluate the measurement results and compare them with the ideal parameters. If deviations are detected, the system provides recommendations for action. With the help of the retrofitted sensors, deviations from the ideal height of the semifinished product can be detected, for example. A few millimeters can have major consequences here! If too little material is placed in the press, the mold will not be filled, resulting in scrap. If the semifinished product is placed too high, material and energy are wasted. In both cases, the AI is able to provide a recommendation on how many millimeters the height of the semifinished product must be corrected to avoid either scrap or waste.