Table of contents:
- The feeling machine tool
- Control quality with augmented reality
- The grinding process becomes intelligent
Video: On The Way To Autonomous Production
Autonomous production systems have the ability to plan and control production themselves. In addition, they are able to adapt to unforeseen changes in the environment. As a result of the ongoing trend towards individualized products, the proportion of individual parts is increasing. That requires first time right in production.
Update: EMO 2019
World's leading trade fair for metalworking is in the starting blocks
The basic requirement is process monitoring for individual parts. In contrast to series production, in which the processes are monitored using previous process data, it is not possible to use existing measurement values in single-part production. In order to still be able to monitor the first copy, reference values must be determined via process simulation. Using the example of the production of an impeller, the spindle current is forecast based on a highly precise real-time material removal simulation and with a neural network. The comparison of simulated and measured process data enables the detection of process errors and thus the monitoring of individual parts.
Permanent status monitoring for valuable and complex capital goods, such as high-pressure turbine blades, is the current state of the art. In contrast to this, however, is their manual repair - also state of the art - which often only depends on the subjective assessment of the processing employee and thus on his training and experience. To regenerate complex capital goods, the scientists are building an automated process chain that is supposed to change exactly that.
In addition to the real level in which the machine tools are located, there is a virtual level. Here, the turbine blades that need to be repaired are simulated and their condition, i.e. the actual performance and the actual service life, are determined. The individual repair paths are generated from this state and the customer's wishes and carried out on the real level using newly developed repair processes.
Emo 2019
Gestalt Robotics - a start-up introduces itself
construction
Design smart machines efficiently
The feeling machine tool
Individual products are becoming increasingly important in modern medical technology. Customized implants can prevent complications after the operation and lead to a higher quality of life for the patient. In order to produce the final contour of the complex implant geometries using milling machines, the entire process chain from CAD design and path planning to fine adjustment of the process parameters must be run through and adapted - over and over again. Since this is very time and cost intensive, individual implants are usually not economically viable.
The scientists at the IFW have now developed the sensing machine to determine process forces. By merging the measured process force and the process-parallel removal simulation (digital twin) developed at the IFW, the intervention conditions can be exactly reproduced and recorded. Problems, for example excessive tool deflection, can thus be identified early and resolved in a targeted manner.
SEMINAR TIP With the online course "Machine Learning - how does the machine learn?" You can delve deeper into the methodology. Professor van der Smagt - Head of Artificial Intelligence at Volkswagen - explains to participants in two hours the most important concepts of machine learning in a concentrated and compact manner.
Further information
A cross-process learning system autonomously makes the necessary path adjustments. The system significantly reduces the effort required to set up new processes. This reduces the manufacturing costs. Thanks to the new technologies, individualized implants will also be available to the general public in the future.
roller bearing
Smart machine tools thanks to digital roller bearing solutions
Control quality with augmented reality
The individual production of customer-specific requirements, an important goal of Industry 4.0, requires more effort for quality assurance. In order to continue to guarantee the cost-effectiveness of manufacturing individual components, new methods for better quality assurance must be derived. For this purpose, the use of augmented reality is being researched at the IFW. The real world of manufacturing is expanded with virtual data about the process. This is intended to enable the machine operator to check the finished component efficiently. By providing additional information that is obtained by means of process simulation, the quality should be checked in a shorter time.
Current research content is, for example, the virtual representation of the process forces during machining on the real component. Using this, the machine operator can draw conclusions about the quality of the manufactured component in the machine room after the process and adjust the process if necessary.
The grinding process becomes intelligent
The tool grinding process is becoming more diverse in the single-part and small series production of milling and drilling tools. In addition, the evaluation of the grinding wheel condition is linked to the operator experience. In combination with conservative deliveries during the dressing process, this leads to longer non-productive times. By introducing a condition assessment based on measurement data and digitizing the decision-making process, significantly fewer resources and less time are required, and this with the same workpiece quality. In order to achieve this goal, intelligent grinding is being developed at the IFW.
Integrated measuring technology in the machine enables the condition of the tools to be determined. This information is used in conjunction with a self-learning wear model to assess the current wear status of the grinding wheel. On this basis, the grinding process can be adjusted autonomously, productivity and process stability can be maximized. (ud)
