Session: 02-03-02: Optimization

Paper Number: 141505

141505 - Design Space Optimization for Sensor-Integration in Standardized Machine Elements 

In the context of digitization and Industry 4.0, data-driven analysis such as condition monitoring and predictive maintenance are based on extensive measurement data. Nevertheless, there is a lack of data on machine or plant health. Sensor-integration into machine elements (SiME) can fill this gap due to their widespread usage, standardization and location that enables in-situ measurement of state-relevant data. Sensor-integration usually compromises the mechanical function of the machine element. To ensure broad usability and retrofitting of SiME it is crucial that their mechanical functions are maintained to a defined degree.

There are various approaches to develop SiME, such as bolts or jaw couplings, as outlined in the state-of-the-art. Design space identification is particularly challenging in the development of SiME due to conflicting objectives between the mechanical and electrical domains.

There is a lack of procedures to systematically find an optimal solution for design space. Therefore, it is not guaranteed that all relevant boundary conditions are respected. A procedure to identify an optimal design space that considers multi-axial force sensor-integration into M20 bolts was developed in prior work.

However, this procedure is not yet validated for use on other bolt sizes or other machine elements. Therefore, this contribution applies the procedure to (1) M12 bolts for uni-axial force measurements by strain gauges and (2) jaw couplings for torque measurement by capacitive sensors. Since the dimensions and the measurement task vary, specifications to the procedure are required, which are described in the following steps.

First, quantities to evaluate mechanical and sensory function fulfillment are identified. The v. Mises stress is suitable for describing the mechanical function for bolts. For the jaw-coupling however, the change of displacement is more suitable. Regarding the sensory function, the sensitivity and accuracy is linked to the changes in strain or displacement at the sensor position.

Second, a design space parameter set for sensor and electronic components within the machine element is defined, impacting both mechanical and sensory functions. A distinction between design space for sensors and electronics is essential. Sensors require placement in high-strain regions for signal acquisition, whereas electronics necessitate strain avoidance for protection.

Third, boundary conditions are defined to consider only viable solutions to the design space problem. For bolts, stress increase caused by the design space cavities is limited to the difference between two consecutive bolt material classes. For couplings, maximum torque is limited to the next smaller size. Electronics necessitate minimal space for desired functionality if off-the-shelf components are used.

Fourth, a finite element model is set up. Numerical simulations are run with varying design space parameter sets, extracting the evaluation quantities defined before.

Fifth, functions are fitted on the evaluation quantities specified before with respect to design space parameter sets by usage of curve fitting and machine learning approaches.

Sixth, an objective function combining the fitted mechanical and sensory functions is constructed. Weights enable shifting of the priorities between mechanical and sensory parts.

Finally, optimal solutions for the cases of the M12 bolt and the jaw coupling are retrieved by minimizing the objective function. As a result, design space parameter sets for the bolt and the coupling are identified that fulfill the formulated boundary conditions and balance the mechanical versus the sensory function. Also, new boundary conditions are discovered that need to be considered in the objective functions, for example manufacturing restrictions or the bucking of the sensor stack in the coupling. With knowing the effect of changing the design space parameters on the mechanical and sensory function trade-offs can be made.

With such optimized design spaces sensor-integration into machine elements (SiME) can be taken to the next level, building a base for capturing extensive measurement data needed for digitization and Industry 4.0 approaches. In future, physical testing is required to validate the solutions. Also, more design space parameters with influence on the mechanical or sensory function need to be included in the objective function.

Presenting Author: Julian Peters Institute of Product Engineering, Karlsruhe Institute of Technology

Presenting Author Biography: Julian Peters received his M. Sc. in the field of mechatronics and information technology at the Karlsruhe Institute of Technology (KIT) in 2020. During his studies he did student jobs in and out of Germany dealing with automation solutions using industrial robots, robotic legs, and prototyping for assembly lines, among others. Since 2020 he is working as researcher and PhD candidate at the chair of Prof. Sven Matthiesen at KIT in Karlsruhe, focused on integrating sensors into machine elements to gain insight in health states and processes of machines and support comprehensive digitization.

Authors:

Julian Peters Institute of Product Engineering, Karlsruhe Institute of Technology
Johannes D. M. Menning Institute of Solid Mechanics, Technische Universität Dresden
Arthur Ewert Institute of Machine Elements and Machine Design, Technische Universität Dresden
Artem Prokopchuk Institute of Semiconductors and Microsystems, Technische Universität Dresden
Richard Breimann Institute for Product Development and Machine Elements, Technical University of Darmstadt
Felix Herbst Measurement and Sensor Technology Group, Technical University of Darmstadt
David Riehl Integrated Electronic Systems Lab, Technical University of Darmstadt
Markus Döllken Institute of Product Engineering, Karlsruhe Institute of Technology
E.-F. Markus Vorrath Institute of Semiconductors and Microsystems, Technische Universität Dresden
Thomas Wallmersperger Institute of Solid Mechanics, Technische Universität Dresden
Berthold Schlecht Institute of Machine Elements and Machine Design, Technische Universität Dresden
Eckhard Kirchner Institute for Product Development and Machine Elements, Technical University of Darmstadt
Mario Kupnik Measurement and Sensor Technology Group, Technical University of Darmstadt
Klaus Hofmann Integrated Electronic Systems Lab, Technical University of Darmstadt
Sven Matthiesen Institute of Product Engineering, Karlsruhe Institute of Technology