No abstract

In modern machine architecture, complex mechanical components guarantee the function of the system. Specific maintenance rates reduce maintenance effort and cost. The combination of load and failure monitoring provide data to establish an individual maintenance rate depending on component specific remaining lifetime. In addition, algorithms can optimise the operation strategy to relieve the weakest components and extend the system lifetime. Adapted ball bearings contain a sensor mechanism depending on the relation of beating load and electric bearing impedance to measure load and failure data.

This paper discusses the difficulties caused by production precision for a widespread application of sensorial Mechatronic Machine Elements to expand existing machines or designs to Cyber-Physical-Systems. The production induced tolerances lead to an undetermined flow of forces in machine elements and prevent a reliable sensing of forces in Mechatronic Machine Elements. A promising approach to overcome these difficulties is Robust Design, which leads to determined mechanical functions that reduce uncertainty of force measurement in Mechatronic Machine Elements.

The present paper describes a measurement setup and a related prediction of the electrical impedance of rolling bearings using machine learning algorithms. The impedance of the rolling bearing is expected to be key in determining the state of health of the bearing, which is an essential component in almost all machines. In previous publications, the determination of the impedance of rolling bearings has already been advanced using analytical methods. Despite the improvements in accuracy achieved within the calculations, there are still discrepancies between the calculated and the measured impedance, leading to an approximately constant off-set value. This discrepancy motivates the machine learning approach introduced in this paper. It is shown that with the help of the data-driven methods the difference between analytical prediction and measurement is reduced to the order of up to 2% across the operational range analyzed so far. To introduce the context of the research shown, first the underlying physics of bearing impedance is presented. Subsequently different machine learning approaches are highlighted and compared with each other in terms of their prediction quality in the results part of this paper. As a further aspect, in addition to the prediction of the bearing impedance, it is investigated whether the rotational speed present at the bearing can be predicted from the frequency spectrum of the impedance using order analysis methods which is independent from the force prediction accuracy. The background to this is that, if the prediction quality is sufficiently high, the additional use of speed sensors could be omitted in future investigations.

In this paper the development process of a sensing rolling bearing is presented, from which finally design rules for sensing machine elements are derived. In the first step, the requirements of the users are determined. It turns out user of sensing machine elements want to continue to use the advantage of the standardized machine elements and costs should not be incurred by redesign or complex assembly. With these requirements the development of the sensing rolling bearing is started, in which the different presented technologies are reviewed for their suitability regarding the requirements. With the selected technology measuring the electric rolling bearing impedance to estimate rolling bearing loads, a first prototype is developed by creating a functional structure of the product and focusing on the partial solution of the most relevant partial functions. This prototype is then tested with regard to its functionality. Finally, generalizable design rules for sensing machine elements are derived from the development.