Global load determination in linear guides based on the fusion of local rolling element loads determined from strain sensitive sensor groups

Abstract

Abstract Measuring the mechanical load on linear guides provides many possibilities regarding predictive maintenance and process monitoring. In this contribution, we provide an in depth evaluation of a Diamond Like Carbon (DLC) based sensor system integrated into the runner block’s raceway that is capable of directly measuring the load on individual rolling elements. An efficient algorithm based on an Extended Kalman Filter (EKF) for local sensor fusion and load estimation is presented and proven to reliably retrieve the load regardless of the rolling element’s position. Afterwards, we compare locally measured loads to results from a theoretical load distribution model, providing valuable insight into modeling parameters and a verification of the sensor measurement principle. In a final step, an algorithm to invert the load distribution model is derived and used for an evaluation of the sensor system, achieving Root-Mean-Square (RMS) estimation errors of equivalently 1.4 kN in the preload range and 2.75 kN overall for one dimensional loads. Load mode distinction was equally successful with a suppression RMS error of 0.7 kN in the preload range and 2.87 kN in total.