Intelligent vision based wear forecasting on surfaces of machine tool elements

Abstract

To realize autonomous production machines it is necessary that machines are able to automatically and autonomously predict their condition. Although many classical as well as Deep Learning based approaches have shown the ability to classify faults, so far there are no approaches that go beyond the basic detection of faults. A complete, image based predictive maintenance approach for machine tool components has to the best of our knowledge not been investigated so far. In this paper it is shown how defects on a Ball Screw Drive (BSD) can be automatically detected by using a machine learning based detection module, quantified by using an intelligent defect quantification module and finally forecasted by a prognosis module in a combined approach. To optimize the presented method, it is shown how existing domain knowledge can be formalized in an expert system and combined with the predictions of the machine learning model to aid quality of the prediction system. This enables the practitioner to forecast the severity of failures on BSD and prevent machine breakdowns. The work is intended to set new accents for the development of practical predictive maintenance systems and bridging the fields of machine learning and production engineering. The code is available under: https://github.com/2Obe/Pitting_Pred_Maintenance.

Data mining in predictive maintenance systems: A taxonomy and systematic review

“…This feature makes them especially suitable for detecting spatial features in the data, which makes them very relevant in image analysis. In this sense, in Schlagenhauf and Burghardt (2021), a framework is developed that uses CNNs to detect if there is damage on the imaged surface. In the next phase of the model, classical vision techniques are used to establish the severity threshold of the detected damage.…”

“…Thus, in Oliveira et al (2020) a pulse‐echo ultrasonic technique based on local immersion was used to acquire data from wind turbine blade test specimens. Schlagenhauf and Burghardt (2021) proposed an image‐based analysis from the photographs of the degradation of the analyzed surface of a ball screw drive. In Lasisi and Attoh‐Okine (2018) and Consilvio et al (2020), the track geometry measures of railway assets are used, specifically in Consilvio et al (2020), as a complementary source of data together with inspection logs.…”

“…Another example of use can be found in Saranya and Sivakumar (2020) and Kang et al (2021), where different multivariate LiR are employed, respectively, to estimate RUL. In Schlagenhauf and Burghardt (2021), LiR is the last step in a decision support system that estimates the evolution of a damage area in the monitored surface. In Chang et al (2021), different regression methods are compared to analyze complex manufacturing data and predict the future behavior of the monitored machine.…”

“…Another common domain is the ball bearing, with the Case Western Reserve University bearing dataset ( Bearing Dataset , CSE Groups, 2008) as the most popular dataset. Component surface defect dataset (Schlagenhauf & Landwehr, 2021) and Alarms logs dataset (Tosato et al, 2020) have been generated from PdM article included in this review (Pezze et al, 2021; Schlagenhauf & Burghardt, 2021). Some of the datasets listed in Tables 7 and 8 do not report usage in the period reviewed, which is partly because they are too modern or offer similar features to more popular ones.…”

Software-Defined Workpiece Positioning for Resource-Optimized Machine Tool Utilization

“…In order to detect damages on the surface of ball screw spindles at an early stage, Schlagenhauf et al [31] developed a monitoring approach using an integrated camera system. To enable machines to detect and predict the spindle condition, a method based on machine learning was developed in [32] to interpret defects in ball screws automatically and autonomously. An intelligent defect quantification module quantifies the defects, which are then predicted by a prognosis module in a combined approach.…”