Nowadays, hydropower plants are forced to have more frequent power control and the self-lubricated bearings used in the applications are one of the most critical components affected by the continuously changing operating conditions. In this study, microstructure and composition of two commercially available bearing materials (Orkot TXM Marine and Thordon ThorPlas) used in hydropower turbines were studied. In addition, the influence of sliding speed and applied pressure on the friction and wear behavior of the materials was investigated systematically for dry sliding conditions. The bearing materials were characterized using X-ray microtomography, Nuclear Magnetic Resonance (NMR) spectroscopy and Inductively Coupled Plasma–Sector Field Mass Spectrometry (ICP-SFMS) techniques. Friction and wear tests were carried out with a polymer pin sliding against a stainless steel (SS2333) plate with a linear reciprocating motion. Test conditions were: room temperature, 9–28 MPa pressure and 10–40 mm/s sliding speed ranges. Surface analysis of the polymer pins and the wear tracks were performed by optical profilometry, Scanning Electron Microscope (SEM) and Energy Dispersive Spectroscopy (EDS) techniques. Test results show that, for both materials, the coefficient of friction (COF) is decreasing at higher pressures. Surface analysis reveals higher concentrations of solid lubricants in the transfer layers formed at higher pressures, explaining the decrease in COF. Furthermore, the specific wear rate coefficients are increasing at higher sliding speeds, especially at lower pressures. Results of this study demonstrate that, under dry sliding conditions, changes in sliding speed and pressure have a significant influence on the tribological behavior of these bearing materials.

There are many moving machine assemblies with conformal tribological contacts at very high contact pressures, e.g., sliding bearings, propeller shaft bearings and machine guideways. Furthermore, applications such as trunnion and guide vane bearing in Kaplan turbines have very low sliding speeds and oscillatory types of motion. Although there is a vast selection of tribology test rigs available, there is still a lack of test equipment to perform friction and wear tests under high contact pressure, reciprocatory sliding and large area contact. The aim of this work is thus to develop a novel reciprocating tribometer and test method that enables friction and wear tests under low-speed reciprocatory sliding with contact pressures up to 90 MPa in a flat-on-flat contact configuration. First, a thorough description of the test rig design is given. Secondly, the influence of contact pressure and stroke length on the tribological properties of a stainless steel and polymer composite material combination is studied. The significance of considering creep, friction during the stroke and contact temperature is specifically highlighted. The novel tribometer can be used to screen different bearing and shaft material combinations and to evaluate the friction and wear performance of self-lubricating bearings for the specific operating conditions found in Kaplan turbines.

Technical failures in machines are major sources of unplanned downtime in any production and result in reduced efficiency and system reliability. Despite the well-established potential of Machine Learning techniques in condition-based maintenance (CBM), the lack of access to failure data in production machines has limited the development of a holistic approach to address machine-level CBM. This paper presents a practical approach for failure mode prediction using multiple sensors installed in a bearing ring grinder for process control as well as condition monitoring. Bearing rings are produced in a set of 7 experimental runs, including 5 frequently occurring production failures in the critical subsystems. An advanced data acquisition setup, implemented for CBM in the grinder, is used to capture information about each individual grinding cycle. The dataset is pre-processed and segmented into grinding cycle stages before time and frequency domain feature extraction. A sensor ranking algorithm is proposed to optimize feature selection for failure classification and the installation cost. Random forest models, benchmarked as best performing classifiers, are trained in a two-step classification framework. The presence of failure mode is predicted in the first step and the failure mode type is identified in the second step using the same feature set. Defining the feature set in the failure detection step improves the predictor generalization with the classifiers’ performance accuracy of $$99\%$$99% on the test dataset. The presented approach demonstrates an efficient failure mode classification by selecting crucial sensors resulting in a cost-effective CBM implementation in a bearing ring grinder.