Accurate wear predictions and estimation of the total cost of ownership for centrifugal pumps are critical for the slurry industry. Although the cost of pump components, such as impellers, casings, and liners, is a concern for predicting wear, unplanned shutdowns in production due to failure of a component can be far more expensive. They negatively affect the overall productivity, and many plants estimate unplanned shutdowns can cost hundreds of thousands of dollars per hour. Multi-phase computational fluid dynamics (CFD) provides accurate predictions for slurry erosion wear of different pump components, for example, pump impellers and pump casings. The erosion wear can be used to accurately estimate the downtime of the pump so that the maintenance can be synchronized to minimize the shutdown period. However, predictive CFD models come with a high computational cost that is not typically affordable by normal users, such as designers and engineers, in a timely manner. Fortunately, data-driven approaches, also known as machine learning (ML) approaches, are capable of providing accurate wear predictions, at a fraction of the computational cost, if the ML wear models are sufficiently and well trained using CFD wear models. Utilizing ML techniques to reduce the high computational cost of CFD while maintaining its predictive capability is a practical and effective approach to bring the power of CFD tools to more interested users. Furthermore, the low computational cost of the ML wear models also allows one to efficiently explore the high-dimensional input space. In this paper, we apply the WearGP framework (*Tran, A., Furlan, J. M., Pagalthivarthi, K. V., Visintainer, R. J., Wildey, T., & Wang, Y. (2019). WearGP: A computationally efficient ML framework for local erosive wear predictions via nodal Gaussian processes. Wear, 422, 9-26.*) to accurately predict wear rates as the global quantities of interests (QoIs), and explore the wear behaviors as a function of pump operating conditions.


Weargp: A Machine Learning Wear Prediction Framework for Slurry Pump Impellers and Casings

Division / Sponsors: 
Heat Transfer Division (HTD), Fluids Engineering Division (FED)

Heat Transfer Summer Conference
Paper Proceedings: https://asmedigitalcollection.asme.org/HT/HT2020/volume/83709

18th International Conference on Nanochannels, Microchannels, and Minichannels
Paper Proceedings: https://asmedigitalcollection.asme.org/ICNMM/ICNMM2020/volume/83693

Fluids Engineering Division Summer Meeting
Paper Proceedings: https://asmedigitalcollection.asme.org/FEDSM/FEDSM2020/volume/83716

HT-ICNMM2020 - Virtual Conference, SHTC - Virtual Conference, FEDSM2020 - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.