Gaining insight into tribological systems is crucial for optimizing efficiency and prolonging operational lifespans in technical systems. Experimental investigations are time-consuming and costly, especially for reciprocating seals in fluid power systems. Elastohydrodynamic lubrication (EHL) simulations offer an alternative but demand significant computational resources. Physics-informed neural networks (PINNs) provide a promising solution using physics-based approaches to solve partial differential equations. While PINNs have successfully modeled hydrodynamics with stationary cavitation, they have yet to address transient cavitation with dynamic geometry changes. This contribution applies a PINN framework to predict pressure build-up and transient cavitation in sealing contacts with dynamic geometry changes. The results demonstrate the potential of PINNs for modeling tribological systems and highlight their significance in enhancing computational efficiency.

In many technical applications, understanding the behavior of tribological contacts is pivotal for enhancing efficiency and lifetime. Traditional experimental investigations into tribology are often both costly and time-consuming. A more profound insight can be achieved through elastohydrodynamic lubrication (EHL) simulation models, such as the ifas-DDS, which determines precise friction calculations in reciprocating pneumatic seals. Similar to other distributed parameter simulations, EHL simulations require a labor-intensive resolution process. Physics-informed neural networks (PINNs) offer an innovative method to expedite the computation of such complex simulations by incorporating the underlying physical equations into the neural network’s parameter optimization process. A hydrodynamic PINN framework has been developed and validated for a variant of the Reynolds equation. This paper elucidates the framework’s capacity to handle multi-case scenarios—utilizing one PINN for various simulations—and its ability to extrapolate solutions beyond a limited training domain. The outcomes demonstrate that PINNs can overcome the typical limitation of neural networks in extrapolating the solution space, showcasing a significant advancement in computational efficiency and model adaptability.

A comprehensive understanding of the dynamics of tribological interactions is essential for enhancing efficiency and durability in a multitude of technical domains. Conventional experimental techniques in tribology are frequently costly and time-consuming. In contrast, elastohydrodynamic lubrication (EHL) simulation models present a viable alternative for calculating frictional forces in sealing contacts. These calculations are based on the hydrodynamics within the sealing contact, as defined by the Reynolds equation, the deformation of the seal, and the contact mechanics. However, a significant drawback of these simulations is the time-consuming calculation process. To overcome these experimental and computational limitations, machine learning algorithms offer a promising solution. Physics-informed machine learning (PIML) improves on traditional data-driven models by incorporating physical principles. In particular, physics-informed neural networks (PINNs) are as effective hybrid solvers that combine data-driven and physics-based methods to solve the partial differential equations that drive EHL simulations. By integrating physical laws into the parameter optimization of the neural network (NN), PINNs provide accurate and fast solutions. Thus, unlike traditional NNs, PINNs have the potential to make accurate predictions beyond the limited training domain. The objective of this study is to demonstrate the feasibility of spatial and temporal extrapolation of the PINN and to analyze its reliability, both with and without consideration of cavitation. Two test cases are employed to examine the pressure and cavitation distribution within a sealing contact that extends beyond the spatial and temporal training range. The findings indicate that PINNs can surmount the typical constraints associated with NNs in the extrapolation of solution spaces, which represents a notable advancement in terms of computational efficiency and model flexibility.

"In numerous technical applications, gaining insights into the behavior of tribological systems is crucial for optimizing efficiency and prolonging operational lifespans. Experimental investigations of such systems require considerable costs and time investments, particularly in the field of sealing, notably reciprocating seals for fluid power systems. A more feasible method is the application of elastohydrodynamic lubrication (EHL) simulation models, such as the dynamic description of sealings (DDS) model, which compute friction of seals by the hydrodynamics within the sealing contact according to the Reynolds equation, the seal’s deformation, and the contact mechanics. The main drawback of these distributed parameter simulations is the necessity of a time-intensive resolution process. Given these experimental and computational constraints, machine learning algorithms offer a promising solution. Physics-informed machine learning (PIML) represents a noteworthy advancement in machine learning in tribology, extending traditional models with physics-based rules and enhancing accuracy in determining phenomena such as friction, wear, and lubrication. Within this field, physics-informed neural networks (PINN) emerge as a powerful class of hybrid solvers, combining data-driven and physics-based approaches to solve partial differential equations, the governing equations in EHL simulations. By integrating physical principles into the neural network’s parameter optimization, PINNs provide an accurate and accelerated solution. In this contribution, a PINN framework is applied to predict pressure build-up and cavitation in sealing contacts with housing. The capability of PINNs to determine transient and cavitation effects is thoroughly investigated and validated by the solution of the Reynolds equation obtained by the DDS. The results demonstrate the potential of PINNs for modeling tribological systems and highlight their significance in enhancing computational efficiency."

Gaining insight into tribological systems is crucial for optimizing efficiency and prolonging operational lifespans in technical systems. Experimental investigations are time-consuming and costly, especially for reciprocating seals in fluid power systems. Elastohydrodynamic lubrication (EHL) simulations offer an alternative but demand significant computational resources. Physics-informed neural networks (PINNs) provide a promising solution using physics-based approaches to solve partial differential equations. While PINNs have successfully modeled hydrodynamics with stationary cavitation, they have yet to address transient cavitation with dynamic geometry changes. This contribution applies a PINN framework to predict pressure build-up and transient cavitation in sealing contacts with dynamic geometry changes. The results demonstrate the potential of PINNs for modeling tribological systems and highlight their significance in enhancing computational efficiency.