Development of a Multi-Phase Tank Model for the One-Dimensional Hydraulic Simulation

  Metamodel for the Integration of Tank Simulation into 1D Models

Undissolved air in the oil causes various problems for hydraulic systems. Therefore, the hydraulic tank as a component that performs the task of separating air from the system is being investigated in more detail.In recent years, the behaviour of the hydraulic tank with respect to air separation has been investigated experimentally as well as with CFD simulations.In this project, a metamodel is being created which, depending on identified influencing factors, depicts the air portion at the outlet of a hydraulic tank in a much more time-efficient manner compared to CFD simulations and experiments.



Determination of the air content at the outlet of a hydraulic tank without high costs and special knowledge Quantification of the influencing factors and selection of an experimental design
Selection of a suitable tank for hydraulic systems with regard to the air content at the outlet Performance of CFD simulations and creation of a metamodel
Simple and fast application of the multi-phase tank model in industry due to the lower computing capacities and times compared to CFD simulation Validation and optimization of the model
Preparation of the metamodel for software transfer



+49 241 80 47735



CFD Solver, Test Bench and Model Concept

CFD of an Oil Tank CFD Tank Model

CFD Model

For this project a solver in OpenFOAM® was developed, which is based on the Euler-Lagrangian approach. The multiphase flow in a hydraulic tank is assumed to be a dispersed two-phase flow consisting of fluid and particles. The fluid represents the continuous liquid carrier phase (oil) and the particles represent the dispersed gas phase (air bubbles).

The simulation is based on a four-way coupling, i.e. the influence of the flow on air bubbles and vice versa as well as bubble interactions (collisions and coalescence) are considered.

The solver uses a coupled Euler-Lagrange method to simulate the dynamic change of oil and air volume flows at the inlet. With the help of the simulation, the air share at the outlet can be dynamically calculated as a function of the system parameters.

  Tank Test Bench Tank Test Bench

Tank Test Bench

With the tank test bench, the air content at the outlet of a hydraulic tank can be measured and then compared with the results of the CFD simulation. With a shadow recording system the oil flow is photographed in a bypass after leaving the test tank. The resulting images are post-processed to calculate the air content and bubble diameter distribution.



The direct use of complex CFD models is only possible to a limited extent for the analysis of the tank due to long computing times and the required knowledge. With a so-called meta model, the target value (here air share at the outlet of the tank) can be predicted with sufficient accuracy in the range of milliseconds. Such a metamodel is being developed within the context of this research project.

This model can be based on a variety of complex systems of equations. The investigated influencing factors are the input variables of the model. A multi-dimensional structure is created from the adjustment range of these variables, which is called factor space. Each point in the factor space is a combination of the numerical values of the factors. The selected combinations from the factor space form the experimental field.

The DoE method is used to select a more efficient experimental field. The aim is to select a trial field that provides a maximum amount of information with a minimum number of trials. This provides the complex relationships between the factors and the target value. The metamodel is obtained from the data determined on the basis of a test field.

  Flow Diagram Procedure for Modelling Modeling

Software Transfer

The multi-phase metamodel of the tank should be able to be implemented as an independent component module in 1D simulation software (for example AMESim or DSHplus). For this purpose, an FMU model is provided from the multi-phase tank model with the help of FMI, which can be used by any 1D simulation program.

The accuracy of the formed meta model is assessed by the validation calculations and if necessary model optimizations are performed with the existing methods. Finally, the influence of the factors on the target variable can be determined with a sensitivity analysis. The developed metamodel is the multi-phase tank model.



The IGF project 19612 N / 1 of the Forschungsvereinigung Stahlanwendung e. V. -FOSTA, Sohnstraße 65, 40237 Düsseldorf was funded via the AiF as part of the programme for the promotion of joint industrial research (IGF) by the Federal Ministry of Economics and Energy on the basis of a resolution of the German Bundestag.

Simulations were carried out with computing resources granted by RWTH Aachen University in project rwth0344.

The ifas would like to thank you for the funding and support.


*** AIF Logo ***

Copyright: AIF  


Mostafavi R., Tiffin D., Schmitz K.: "Determination of The Dynamic Characteristics of a Hydraulic Reservoir for Its Air Release Efficiency Using Multiphase CFD Model“, BATH/ASME Symposium on Fluid Power and Motion Control (FPMC2018), 2018, Bath, UK