Log in

OnScale Solve™ – Parametric studies

By Gergely Simon 24 December 2020

Having the ability to carry out parametric sweeps for finite element simulations is essential to aid design workflows and test tolerances. Dependence of variables of interest, such as stresses or displacements on input variables like temperature, forces or pressure can aid the designer in locating weaknesses in the design and making improvements. Once a design reaches the final stages of the planning cycle, it is important to carry out similar parametric sweeps to see sensitivities of that particular design to changes in operating environment.

In this blog post the main aim is to showcase the importance of these types of simulations and to illustrate how straightforward these analyses are using OnScale Solve. The example we chose is a pump house, depending on application, this pump house can circulate fluids of various temperatures (in harsh conditions even sub-zero, or when used for machinery coolant up to 100s of °C), and can be located underwater (or any other liquid), experiencing pressure on its outer surface. The simulation is therefore a coupled thermomechanical one.

A standard pump house CAD model has been obtained from a public OnShape example. It has a single part, and for simplicity, this was assigned Titanium from our material database. To set up boundary conditions: the bottom of the model is fixed (restraint), a pressure is applied on the larger outer shell surfaces, temperature applied to the inner surface of the smaller inlet ring, and a heat convection on the same surfaces as the pressure (see Fig. 1).

Parametric Studies

Figure 1 Boundary conditions applied to the pump house model

For the temperature boundary, the typical range of negative tens of °C to positive hundreds of °C suggests a linear parameter to be used, as shown in Fig. 2.

Parametric Sweep

Figure 2 Linear parametric sweep setup for temperature

However, pressure can vary in a much wider range on the housing depending on environment, so here a geometric sweep seems more adequate (Fig. 3).

Geometric Sweep

Figure 3 Geometric sweep setup for pressure load

Afterwards, the usual meshing and estimate steps can be carried out, followed by execution of the model.

Temperature and von Mises stress results for the four extreme cases (-60°C, 100 Pa; -60°C, 100 kPa; 160°C, 100 Pa; 160°C, 100 kPa) can be seen below. These can be easily obtained once the Results tab is available. The user can click through all simulation results, check input variables corresponding those and display results.

Figure 4 Temperature plots for extreme simulation cases (left column -60°C, right column 160°C; bottom row 100 Pa, top row 100 kPa pressure)

The temperature plots do not change significant variation with external applied pressure, both for the lower and higher extreme temperatures. The temperature distributions are as expected: gradually increase/decrease from the temperature boundary towards the back side of the model, where they approach the ambient 20°C.

Figure 5 Von Mises stress plots for extreme simulation cases (left column -60°C, right column 160°C; bottom row 100 Pa, top row 100 kPa pressure)

The effect of temperature and pressure on stress is equally interesting. The pressure has small effect on the stress in general, and it mainly comes from thermal expansion of the structure. Nevertheless, in all cases it concentrates towards the temperature boundary, which is again expected. The magnitude of the stress is larger for the 160°C case, as this value is further away from the ambient 20°C than the -60°C. Finally, the external pressure increased stress for the lower temperature case: the model shrinks with reduced temperature and the external pressure adds to this. The reverse happens for the larger temperature: the external pressure lowers stress as the compression of the model acts against the thermal expansion.

Parametric sweeps offer a great tool to investigate designs and analyze how they behave and perform at various operating points or how reliable their design is. We are excited to offer an easy to use, lightweight and fast solution: OnScale Solve.

Getting Started with OnScale Solve

Engineers, designers, analysts, and current OnScale users can learn more about OnScale Solve and run their first cloud engineering simulation study by accessing these resources:

With OnScale Solve you can work, collaborate, and share from any location and device. Ready to go? Create your free account today and start simulating with OnScale Solve! If you have any questions please contact us at info@onscale.com!


Gergely Simon
Gergely Simon

Gergely Simon is an Application Engineer at OnScale. He received his PhD in Smart Systems Integration from Heriot-Watt University. As part of our engineering team Gergely assists with developing applications, improving our existing software and providing technical support to our customers.