Figure 1. Simulation of a capacitive touchscreen. The user can control the position and orientation of a finger pressing the touchscreen, and then compute the capacitance matrix. (Simulation example made using COMSOL Multiphysics® software; image courtesy of COMSOL)

As new concepts need to be investigated, and a wide range of operating conditions has to be considered, the design of new products and processes must push the limits of technology. Multiphysics simulation software can boost product design in many ways, and help bring innovative products to market while optimizing the use of resources throughout organizations.

Multiphysics simulation allows for a reduced need for testing and physical prototyping, as high-fidelity modeling allows you to take into account any physical interactions as they happen in the real world. Additionally, multiphysics simulation enables collaboration between simulation specialists and experts in product design, as well as other departments, through apps that bring the predictive power of multiphysics into the hands of simulation experts and product designers alike.

Less Physical Prototyping

One of the key advantages that simulation offers is the ability to explore designs and reduce physical prototyping. Challenging designs and new ideas can be built and tested in the virtual world of numerical simulation without having to be physically constructed or tested repeatedly. In every industry, safety is of paramount importance; the capability to investigate different scenarios by specifying boundary conditions, material properties, and working mechanisms allows for early and harmless correction of design mistakes. This is an important benefit for industries such as medical, space exploration, buildings, emergency management, and products operating at very large or small scales.

Figure 2. Vibration and noise analysis in a 5-speed synchromesh gearbox. A transient multibody analysis is performed to compute the gearbox vibration. An acoustics analysis is then performed to compute the sound pressure levels in the near, far, and exterior fields. (Simulation example made using COMSOL Multiphysics® software; image courtesy of COMSOL)

Tests are also expensive, so the opportunity to reduce — sometimes by an order of magnitude — the number of prototypes created or tests run leads to significant cost savings. Additionally, simulation measurements of any variable, such as temperature or flow velocity, can be taken at any point in a model, enabling a better interpolation of available experimental data. The fact that simulation results can be accessed in locations where it would be impractical if not impossible to place sensors on a physical prototype is greatly appreciated, as physical testing cannot reveal information in those cases.

There are even scenarios in which further testing could possibly be eliminated with a validated model. If a virtual prototype represents reality with high accuracy, then the effect of very small design changes can be addressed by numerical simulation. Of course, these design changes need to be within the parameter ranges for which the virtual prototype has been initially validated. Sometimes testing cannot be performed at all. This could be due to very harsh environmental conditions where testing would be impossible, or for small devices where testing is difficult, too costly, too dangerous, or impractical as it is in the case of a touchscreen (Figure 1), where thousands of use cases need to be considered, from the size of the stylus used, its position, and material, to the cleanliness of the surface and external interferences. Design challenges can be surpassed with ease when thousands of ideas and use cases can be tested and measured virtually, even before the first physical prototype is built.

More Accuracy with a Multiphysics Approach

Figure 3. Multiphysics simulation of tumor ablation. The localized heating of malignant tissue is achieved through the insertion of a four-armed electric probe. This model couples the bioheat and electric field equations, and models the temperature field in the tissue. (Simulation example made using COMSOL Multiphysics® software; image courtesy of COMSOL)

The large majority of products are inherently multiphysics, in which multiple physical effects influence each other (Figure 2). Hence, they have to be designed in that context to function properly and effectively. A multiphysics approach to simulation allows for high-fidelity modeling of products and the systems they operate in serving as an accurate predictor of a design’s performance by including all of the physical phenomena involved, and being able to describe the interactions as they happen in the real world. A practical example would be the tissue ablation problem shown in Figure 3.

In this case, multiphysics modeling work shows how precise control of the shape of tissue ablation zones is achieved, regardless of the target location or tissue type. To successfully design such an ablation system, the engineers had to take into account different electrical conductivities of tissues, optimize the delivery of electromagnetic energy within the radio frequency and microwave range, and include perfusion to predict the temperature distribution inside and around the tissue being ablated. The level of accuracy delivered by adopting a multiphysics approach to modeling and simulation has become the norm for many specialists who rely on multiphysics software.

COMSOL Multiphysics® software, for example, receives user input and generates a mathematical model, consisting of differential equations, to describe the physics phenomena of interest to the simulation specialist. Any CAE software today is based on predefined numerical models, which are approximations of differential equations. These approximations are necessary as, in most cases, the relevant differential equations cannot be solved analytically; that is, an exact solution cannot be determined. Instead, different types of discretization — such as finite differences, finite volumes, and finite elements methods, among others — are used to approximate the relevant differential equations. It is difficult to add phenomena and descriptions of variables and multi-physics couplings to a numerical model if they are not considered in the differential equations from the beginning.

With this software, a full mathematical model is generated on the fly, based on the user input, before the discretization is created when the user runs the simulation. This core technology allows simulation specialists to create their own expressions and multiphysics couplings by directly typing the mathematical expressions in the user interface. If that is not the case in a CAE software, then descriptions that are not built in must be done at the numerical level and after the discretization happened using user-defined subroutines, which may be inaccurate and/or difficult to produce.