Next Generation Software: How Simulation Process Modeling Enables Integrated Simulations of Complex Products
June 29, 2015:
How does it work to integrate simulations of complex products successfully – where components from different physical areas and thus diverse engineering disciplines are coupled to each other? If conventional simulation products reach a point at which no more improvement can be achieved? If electromagnetism, mechanics, acoustics – thermal effects, path-dependent dynamic effects and nonlinearities have to be considered?
The development of simulation software, and its application in product engineering, are typically features and functions driven, and more or less neglect strategic aspects that deal with methodology development and integration of simulations tasks into development processes.
When we think about the content of software training or discussions with software vendors, we note that they are naturally focused on the explanation of functions and features.
They illustrate the respective functions by selecting training examples of special applications. A dilemma, in our opinion, that comes to the surface. Mostly, they don’t pay much attention to the efficient use of the simulation tools. The potential of simulations is misjudged. The integrated use of simulation in all phases of product development is not exhausted. However, this is exactly what users should focus on in our opinion – a consideration throughout the entire product development cycle for an even more efficient engineering.
Konzept-X (Mvoid) is convinced that in carrying out complex engineering analysis not a specific software product nor the programming of a specific application provides the best solution. For Konzept-X (Mvoid) – there is no doubt – the best solution can be found by method development and by integration of various simulation applications within the entire development process. Strategic aspects should be in focus, while, of course, the simulation software products still provide the foundation. Significantly different is the way the product improvements are achieved: Suitable packages are to be selected to accompany the development process at all phases of the product development, that connect technically similar data closely to each other, as well as changes and identify weaknesses immediately.
In the eyes of Dr. A. Svobodnik, the optimization of the application and the interaction of appropriate simulation software tools represent the next generation of simulation software. It’s about analyzing the product development cycle in all its stages of development by means of integrated simulation methods continuously. Important is the intelligent use of simulation software tools from the onset. Each product line, each product is unique and has special features. They have to be analyzed by intelligent applications of simulation software – thinking for example of the various components with different materials and different placement in the various models, like SUV, Limousine or even convertible.
Difference between Simulation Software and Simulation Process Modeling
Thus, when we talk about the next generation of software we don’t refer to the replacement of the traditional engineering analysis software. It is about the intelligent use of CAE software packages across the entire product development process. The simulation process modeling is to be understood as a logical development.
Let us look back briefly: In the past 25 years of development of finite element methods, we find the following evolutionary stages:
In recent years we increasingly observed that company’s force simulation driven engineering and that they are working on appropriate solutions. For Konzept-X (Mvoid) the simulation process modeling is the key, the “door opener”, for the simulation driven engineering. It is far more than the next logical step in the development of FEM methods. It’s a groundbreaking development in Konzept-X (Mvoid) eye.
Let us explain in more detail: While the simulation software focuses on the modeling of the functional performance of components and, starting to analyze a component up to the whole product and thus follow a “bottom-up” approach, the simulation process modeling initially considers the major phases of the product development process and is using intelligent methods to develop appropriate (sub-) processes for each phase.
The product development process follows the “top-down” approach, as the simplified diagram below demonstrates.
Product development process in line with Western, Mike (2008) The Product Development Process
The simulation process modeling also follows the “top-down” approach. Already in the early stage, such as the concept development phase, it is possible to identify valuable insights. With the help of previously defined methods and processes and with the use of conventional simulation software it can be analyzed, whether ideas are realizable. For each major stage of the product development process a detailed simulation process is defined, which varies in the following categories/dimensions in modeling approach (Mvoid® Simulation Process Technology, Automotive Audio Version):
1. Dimension Detail– from Lumped-Parameter (sub-)models to 3D fully detailed FEA models
Lumped parameter models are based on one-dimensional, scalar equations to describe a physical area. For example, they describe the behavior of a speaker that interacts with an enclosure and the resulting sound radiation. For the mechanical and acoustic (airborne sound) area, however, there are substantial limitations. For the mechanical system, for example, only piston-like movements can be described. Thus, the applicable frequency range is very limited. At higher frequencies, the vibration behavior is no longer piston-shaped and a multidimensional modeling approach has to be used. In that case the finite element method is recommended. Today the FEA method is a widely used calculation method. Issues from various physical disciplines can be calculated.
In particular, in the very early stages of concept development mainly LPM (sub) models are used that are refined in parallel with the entire vehicle development permanently by FEA models.
2. Dimension Linearity – from linear to fully nonlinear models
Upon excitation of linear systems with a sine signal again a sinusoidal signal at the same frequency will be obtained at the output, but with a changed phase position and amplitude. Generally, nonlinear systems do not have these properties. Nonlinear systems can have on their system output frequency ranges that are not included in the input signal (distortion).
The closer we get to the SOP, the more a refinement is performed, similar to the dimension “Detail”.
3. Dimension Scale – from component to system level
In the automotive audio area the different dimensions are a particular challenge. The air gap between the coil and magnet in the drive system of the loudspeaker is in tenths of a millimeter range for example, while the main dimensions of the cabin – in the listening room – is in several thousand millimeters. We therefore need to map four dimensions in the model. A real challenge!
The above mentioned dimensions “Detail, Linearity and Scale” vary depending on the product requirements and they are used in combination in the respective phases of the product development process. They serve as a starting point for the development of the optimal simulation process. As mentioned above, conventional simulation applications are used. Due to the specific components respectively process phases, different simulation software packages have to be used. The challenge is to couple the different applications with each other and to convey all relevant information as well as to consider all physical disciplines. This is the only way an accurate simulation flow can be achieved throughout the entire product development process.
The business benefits of simulation process modeling
Getting efficient virtual development results of specific products or classes of products requires sufficient accurate simulations, no question about it. For the modeling of the simulation process a minimum of data on detail level and on the full level is necessary. It requires a so-called “critical mass” of parameters, so that such a process simulation model generates a benefit for the development. Consequently, the simulation modeling process requires initially additional time and work for the strategy development and consideration of the necessary parameters. The usage of potential process models is more limited compared to the availability of simulation software, at least in the near to medium-term time frame.
The additional time and work can be offset quickly, because the benefits of simulation process modeling are broad and clear. Typically, the benefits are quickly realizable and noticeable:
- Virtual systems can be used as test beds for innovations
- Virtual systems enable more freedom in design decisions
- Design changes can be evaluated without building prototypes
At first glance, the benefits do not appear new. We have seen similar benefits already in the first introduction of CAE-based simulations. However, these advantages are now not only limited to component level, but affect the development of whole products as a system. We therefore see an intensification of the following business benefits:
- Speed up development process
- Reduce development costs
- Minimize prototype expenses
- Improve product quality
Konzept-X (Mvoid) developed a simulation process model for the virtual development of automotive audio systems. The whole development process is virtualized by means of advanced CAE methods and other numerical schemes for all disciplines involved in the product development process:
- Industrial design
- Multiphysical engineering analysis
- Digital signal processing
- Tuning (Equalizing)
- Psychoacoustics and subjective evaluation
- Binaural Audio
Konzept-X (Mvoid) is convinced that the application of the above-described process simulation model in projects of the automotive industry provides the above mentioned benefits and can help to implement product enhancements immediately.
More information of the practical implementation of such a simulation process model is given in former technical papers of Dr. A. Svobodnik: 2013, Do we still need (new) Software?– On the way from Software Development to Process Development und 2013, Virtual Development of Audio Systems– Application of CAE Methods.