Advertisement

Lightweight Design worldwide

, Volume 11, Issue 5, pp 48–53 | Cite as

Strategic Development of Lightweight Platforms Made of Steel

  • Krister Önnermalm
  • Lars Fredriksson
  • Thomas Vorberg
  • Mirko Bromberger
Design
  • 127 Downloads

Successful lightweight steel design in large-scale automotive production requires the strategic use of optimization methods already in the early development phase. The Swedish automotive company NEVS relies on Altair's C123 method for simulation-driven and cross-vehicle lightweight design.

Steel Dominates Mass Production

While there is an ever-increasing trend to apply new materials and material mixes to car body designs in order to optimize vehicle weight, the automotive industry still mainly uses steel - especially when it comes to mass production. On the one hand, this is due to the material's excellent workability, and on the other hand, this is because of the numerous new developments in the area of high strength steels, which enable a thinner wall thickness and hence less weight while offering a constant stiffness. This is why the automotive industry expects steel to be the dominant material in the automotive sector also in the next 20 years. In order to reach the ambitious weight and safety goals car makers have set, it is time to find out how to efficiently use these high strength steels.

To successfully create lightweight concepts with steel and to use them in mass production, appropriate development methods are required. These methods have to be applied very early in the development process - focusing on the set weight goals for the body in white during the entire development process.

Virtual development methods, such as numerical optimization, are particularly appropriate to investigate development concepts and for defining suitable design directions early on. In addition to traditional FE simulations investigating the structural design, numerical optimization enables engineers to actively affect the equilibrium between development attributes and thus offers the possibility to identify and realize potentials for weight reduction. For many years, numerical optimization has been a well- established method in automotive development as this technology enables, among other things, a load-optimized structure applying as little material as possible. However, optimization is still being used as a tactical rather than strategic method, i.e. a component or assembly is only weight optimized with respect to the adjacent loads. By means of an example from NEVS, it will be demonstrated how to handle a strategic use of optimization technology, what kind of benefits this offers for the development process and hence for the final product. Here, NEVS applied the so-called methodical C123 approach offered by software and service provider Altair, Figure 1.

Figure 1 The C123 process for body optimization is divided into three different levels of abstraction (© Altair)

Optimization Methods

For many years, the automotive industry has been using simulation in various areas - within design, development, and test - to avoid unnecessary development cycles and to receive validated results about necessary design changes. This helps to detect faults early on and to find the ideal solution. Although a close collaboration between designers and engineers has become the standard today, development engineers are still facing the challenge to keep up with the current design status when creating a new simulation model. As they need a certain amount of time to create this model the design team is usually one step ahead of the simulation team. In this case, you cannot speak of a simulation-driven development process, the simulation rather only helps to validate in parallel the design changes performed. However, the ultimate goal is to establish a simulation-driven design process in order to benefit from all the advantages numerical simulation can offer to automotive development. Technically speaking, a development process can be called a simulation-driven development process if simulation and optimization are being applied as design tools and if optimization is specifically used to quickly include relevant knowledge from "what if" studies, sensitive analyses and Pareto charts in the decision process for vehicle design, Figure 2. Hence, simulation and optimization are no longer being used as mere validation tools but as design tools. To follow this approach, development engineers need the appropriate tools to gain valuable insights even before the design engineers start working on the model. Ideally, this results in taking full advantage of the simulation's potential for weight optimization already in the early concept phase when design changes are still possible and not too costly. Once the design phase has started many decisions have already been made and a major change in the overall design is either impossible or far too expensive.

Figure 2 Only by applying a simulation-driven development process does simulation become a design tool (© Altair)

Concept Design Process C123

To address this fundamental challenge, Altair has created the C123 development approach, which provides engineers with valuable insights early in the development, thus allowing them to decide for the most promising development direction at an early stage. The C123 method is a very flexible approach performing the optimization on the bases of three different levels of abstraction of the body in white, Figure 3. On the first abstraction level C1, the vehicle is built as a model representing the design space to define load paths. On the abstraction level C2, the model, containing special beam elements and coupled cross section information, is used to quickly perform optimization of variants. The abstraction level C3 is basically equivalent with the traditional method using shell models. In order to quickly simulate and optimize shell models, details which are not needed for fundamental design decisions are often neglected. The objective of simulation-driven development is to give support, to show the right direction, to evaluate design alternatives and to find the best possible balance between all development goals. For a detailed description of the abstraction levels, please see below.

Figure 3 Overview of the C123 process (© Altair)

Following this approach, various vehicle configurations sharing the same platform can be integrated into the numerical design process without the need for a detailed view of each variant.

In the first phase of load path determination (C1), Figure 4, fundamental issues such as the available design space and packaging for the desired structure are being discussed along with questions about which design spaces are required for which vehicle architecture and variant (coupé, van, limousine). In combination with the applied loads and the various load cases of the vehicle architectures, many different load paths are being created. The aim of the C1 phase is to determine the best load paths, employing at this point topology optimization, in this case, the optimization tool Altair OptiStruct, as the key discipline.

Figure 4 Coupling of design variables for equal load paths in shared design spaces. (© Altair | NEVS)

Once the preliminary load paths are determined, geometry requirements are added in the second phase (cross sections of door sills or A- or B-pillars) to pre-dimension load paths and apply data of real cross sections (C2), Figure 5. In the simulation, a wire model made of beam elements is used and the corresponding cross section properties are applied. The simple beam model can be used for a quick estimation if the cross section properties of a beam element have to be changed or if they are already sufficiently dimensioned. For this development step, a library with possible actual cross sections is available. For each cross section, a direct mathematical connection between the actual properties of a cross section (width, height and sheet thickness) and the properties for the description of the beam (geometrical moment of inertia and cross section area) is created and can therefore be directly optimized for the actual cross section properties. In addition, essential insight is gained about the node rigidity of the connection points which are considered when designing the details of the connection. In this phase, both optimization and Design-of-Experiment (DoE) methods are employed to provide the design with decision making and "trade-off" information.

Figure 5 Pre-dimensioning of load paths and applying data of real cross-sections (© Altair | NEVS)

In the third phase (C3), more details are added to the model to increase the maturity on the basis of simplified 3-D models while employing Multi-Disciplinary Optimization (MDO), Figure 6. By taking a close look at the connections in the nodes the connection method can be defined more precisely. This phase is of particular importance for the project's development and realization costs because decisions are being made on the manufacturing of the component (for example sheet metal design versus casting design). Also, existing manufacturing lines such as existing forming machines can be considered as boundary conditions in the decision process. In the C3 phase, all relevant information - such as load paths, actual cross sections, and structural nodes - is brought together. Subsequently, further studies can be derived from this data. It is not until this phase that the actual geometry of the chassis is created.

Figure 6 Increasing maturity on the basis of simplified 3-D models while employing Multi-Disciplinary Optimization (MDO) (© Altair)

In summary, this means that the process begins with creating design spaces for the different vehicle variants and that the vehicle models are already connected in the areas where identical components such as a front module or an underbody structure will be used. In a second step, the occurring load paths are used to create a beam-based structure which already includes all needed cross section properties. In the final step, all connection details are defined and further studies derived.

The C123 Process at NEVS

NEVS has inherited the legacy of Saab Automobile and has set itself the goal to offer an entire portfolio of fully-electric premium vehicles, mobility solutions, and sustainable urban city solutions. To this aim, the company relies on the simulation-driven development approach C123, Figure 7. The most important aspects for their development processes, especially to develop fully electric vehicles, are:

Figure 7 Design spaces of all NEVS vehicle types (SUV, car, station wagon), topology concepts and conveyed detailed models of the C2 and C3 phases (© NEVS)

  • the development of an ideal structural concept as a combination of vehicle platform, battery carrier, automotive body, and other load-carrying structures

  • the guidance and monitoring of load paths during the concept phase in order to successfully develop the vehicle architecture as well as a support concept to treat individual vehicle configurations (panoramic roof, motor configuration, vehicle dimensions etc.)

  • good monitoring and control options to develop an optimum balance between different attributes such as weight, costs, performance etc. and the structural feasibility studies

  • a maximum modularity between the single segments and vehicle types.

Since the development engineers at NEVS had to consider a wide spectrum of vehicle variants, they first chose and investigated a basis variant. This step was carried out jointly by NEVS and Altair. They decided on an SUV for the global market as a starting point on which to base the platform.

At first, they applied the C123 process to develop the basic concept. Since all vehicles of the company are operated fully electrically, the battery carrier was especially important for NEVS. To ensure that all vehicle variations and platforms can be considered simultaneously they performed so-called multi-model optimizations allowing the engineers to consider several questions at the same time.

In one of their studies, they wanted to assess which consequences a certain variant of the battery carrier would have for the performance of other vehicle structures. Due to the effect on the different vehicle structures, this process was very dynamic. Especially during the crash simulation, it became apparent that the battery carrier could also be used as an additional load path, with the result that the load distribution between the different load paths could be better resolved. Subsequently, the connections were investigated to find the ideal connection for the battery carrier and to reach the best possible performance.

Basis for Development Decisions

Using the C123 development approach it was possible to actively support the concept development at NEVS with simulation and optimization. By systematically employing optimization technology the engineers could highlight design alternatives, determine design sensitive factors, solve trade-offs, and run what-if scenarios to identify the optimal design balance between partly competing attributes such as performance, weight, and costs. In addition, NEVS obtained significantly more information providing a basis for important design decisions, especially in the early concept phase. Implementing the C123 process the development team was able to rely with confidence on the created concept and the well-balanced design.

Particularly, the design teams benefitted from the new approach since now they are able to quickly assess how design changes will affect other components of the vehicle. This resulted in an unprecedented degree of concept maturity and convinced the NEVS developers to employ the C123 process in future projects as a standard procedure.

In contrast to using simulation to validate single components, which is still the main area of application in many companies, the C123 process offers a real simulation-driven and cross-vehicle design approach. Using such a method is the only way to identify and follow-up the most promising development directions for a lightweight design concept very early in the development process of a vehicle. This is particularly the case when developing a platform that should use as many synergies as possible across vehicle variants. As a result, this process leads to lightweight concepts which are also optimized regarding costs and performance. |

Copyright information

© Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2018

Authors and Affiliations

  • Krister Önnermalm
    • 1
  • Lars Fredriksson
  • Thomas Vorberg
  • Mirko Bromberger
  1. 1.National Electric Vehicle SwedenTrollhättanSchweden

Personalised recommendations