How Conceptual Optimizations Improve Your Designs
Complex multiphysics optimizations are more practical using more computational power. But much of this is more useful later in the design cycle. Keep concept design simple.
When performing simulation in the design phase of a development cycle, engineers care a lot about accurate numbers and predictions of performance. However, there is a fundamental difference in the concept stage of a product’s development. Here, engineers should aim for high-level concepts and worry about the finer details later. But how can a high-level simulation be useful to a product’s design?
“Concept design values should not be taken at face value,” warned Fatma Koçer , director of business development at Altair. “This is a common error. Systems level concept design looks at trends, not the values. Validation simulations are done to determine accurate representations as they are more detailed to assess design performance. Here you want accurate values.”
During the early design phase, engineers should use equations and simple models to better understand the physics of the system and how the parts and design variables interact. As a result, optimizations at the concept design stage will look to find global performance, trends and trade-offs.
When looking at the system optimizations, it is important to remember various disciplines, from manufacturing to structural performance. By connecting these system-level simulations, engineers can determine quick trade-offs and pick a few configurations to look at more in-depth.
Inspire has a graphical user interface that is targeted to be user friendly for design engineers. This is beneficial for topological concept design.
“If you look at topology results to meet the weight reduction targets of a seat or bracket, the goal is to use material most efficiently along the load paths,” said Warren Dias, business development manager at Altair. “If you add material where there is no load path, you are wasting material. If there is a load path without material, you are risking your design’s structural integrity.”
Altair offers a few tools for looking into these early concept designs.
How Altair Can Bring Conceptual Design Into Your Workflow
Bringing concept design, like Inspire’s topology optimization, into the design cycle can produce more innovative designs.
Inspire offers a simplified user interface and stripped down capabilities of OptiStruct to target design engineers. This is optimal for conceptual design as it helps to keep the analysis simple.
OptiStruct, on the other hand, is more for analysts. It looks into more complex designs with load cases too difficult to be incorporated into Inspire. As a result, it can be used in conceptual design and more complex part assessments later in the design cycle.
HyperStudy allows for the exploration of a variety of design changes with its DOE and optimization capabilities.
After the initial topology assessments are done, design fine tuning for multiphysics can be performed using HyperStudy.
HyperStudy is a multi-disciplinary design tool for optimization and design of experiments (DOE). As a result, it is used to assess the system-level trade-offs.
Concept design points engineers into a certain direction. Using simple simulations, engineers are able to look into trade-offs and understand how their design variables affect the system’s performance. Engineers will need to make assumptions about the system to speed up the simulations and account for parts that are not finalized. These assumptions will typically be 0D and 1D simulations.
"With HyperStudy you can perform what-if studies and set up targets for multi-disciplinary applications," said Koçer. "The key to this process is that you move from a typical trial and error optimization to a more informed mindset."
There are some common mistakes when performing conceptual design studies, noted Koçer. For instance, she said, “Concept designs might have local stress or buckling, but this isn’t always what we want to focus on at this stage. They are local issues that can be solved late in more detailed simulations. Many users get stuck at this point and miss the point of looking at the trends.”
Your Design Workflow After Conceptual Design
Moving from single-run analysis to multi-run analysis with multi-disciplinary design will help to better optimize the design.
“People expect optimization is a run once and done, but you can’t just optimize at one stage and hope for an optimal design,” said Dias. “It is recommended to optimize throughout the design process.”
Users can pick a few variables or concepts at a time and look into the design space. Dias said, “You will gain more information into your system and move to final assessments from there.”
Once a concept has been created within Inspire or OptiStruct, the next step is to move into in-depth part designs. Here you will start to interpret your system-level and topology optimization at the part level. It is important to include various experts that will be involved in the creation of the product at this stage.
For instance, the results from Inspire may not be feasible to manufacture. Inspire and OptiStruct do allow for designers to add manufacturing constraints to the algorithm. However, this will only make a topology that is close to being manufacturable. As a result, the CAD design team and manufacturing team will have important inputs to develop the design.
Next the team will need to validate the product and parts using CAE tools like HyperWorks to perform FEA, CFD and multiphysics simulations. “As computational power has increased,” said Koçer, “these more complex multiphysics studies are now more practical.”
“It is easy to move from Inspire to OptiStruct when performing more complex design assessments,” noted Dias. “The output files are of the same framework. Once the analysts assess the stress and frequency of the design then the manufacturing team can start producing a prototype for physical testing.
Optimization Challenges? Call a CAE Expert.
Altair’s Optimization Centers offer users the chance to collaborate with optimization experts.
“When performing your interpretation step you might be conservative and increase the material and weight of the product,” explained Koçer. “But in the design stage you can start to correct this and reduce the weight again by fine tuning the product as you learn more. And the optimal design is always evolving as you learn more.”
Users should understand that keeping track of all of these changes and all of the programs used in the product development can be a challenge. Without product lifecycle management (PLM) tools and communication, various team members might end up reinventing the wheel, or missing vital information needed for their step in the design cycle.
Another challenge that Koçer mentioned was to set limits for automated optimization tools. She said, “When looking at design optimization, look at what can change in the design and by how much. Next ensure that your performance is moving in the direction you intend it to. Then iterate as needed.”
It is important to note that optimization and design exploration technologies may not have the ability to find the small changes in many simulations that are performed. Koçer’s answer to this problem is to focus on what is important and critical to the design.
“There are tricks you can learn to be more effective as you gain more optimization experience,” said Koçer. “For instance, complex designs are hard to work with, so make your designs as simple as possible. But if you are lost and need assistance, Altair offers various optimization centers where users can contact and collaborate with optimization experts.”
To learn more about Altair’s optimization and concept design capabilities, follow this link.
Altair has sponsored this post. They have no editorial input to this post - all opinions are mine. Shawn Wasserman