Innovations Product Design
Discover our latest innovations on product design
The demand for customised production is increasing dramatically and bringing a single product to market is no longer enough. Using model-based design methods and supporting software, we help developers improve the increasingly complex design process. We use Artificial Intelligence, digital twins, but also tools that check the manufacturability of CAD designs, for example.
Other themes:
50% faster mechanical redesign using high-dynamics digital twin
In product design, we identified an industrial need to simplify the operator-designer interaction. We do this by bridging the two in the form of a central digital twin. By using a flexible multi-body model as the digital twin, we can establish an effective and integrated framework that serves the entire value chain. From identification to design to operational monitoring. With this central digital twin, we achieve up to 50% faster redesigns and benefit from real-time status and input estimates. The central digital twin streamlines the entire development process, reduces modelling and identification efforts and thus leads to better mechatronic products.
Optimising the early design phase of an electric vehicle's powertrain
The design and development cycle of mechatronic systems is becoming shorter and shorter. As a result, companies are finding it increasingly difficult to design high-performance systems in the early design stages, where requirements and component choices are still uncertain.
To help designers with this, we have developed a design workflow that helps them to quickly identify the most promising powertrain designs for an electric vehicle (EV). The workflow does this by understanding and optimising the relationship between different parameters and powertrain performance. This results in a shorter and optimised development process and leads to energy and cost savings for the end-user of the vehicle.
Manufacturability check for CAD designs
Every product development process faces constraints around product geometry, material selection and the machine used. To speed up these processes, we have developed a method that collects in-house manufacturing expertise and applies it to the CAD design process.
This new method allows for an easy evaluation of the manufacturability of a CAD design, based on the collected manufacturing knowledge. For example, our tool can evaluate a housing component of a beamer. The tool assists the CAD designer in assessing the limits for different manufacturability aspects in a simple workflow. Based on the feedback from the tool, the CAD designer can make several design iterations leading to a final design that no longer contains manufacturing errors. By avoiding design errors in the design phase, a lot of costs can be saved. In addition, with our tool you reduce the complete development time for a complete CAD design process.
Hands-on topology optimisation
Topology optimisation is a technique that allows an industrial designer to use the available design space as effectively as possible. This often involves finding the stiffest component for a desired weight or, conversely, the lightest component for a desired stiffness. Unfortunately, conventional topology optimisation methods are rather expensive and unsuitable for thin-walled structures that we often find in contemporary designs.
To make these topology optimisation methods more accessible, we have developed a framework that can deploy thin-walled parts by using shell elements instead of solids. This leads to a faster design that is also easier to produce with conventional production methods. With our approach, the designer starts with an arbitrary 3D geometry and can find the best distribution of the thin-walled material from there.
Designing products for optimal performance and assembly
In a traditional product design process, companies usually focus first on achieving all product specifications. Only then are production and assembly aspects taken into account. This often leads to the design being sent back and forth between the production and design teams, which costs a lot of time and effort. Ideally, you should be able to load assembly knowledge into the design process. This way, you can take into account both the limitations of the assembly process in the design and the impact of the design on the assembly process.
To solve this problem, we have developed an automatic modelling tool with which a designer can optimise a product. We take into account both product performance and assembly complexity. With this tool, the designer gets instant feedback during the design process when a design is difficult, or impossible, to assemble. Moreover, new and sometimes unconventional designs emerge that form an ideal compromise between product performance and manufacturability.
Speeding up controller commissioning using AI methods
Production efficiency and product quality are important in any manufacturing environment. To achieve this, a commissioning process is often used to fine-tune production machines. However, this process is lengthy and can only be carried out by experienced operators, especially for machines that perform multiple tasks or work in variable conditions.
To speed up this process and reduce the dependency on experienced personnel, we have developed an AI-based method that provides a good initial estimate of the correct control parameters. We do this by using information from similar machines and tasks that have already been tuned. We applied this method to three rod mechanisms, each performing a different task. Comparing our method with a (model based) benchmark method we can see that the time needed for tuning the machines decreases by about 35%. At the same time, performance improves by about 70%. Yet we use only 20% of the available data to train our AI method. This is ideal for tuning an entire fleet of machines with a wide range of tasks.