Wireless power transfer (WPT) is much researched and discussed in the context of IoT, electric vehicles and mobile electronic devices. The methodology of powering a device without a physical connection is well known. However, designing the coil shapes and their placement, maximizing efficiency and validating behavior at the system level still represent challenges that cannot be achieved without simulation. The next frontier to be explored is extending and applying wireless power transfer systems to more applications, such as continuous charging of multiple devices, increasing the range of efficient power transfer and ensuring the WPT system design meets regulatory guidelines. Continue reading
Recently I was interviewed for a Desktop Engineering article about how designers can expand their FEA skills. My first thoughts were the obvious — you could attend a live training or a web-based course, or use reference books. But I also thought about how I would gain more understanding of how a given model behaves. I may want to know the influence of the mesh on the results or how the material could affect the behavior of the design. So, during that interview, I mentioned parametric analysis.
Every time I discuss this topic with simulation users, it seems to resonate pretty well. However, very few effectively perform parametric analyses in an automated manner. But everyone agrees that for a given design, they will need to compute variations, perhaps because the initial design fails or because after meeting initial requirements they want to optimize a design for cost or weight reduction.
Whether you choose to perform parametric simulation with tools that are integrated in your simulation platform, such as ANSYS DesignXplorer, or with external optimization tools, all it usually takes only a few mouse clicks to convert a “one shot” simulation model into a parametric model. After that, variations are just a matter of changing the parameters’ values and rerunning the model. Of course, this requires that your simulation tools allow for parameterization of geometric changes, material properties, loads and boundary conditions, or results. But it will lead to a much better understanding of your product, as you will not only have the results for a nominal design but you will be able to answer the “what-if” questions that your colleagues will surely ask you.
So, what prevents you from going parametric?
For more information on parametric simulation, look at these contributions from my colleagues: