A seemingly minor investment announcement might have big implications for additive manufacturing.
The news is that a London-based startup, PhysicsX, is now “out of stealth mode”, and raised a notable US$32M in an investment round.
What does PhysicsX do? They are a software company that produces AI-powered simulations of complex mechanical scenarios.
Simulation has long been present in the industry. It’s a way to explore the performance of a part in a virtual way before committing to actually building the part. Sometimes this step is called optimization, because iteratively generated simulations can gradually lead towards the optimal design.
PhysicsX has already appeared in the 3D print world, as they recently partnered with Velo3D to solve the issue of smoke flow produced during LPBF metal 3D printing.
Simulation and optimization are of great interest in the 3D print world because the highly complex designs that inevitably result are almost always only producible on 3D printers. 3D printing’s additive approach means that highly complex geometries are simple to make, unlike traditional subtractive technologies.
That’s why many 3D printer manufacturers have an affinity for generative and simulation software: it provides a way to sell more equipment to manufacturers seeking complex parts made by that software.
However, there’s always been a huge problem: simulation is highly resource intensive. In other words, you require a whopper of a computer to run the simulations, some of which are basically plotting the movement of particles within the scenario, second by second.
Simple simulations can be done in minutes or hours, but for complex simulations it may take days or even weeks to complete. Because of the extraordinary long durations, the number of iterations tends to be low. That means the optimal solutions may not actually be achieved in all cases.
Enter PhysicsX. Their AI-powered system is able to perform simulations in a very different manner that is far, far faster. They say that their platform can execute simulations incredibly fast.
In TechCrunch, PhysicsX Co-founder Jacomo Corbo explained:
“We are building an enterprise platform to support a pretty broad range of domain applications that are tied to building and optimization problems, physics simulation bottlenecks. What PhysicsX buys you is the ability to be able to predict the physics (of a system) with very, very high accuracy and fidelity, doing it, anywhere from 10,000 to a million times faster. Now we can be a whole lot more sophisticated about, for example, mining, across a very high dimensional space.”
Hold on, did he say “a million times faster”?
He did.
A simulation that previously took five days to complete might now be done in as short as only HALF A SECOND!
To put it another way, during a single work week, a million variations of a part design could be evaluated.
The implications are incredible: a designer should be able to produce a truly optimal complex part design in far, far less time. This will almost certainly lead to an explosion of highly complex part designs, all of which will require 3D printers to produce them.
This is incredibly good news for the 3D print industry.
Via PhysicsX and TechCrunch
منبع: https://www.fabbaloo.com/news/accelerating-additive-manufacturing-physicsxs-ai-simulations-promise-a-design-revolution