Artificial Intelligence Is Pushing Boundaries in Design Engineering

Artificial Intelligence Is Pushing Boundaries in Design Engineering

Today’s design engineers are always looking for ways to innovate and compete in today’s global design environment. The 21st century has opened up a plethora of connected workflows that allow for seemingly real-time data interaction, no matter where you are in the world. The days of the central data center and being tied to the office or factory are over. Powerful mobile devices are in our pockets, with us at our bedside tables and more cost effective than ever.  We are at an inflection point; forward thinking engineers will have the advantage if they implement technology from the edge, AI and the IIoT worlds.

3D Design Platforms Reduce Design Time

Have you ever wanted to design a custom component for a product you don’t have the 3D CAD data for? If so, you are most likely familiar with the almost never-ending process of acquiring an expensive 3D scanner, getting a point cloud and manipulating it in high-priced specialty software. Next steps include converting that point cloud into a mesh, taking that into a mechanical CAD system to use finally. 

With LiDAR capabilities now built into many of these devices on the edge, 3D design platforms like Onshape, a PTC technology, can drastically reduce the amount of time and effort it takes to get useful dimensional information about an object without even sitting in front of a traditional CAD workstation. Simply point your iPhone for example at your target, walk around it and hit go, and you will then immediately have a 3D Mesh object instantly available to share with your fellow colleagues, customers, and suppliers. These consumers of the information can use whatever plane of glass you bring to the party; any modern device that connects to the internet, like a Chromebook, Mac, Windows PC, Android, or iOS device are options now for product designers who want to get their work done wherever they are.

Image courtesy of OnshapePicture3.jpg

AI Determine’s a Machine’s Performance

When products fail in the field, they can cost a company in downtime or lower productivity.  Imagine having unplanned downtime on a piece of machinery that labels and caps a can of your favorite product. If a subsystem of that machine is performing poorly and then breaks, there is nothing to do but to shut the system down and make repairs.

Machine designers are now using myriad sensors and internet connected devices to determine a machine’s performance. Machines might even have more lines of software code than your favorite high-end video game console. These new devices can push and pull information. Software can read output that engineers find important, and over time, AI-powered industrial systems recognize and can act upon multiple performance variables. This prevents unplanned downtime, saves money, and increases speed to market.

AI can also be used to provide design feedback and help engineers with design alternatives that they may not have realized possible. With the input of product conditions like the material choices, manufacturing processes and geometry, along with the loads and restraints that an assembly may encounter, powerful AI tools hosted on the PTC’s ATLAS platform for always-on Simulation tools and Generative Design, which uses AI and High Performance computing techniques to present design alternatives to engineers to make better design choices.  These choices are presented in such a way where an engineer can select from a range of design iterations the AI based solvers came up with. This enables solutions to issues that may not have been considered. 

This process also has familiarity if you are an agile product design advocate. Agile design and iteration are all about failing fast and early in the design process to advance the innovation of your product in short spurts, versus marathon style stage gate processes. If AI is there doubling or tripling your design capacity for improving designs, AI can provide improved time to market plus improve the function of your design.

IioT and the Digital Twin

We all know how orbiting satellites have improved all our lives with GPS, weather forecasting and many more fields. This technology makes all the mobile devices that we use every day much more useful, as information from satellites drive apps that give us driving directions, direct us to the nearest coffee shop and more. Similar analogies can be made to the various devices that surround us in our daily lives at home with smart devices like speakers, light bulbs, cameras, etc. But this is also happening every day in the manufacturing space. Sensors are available to add to your products and machinery that can track anything that will help with performance, maintenance, and repair. All these devices could be considered edge appliances or devices.  And these devices all send and receive data.

Understanding this data requires software that can understand this torrent of information as well as conveying this data in human readable format. Modern cloud-native apps like Onshape employ an open API or Application Programming Interface that can connect to other business systems and services, or even devices. If you know a little Python, you can connect these cloud-native systems easily and securely. For example: you can use python to analyze complex, 3D systems and run simulations of a robotic arm’s motion in the context of the rest of a machine.

These connections to digital and physical data are formed on the Digital Twin. This twin is a virtual representation of a physical product, process, person, or place that can understand and predict its physical counterparts. A digital twin has three components: 

  1. A digital definition of its counterpart (typically a 3D CAD or metadata, PLM-based representation of the design.
  2. Operational/experiential data of its counterpart (gathered from IIoT device data of the real-world telemetry)
  3. An information model displayed in dashboards and reports that correlate and present the data to drive decision making.

With this digital twin representation, you can start connecting applications like Onshape to your edge devices and create a real-time display of data in Onshape to monitor the physical device, based on sensor information from the physical object as well as drive the machine based on inputs in the digital model in Onshape. Once your edge device is connected to Onshape, you can use the model as your interface for controlling the movement and logic of your physical machine.

 

The future is here. With Edge Computing, AI and IIoT in the mainstream, and with the power of modern SaaS infrastructure with public cloud solutions, it is now easier than ever to push the boundaries of what is possible in your engineering design projects.

Source link

Share This
COMMENTS

Leave a Reply

Your email address will not be published.