May 10, 2017
What is a Digital Twin?
A digital twin is the combination of all the organization’s digital information on a specific product with operating data streaming live from the product as it is being used in operation. Merging physics-based understanding with analytics delivers the insights that unlock the true value of the digital Twin.
Using these insights, engineers can understand the operational failure modes of the product, prevent unplanned downtime, improve product performance and seed the next product generation. The digital twin is able to leverage multifidelity simulations from detailed 3-D physics to reducedorder models (ROMs) to compress simulation times and demonstrate key product performance aspects.
For example, a digital twin of a gas turbine installed in a power plant might be designed to highlight energy efficiency, emissions, turbine blade wear or other factors of particular importance to the customer and thus the product development team.
By studying the digital twin, engineers can determine the root cause of any performance problems, improve output, schedule predictive maintenance, evaluate different control strategies and otherwise work to optimize product performance — and minimize operating expenses — in near real time. This is becoming increasingly important as customers shift from buying a product to buying an outcome, with
the performance risk passed to the product developer.
Simulation has long been an integral part of the product development process; it greatly improves product performance, reduces development costs and gets the product to market much more quickly. The technologies that underpin the Internet of Things now make it possible to go a step further by integrating simulation with products as they exist and operate in the real world. This opens up a whole new era in value creation for companies to optimize operations and maintenance, as well as further accelerate the new-product development process. ANSYS worked with PTC, Flowserve, National Instruments and HPE to demonstrate this, showing how a simulation model of an operating pump can diagnose and solve operating problems faster than was ever possible before.