Digital twin innovation cuts energy costs in water purification

Researchers at the U.S. Department of Energy's (DOE) Oak Ridge National Laboratory (ORNL) are charting a new course in water purification to reduce energy consumption and costs while ensuring safe drinking water.

ORNL and its research partners created twin systems-one real, one virtual-linked in real time so the twins operate together. The "digital twin" continuously monitors the physical twin's operations and energy prices remotely, then updates the real-world settings, at least every hour, to improve performance, lower energy use, and avoid maintenance downtime. The "digital twin" thus saves money for utilities and local governments.

"Digital twins are increasingly used as platforms for safely testing how new approaches affect complex systems," said Subrata Mukherjee, who leads the project for ORNL. "This project pairs a digital twin with a physical system, so they provide constant feedback to each other while operating. This unique approach supports data-driven decision making for water utility owners and operators."

Industry partner tests approach in real-world water system

ORNL created the digital twin for the project led by the University of California, Irvine, which developed and manages the physical pilot plant hosted by local municipal partner Orange County Water District (OCWD). The pilot is a precise, miniature representation of the full-scale drinking water reuse facility owned by the regional water provider. Most water treatment plants run at fixed rates and require manual adjustments, but ORNL's digital twin prompts the pilot plant to alter flows as electricity prices fluctuate during the day. For OCWD, the concept explores a forward-thinking approach for how utilities could operate in the future.

Conventional digital twins often depend on exhaustive, physics-based simulations. These approaches require extensive computing power and time-consuming setup using large sets of training data. ORNL avoided these drawbacks by developing data-driven models that use a compact set of flow observations and operating features to predict power demand and minimize costs.

This approach speeds both deployment and responsiveness of the technology, which can be adapted to any type of water system, from a drinking water reuse plant like OCWD's to a plant that removes salt from seawater.

Project funding was provided by the National Alliance for Water Innovation, a public-private partnership supported by DOE, as well as DOE's Office of Critical Minerals and Energy Innovation, the California Department of Water Resources, and the California State Water Resources Control Board. In addition to Mukherjee, other ORNL researchers who contributed to the project include Kris Villez, Alex Melin, and Sally Ghanem. OCWD research and plant staff provided key testbed support for operating the pilot system.

UT-Battelle manages ORNL for DOE's Office of Science, the single largest supporter of basic research in the physical sciences in the United States. The Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science.