HPC Simulations and Machine Learning Help Improve Power Plants

Scientific illustration.
Copyright: IKE, University of Stuttgart

Supercomputing resources and data-driven machine learning helped University of Stuttgart researchers model how coal, nuclear, and geothermal power plants could be retrofitted for cleaner, safer, and more efficient and flexible operation.

In conventional steam power plants, residual water must be separated from power-generating steam. This process limits efficiency, and in early generation power plants, could be volatile, leading to explosions.

In the 1920s, Mark Benson realized that the risk could be reduced and power plants could be more efficient if water and steam could cohabitate. This cohabitation could be achieved by bringing water to a supercritical state, or when a fluid exists as both a liquid and gas at the same time.