Case Study: PLEXOS and Power Modeling Software
Power Market Modeling Improved 1000x
Energy production cost simulations for large-scale power market planning are now possible with substantially higher throughput and greater accuracy to help you make more informed energy grid optimization and expansion decisions.
Working in partnership with the HPC Innovation Center, Energy Exemplar’s power market modeling and simulation software PLEXOS® has been boosted 1000x in performance using HPC scaling methods. Running on LLNL’s leadership-class supercomputers using IBM’s CPLEX Optimizer, this parallel version of PLEXOS® dramatically reduces the time needed for power grid optimization across multiple timescales, using stochastic unit commitment models and five-minute economic dispatch.
This capability is now being made available through the HPC Innovation Center to electric power grid operators to enable more informed capital spending and grid optimization decisions. Physics-based uncertainty modeling of renewable generation can be added to analyze operational impacts of these energy sources.
To find out more, please send us a message using the form to the right.
Case Study: California Energy Grid
Mandated to utilize 33% renewable energy sources by 2020, the California Energy Commission looked to Lawrence Livermore to determine whether new energy storage technologies and demand response initiatives might help balance the load on generators as more solar and wind resources are added to the grid.
A multidisciplinary team of LLNL computational engineers, computer scientists, and atmospheric physicists assembled to create a comprehensive planning system for electric grid research.
Using a stochastic production simulation model running on Livermore's high performance computers, they optimized unit commitment and economic dispatch of 3000 days of operation to cover a range of possible weather scenarios and grid configurations. This analysis campaign would have taken 342 years running on a single processor workstation — an impractical timescale. Instead the study was performed in less than one month using a parallel version of PLEXOS software running similtaneously on thousands of cores, applying innovations developed by Livermore, Energy Exemplar, and IBM scientists and engineers.
Transformational Solutions. Delivered.
HPC Innovation Center engagements include access to high performance computers, parallel software and the expertise of Livermore's staff of more than 2500 scientists and engineers. Projects can be turn-key or collaborative to allow sponsoring partners to expand their in-house adoption of HPC technologies.
Do you have other software that could benefit from scaling? Let's have a conversation to explore how we can help.