Use of Parallel Processing in Applications for Hydro Power Scheduling – Current Status and Future Challenges

AbstractThis paper gives an overview of models for hydro power scheduling and market simulation developed at SINTEF Energy Research. All of the models are in operative use by power producers, transmission system operators, consultants and regulators operating in the Nordic power market. Several of the models have been adapted to use of parallel processing to decrease computation time. The paper gives an overview of existing models where parallel processing has been applied.Finally future challenges and the need for higher level of parallelization are discussed and exemplified by two new models for power market simulation that use two level parallelization. Both of these models have the potential for efficient utilization of hundreds or even thousands of processors (or cores) and can be run on large compute clusters