Abstract
Stochastic energy flows are an increasingly important phenomenon in today’s power system planning and operation. They are – among other reasons – caused by large amounts of stochastic generation such as wind. The inclusion of energy storage devices, distributed in future systems (distributed energy storage – DES), is continuously being mentioned as a possibility to alleviate some of the problems arising from stochastic generation. The authors show that the potential ownership of the DES systems is an important criterion on which probabilistic methods will be applied for assessment. The potential owners are either the grid operators, the generation owners, or the energy traders. For the grid operators being the DES owners, storage operation will have to be integrated into the planning of the system, therefore multivariate nonparametric time series analysis and synthesis methods have to be applied to recorded data of stochastic energy resources. Together with suited storage models, the implications of DES on the planning of the system can then be assessed. For the producers or traders being the owners of the DES, the topic to be addressed is the real-time operation of each storage device in the power system, which is linked to the optimisation of the economic value of the stochastic resources. In this case, forecasting and operations research issues are paramount. Recently developed methods including scenario development from non-parametric forecast models for the following trading period and probabilistic assessment of necessary storage capacities for hedging with given financial risks are explained
It is generally stated that the non-standard distributions of the stochastic infeeds, as well as complex chronological persistence and interdependence phenomena complicate the modelling procedure and leave space for a large range of research activities on DES in the future. The exact description of how the owners of storage assets are embedded into the energy market frameworks of the future is crucial for the probabilistic quantification of benefits introduced by DES