Advancing technology and data modelling for optimisation of non-dispatchable renewable electricity coupled with underground hydrogen storage

Abstract

Intermittent power generation is a considerable obstacle to the integration of wind and solar energy into the electrical grid. To compensate for the seasonal intermittency of wind velocity and solar radiation, large-scale energy storage is required that must be capable of storing energy on a monthly time-scale. Therefore, the feasibility of hydrogen storage in salt caverns coupled to wind and solar power production is analysed. A Mixed- Integer Linear Programming (MILP) optimisation is performed within the boundaries of the Dutch electrical grid by applying the aforementioned technologies. This research is performed in the context of an existing MILP tool. The optimal design and operation is analysed in variation of the methods applied for weather profi?le modelling. The MILP tool calls for hourly weather data on a one-year time horizon. Three methodologies are developed for the creation of a Typical Meteorological Year (TMY) from a long-term weather database; their performances and impact are tested on the Dutch case-study. On the one hand, optimal design determines maximum deployment of wind and solar technologies for all scenarios, with area constraints playing a key role; on the other hand, the optimal system operations highly differ, for instance as respect to trade-off between offshore and onshore power production. Nevertheless, hydrogen storage is deployed on a negligible scale; copious amount of electricity is curtailed rather than stored. The tool is applied with continuous curtailment, which is found to hinder the actual potential of the large renewable capacity installed. To conclude, opportunity for enhancement of the developed methods are reported.