Hydrogen Production from Wind and Hydro Power in Constrained Transmission Grids

Hydrogen (H2) production from variable renewable energy (VRE) sources can be an important technology for reducing CO2 emissions. Integrating electrolytic H2 production into the electricity system can contribute to this objective on two fronts. Firstly, electrolytic H2 production will be important to enable large shares of VRE integration into the electricity system as it can provide demand-side flexibility at scale. Secondly, H2 can mitigate CO2 emissions in end-use applications that is hard to decarbonize by other means, such as direct electrification. The scope of this thesis is limited to study the effects of large-scale H2 production on the electricity system. Mathematical models are developed and used to study how flexible H2 production can enable cost efficient VRE integration and reduction of CO2 emissions. The models include representations of H2 production as flexible demand and sectorcoupling between H2 and electricity systems. The sector coupled model enables an integrated analysis of multiple H2 production pathways, based on electricity and natural gas. A stochastic rolling horizon dispatch capacity expansion problem is formulated and modeled to investigate the impact of short-term uncertainty from VRE sources on investments in the electricity system. This approach also incorporates the modeling of long-term storage. The results shows that more installed capacity will be needed to handle short-term uncertainty compared to the results from deterministic models. This model can provide an important foundation for future models that studies the impacts of flexibility from H2 production on electricity systems. H2 production is studied in two different case studies, based on the electricity systems of northern Norway and the state of Texas in the US. The two electricity systems have different sources of flexibility, where hydro power provides flexibility in northern Norway and natural gas in Texas. Flexibility from natural gas is a source of CO2 emissions while the emissions from hydro power are negligible. Thus, flexibility from electrolytic H2 production has more potential to reduce electricity system emissions in Texas, while it will be important for efficient integration of high shares of VRE in both systems. In northern Norway, the results from the case studies show that it will be important to operate H2 production in a flexible way to not reduce security of supply. This is especially important in regions where the transmission grid is constrained, such that flexible H2 production can help reduce the amount of transmission grid expansion that is needed in relation to wind power integration. Competitive production of low-cost H2 in northern Norway is dependent on sufficient hydro power flexibility and low wind power investment costs. This result in low and stable electricity prices, which represent 77-89% of the levelized cost of H2 production. In the case studies for northern Norway with 24 hours of hydro power flexibility, the future H2 production cost is estimated to be 1.89 e/kg. In Texas, the results from the case study show that CO2 emissions can be reduced by using flexible H2 production to provide some of the balancing energy which is currently provided by natural gas turbines. H2 can also be produced from natural gas in combination with carbon capture and storage (CCS) to utilize the natural gas resources with low CO2 emissions. This is more cost-efficient than using CCS with natural gas turbines. Furthermore, the results show that higher CO2 prices favor electrolytic H2 production while more H2 demand favors natural gas based H2 production. Electrolytic H2 production is competitive with natural gas based H2 pathways for CO2 prices of more than 60 $/tonne, supplying 40-80 % of the total H2 demand. Emissions from H2 production is less than 1.2 kg CO2/kg H2 with a CO2 price of more than 90 $/tonne as H2 is produced from electrolysis and natural gas with CCS. Flexible H2 production is necessary to enable high shares of VRE integration above 80-85% in the Texas case study. With high shares of VRE electricity generation the electricity price will be variable and electrolytic H2 production will be concentrated to surplus hours when the electricity price is low. The results show that H2 storage has a duration of 5-37 hours and provides flexibility which is complimentary to batteries (2-7 hours). The future H2 production cost in Texas was estimated to around 1.30-1.66 $/kg when the CO2 price is 60 $/tonne CO2. This shows that electrolytic H2 based on renewable electricity can be produced at lower costs than the current H2 price which are around 2.8-3.3 e/kg in Europe and the US (late 2020)