Decentralization, regionalization and electricity grids. Meta-study on assumptions, insights and narratives for the Renewables Grid Initiative (RGI)

In the discourse about infrastructure expansion that is robust and for which public acceptance is assured, the relationship between decentralization and the future demands on power grid infrastructures is a critical issue. It includes the whole spectrum of applicable interrelationships, the myriad areas of tension and complexities of centrality, decentralization and so-called ''cellular'' approaches. The issue of decentralization - which is often handled very vaguely and (too) often features rather crude narratives - requires a nuanced, differentiated analysis. In a first step the present study reviews and analyzes the different dimensions and aspects of decentralization of electricity generation based on literature reviews. This finds, first of all, that a purely technical approach to the relationship between decentralization and grid expansion (small vs. large installations, connected voltage level) is not a viable approach. A crucial factor in the context of grid expansion is, firstly, the proximity of power generation plants to electricity customers. If a large share of the power generation is decentralized, the pressures on the electricity grid can naturally be reduced. Secondly, the proximity of the flexibility options (e.g. demand flexibility, storage, back-up capacities) to the electricity customers is of major importance, since such flexibility options will play a fundamental role in an electricity system based on renewable energies. In a second step, data analyses (with a high spatial resolution) were conducted on the limits of potentials for absolute solar and wind power generation and on the corresponding demand structures (in both cases on a district level). These analyses initially completely exclude the cost or availability issues of flexibility options and contain only quantity balances with a high spatial resolution. They show that, firstly, there is a substantial concentration of demand in the industrial regions in the west and south and in the metropolitan regions of Germany. In a third step, a comparative analysis is conducted for a wide range of models of the German electricity system that have different designs and use very different methodologies. Scenarios that calculate a 20% to 50% lower need for grid expansion have the following characteristics: 1. The scenarios assume or determine a strong expansion of onshore wind energy in the ''South'' zone. The scope of the additional grid expansion resulting for 2030 and 2035 is three to four times, and in extreme cases six times, higher than the values assumed in the network development plans. 2. A disproportionate expansion of onshore wind energy in the ''West'' zone is predominantly assumed or calculated. The additional grid expansion amounts to a factor of 2 to 3, and in two extreme cases to a factor of 7, higher than that assumed in the network development plans. 3. Largely, albeit not consistently, a very strong expansion of solar power generation is assumed in the ''South'' zone. The capacities of PV systems in the ''South'' zone exceed that of the network development plans for 2030 and 2035 by a factor of 2 to 3. (4.) For 2030 the relationships between the remaining coal-fired power plant capacities and the necessary grid expansion depend to a great extent on how (additional) renewable power generation is regionalized. For 2035 the amount of coal-fired power generation no longer shapes the dimensions of electricity grid expansion. Viewing the three steps of the analysis overall, a number of recommendations for action can be derived in addition to the above-mentioned conclusions. Firstly, a structured discourse is needed to clarify whether and in which model or at what times decentralized (''cellular'') control approaches - aside from optimization of self-consumption - could be implemented or considered as a variant for grid expansion planning. Secondly, the assumptions for expansion limits of renewable power generation need to be validated. This is the case for onshore and offshore wind power capacities as well as PV power generation in high spatial resolution, at least for the zones and particularly the ''South'' and ''West'' zones in Germany. The real land potentiality and acceptance should receive special attention. Thirdly, there is an urgent need to develop a uniform assessment criteria for calculating all the costs and land requirements (for electricity generation plants, flexibility options and infrastructures) in order to enable comparability in future analyses. Fourthly, to improve the comparability of future studies, it would be helpful to develop a pragmatic metric that can be used to compare the grid expansion needs and take into account the different modeling approaches. The present metastudy is the first comprehensive attempt to analyze the complex fields of tension between decentralization and grid expansion, which have been shaped by different narratives and present many conceptual and data challenges. Further research needs to be conducted on these aspects.