Abstract
The green transition requires electrifying industries with traditionally stable energy demands. Combined with the rise of artificial intelligence (AI) and hyperscale data centers, a significant increase in grid-connected baseload is expected. These loads, with high capital and operational costs, often lack financial incentives for flexibility. This paper explores how the modeling of additional load affects the optimal energy mix under varying nuclear energy overnight construction cost (OCC) levels, highlighting nuclear energy’s potential role in providing the necessary baseload for AI data centers and heavy industry electrification. By utilizing an analytical approach, the study assesses how additional load profiles match variable renewable energies (VRE) outputs to determine the mix of technologies to be responsible for accommodating additional power demands. A stylized case study using the baseload addition (BA) method showed a significant increase in the share of baseplant units, handling 95.1% of the additional load. In contrast, linear load profile scaling (LLPS) of historical loads left the energy mix unchanged. A more detailed case study with the European Model for Power system Investment with Renewable Energy (EMPIRE) confirmed the same trend as found in theory, indicating a 24% increase in nuclear generation using the BA method over historical load scaling. Moreover, a low-cost nuclear scenario (€4200/kW) installed 59% more capacity than a high-cost scenario (€6900/kW). Finally, higher nuclear shares are shown to significantly reduce the need for transmission, storage, VRE curtailment, and land use, emphasizing nuclear power’s potential role in low-carbon power systems.