Break-Even Economics of Nuclear Microreactor Fleets in Southeast Michigan

Abstract

Nuclear microreactors are proposed for remote communities, industrial facilities, and data centers, but their economics remain commercially unproven. Without an operating fleet, first-of-a-kind (FOAK) to (n)th-of-a-kind (NOAK) cost progression relies on bottom-up estimates, scaling assumptions, and assumed learning, leaving overnight capital cost (OCC) among the least-established inputs in direct levelized cost of electricity (LCOE) comparisons. Rather than prescribing OCC, this work asks: what OCC must a microreactor fleet reach for its generation cost to equal the fossil generation it replaces? Fleets of 13-, 10-, and 5-MWe Holos-Quad, Aalo-1, and eVinci units are dispatched against measured 2025 coal, natural-gas combined-cycle, and gas-peaking generation from DTE Electric in Southeast Michigan. Because capital cost does not affect hourly merit order, replacement cost is linear in a common capital multiplier (m), enabling break-even ($m^{*}$) from one dispatch. Ten metaheuristics algorithms screen mixed fleets over ten seeds using 1,000 evaluations per run. Selected economic candidates contain only Aalo-1 units: 185, 111, 35, and 124 for coal, combined-cycle gas, gas peaking, and combined cases. Using midpoint fossil costs from Lazard’s 2026 LCOE+, corresponding NOAK break-even OCCs are 6,484, 1,063, 2,414, and 8,918 $/kW. Replacement fraction alone does not determine the threshold: combined-cycle replaces 96.73% of target energy yet supports the lowest positive multiplier, whereas the combined fleet replaces 29.32% and supports the highest, demonstrating the importance of utilization and displaced-source cost. Candidate-specific reactor references are then evaluated using nonlinear model predictive control. All fleets satisfy hourly-linear tracking and remain below 1% under a separate Midcontinent Independent System Operator (MISO)-informed 5-min sensitivity, although the coal case reaches the 0.5 degrees/s drum-rate limit. These OCCs are economic break-even targets, not future-cost predictions; coal and combined-case Aalo-1 thresholds fall within generic microreactor cost ranges reported by the Nuclear Energy Institute.