This work develops and applies a dynamic Generation Risk Assessment (GRA) framework to quantify the ability of a modular high temperature gas cooled reactor (MHTGR) system to satisfy a high temperature electrolysis facility’s (HTEF’s) steam and electrical demands. The methodology extends conventional probabilistic risk assessment by explicitly modeling demand based performance metrics, non safety derates, safety related reactor trips, common cause failures, and dynamic dependencies. Individual MHTGR subsystems—such as the helium purification system, reactor plant cooling water system, main circulators, steam generators, turbine generator auxiliaries, and electrical distribution components—are parameterized using subsystem level failure and repair models with associated MTBF, MTTR, and derate relationships.

The analysis employs Event Modeling Risk Assessment using Linked Diagrams (EMRALD) dynamic PRA to capture state dependent transitions, time varying interactions, and maintenance driven outages. This includes explicit modeling of startup transients, staggered periodic maintenance, turbine cumulative versus non cumulative derates, and reactor/turbine capacity under varying failure states.
Results show that the baseline four pack MHTGR configuration exhibits no redundancy, causing generation reliability to fall sharply whenever any reactor or turbine is unavailable. Decomposition of unreliability factors indicates that periodic maintenance and random subsystem failures dominate generation unreliability. Sensitivity studies were performed to assess how results vary under different assumptions, including enhanced subsystem reliability expected from modern HTGR designs, alternative turbine derate logic, and increased redundancy up to two four packs of MHTGR. An additional analysis evaluates the trade offs between overbuilding reactor units and supplementing power supply from the grid.

Comparisons with static PRA using Systems Analysis Programs for Hands on Integrated Reliability Evaluations (SAPHIRE) show agreement in results and highlight complementary insights: dynamic GRA captures time dependent dependencies and variations in power output, while static PRA more effectively captures extremely low probability events but lacks temporal fidelity. Collectively, the work demonstrates that dynamic PRA provides essential insights for integrated energy system (IES) design, revealing reliability constraints, key failure contributors, and trade offs between redundancy, derates, and supplemental grid power.