The design of ocean renewable energy systems requires extensive fatigue assessments, often involving thousands of simulations that increase computational costs and the Levelized Cost of Energy (LCOE). Climate change adds further complexity by altering future marine conditions, which can make traditional fatigue analyses based on historical data unreliable. This study addresses these challenges by combining K-means clustering with CMIP6 climate projections to reduce the number of metocean conditions analyzed while maintaining accurate fatigue damage estimation. Sensitivity analysis indicates that 1000 clusters are sufficient, reducing the number of cases by 96.57% without compromising accuracy. The results also reveal a decline in fatigue damage over the 21st century, highlighting the importance of incorporating future climate projections into the design of ocean renewable energy systems to ensure resilient and cost-effective performance.