Abstract: In large-scale offshore wind farms, there exists a coupling relationship between the collection system and high-voltage export system in terms of capital investment, power loss, reliability and other indicators. Traditional independent planning methods hardly yield the globally optimal scheme from the full life-cycle perspective. To address this issue, a hierarchical joint optimization method oriented to full life-cycle cost is proposed for the collection and export system of large-scale offshore wind farms. On the inner layer, based on radial topology, the cable investment, power losses and outage costs caused by faults are comprehensively considered. An improved sector-based clustering method is used to partition wind turbines, and the intra-cluster collection topology is transformed into a mixed-integer linear programming model that can be solved efficiently. On the outer layer, the offshore booster station location is taken as decision variables, and a derivative-free search strategy combining Latin hypercube sampling (LHS) with the Nelder-Mead simplex method is used. Case studies show that the proposed method reduces the total system cost by about 4% compared with the commonly used centroid-based method, demonstrating its effectiveness and engineering applicability.