The optimization of avoiding typhoon routes stands as a crucial challenge in weather routeing. To address this problem, a graph search algorithm for tropical cyclone avoidance route planning is proposed. This approach uses ship speed loss model and ship sailing risks model under the influence of typhoons. It considers a range of ship speed variations, defines the weights for each edge in the graph search algorithm, and enhances the search strategy based on predefined routes. These improvements significantly increase the efficiency of the algorithm's search and the smoothness of the typhoon avoidance routes, ultimately with the aim of minimizing voyage duration. Using Typhoon "Doksuri" as an illustrative example, this paper validates the effectiveness of the algorithm in designing typhoon avoidance routes. Experimental results demonstrate that the proposed algorithm can effectively steer clear of Typhoon "Doksuri" and obtain optimal typhoon avoidance routes that minimize voyage duration. This research carries substantial significance in the realization of intelligent typhoon avoidance in weather routeing.