Abstract: For the 5 MW ITI Barge floating wind turbine,this paper utilizes the dynamic model and multi-population genetic algorithm to determine the optimal values of each parameter of Tuned mass damper（TMD）in nacelle. To begin with,the Lagrange equation was employed to establish the dynamic model of the turbine incorporating TMD,followed by the utilization of the Levenberg-Marquardt（LM）algorithm to identify the unknown parameters within the dynamic model. Subsequently,the optimization of TMD parameters was carried out using a multi-population genetic algorithm and dynamic model,with the objective function being the standard deviation of the tower’s vertical displacement. Finally,TMD parameters were redesigned according to the optimal solution,and the load reducing effect of TMD was verified by the full coupled FAST model under five typical wind-wave combined load conditions. The results show that the TMD with optimized parameters could effectively reduce the fatigue load of the Barge floating turbine. In comparison to the control scenario without TMD,the utilization of TMD results in a reduction of approximately 6% to 48% in the standard deviation of the tower longitudinal displacement. Similarly,the standard deviation of the tower root longitudinal bending moment decreases by approximately10% to 45%. The standard deviation of blade root longitudinal bending moment decreases by about 11%-33%.