Objectives  To solve the issue of thrust loss caused by the interaction between the ship thrusters and the excessive power fluctuation in the power system, an improved A success-history adaptive differential evolution(SHADE)algorithm is proposed.

  Methods  A multivariable optimization problem is established, which takes the power consumption of the propulsion system as the objective function, the thrust and the directional angle of the thruster as the control variables, and the change rate of the thrust and the directional angle of the thruster as the constraint conditions. For the traditional SHADE algorithm, an improved SHADE algorithm based on the degree of iteration is proposed to solve the thrust distribution optimization issue. The simulation results of the traditional SHADE algorithm and the improved SHADE algorithm are compared.

  Results  The simulation results show that under the same simulation environment, the error between the actual thrust torque instruction obtained by the traditional SHADE algorithm and the desired thrust torque instruction generated by the dynamic positioning controller is large, and the power of the thruster fluctuates greatly; while the improved SHADE algorithm can control the torque command error of the expected thrust and the actual thrust within a small range, in which the surge and the sway error is within ±5×10^-5 kN, and the yaw error is within ±3×10^-5 kN·m.

  Conclusions  The improved SHADE algorithm can transfer the dynamic positioning command controller more accurately under the action of environmental disturbance, to solve the problem of ship power system thrust distribution, and lay the foundation for how to use the thrust distribution to maintain the stability of the ship's power system.