Research On Thrust Allocation Of Surface Vessel With Cycloidal Propeller Under Task Constraint

China has a vast ocean.It is very important to develop marine resources and maintain the security of the territorial sea.Not only a number of mining vessels and exploration vessels are needed,but also the auxiliary vessels such as rescue vessels and salvage vessels are needed too.Dynamic positioning system is an important core equipment to ensure the operation of engineering ships in the deep sea.Thrust allocation is an important part of dynamic positioning system.Generally,the controller calculates the expected thrust and moment of the ship at the next moment,and the thrust allocation unit calculates the thrust and angle required by each propeller,so that the propeller generates the corresponding force and moment to meet the expected force and moment of the controller.In this paper,a rescue ship with main propeller-rudder is taken as the research object,and the thrust allocation of the ship with main propeller-rudder is studied by simulation.The research content mainly includes the following aspects:Firstly,the mathematical model of dynamic positioning ship is established,including ship motion model and environment interference model.This paper introduces the characteristics and mathematical model of the propeller,analyzes the force on the rudder and the calculation method of the rudder force and direction.According to the characteristics of the thrust region of the main propeller-rudders,a non-convex region processing method is proposed.The source and calculation method of the thrust loss of the propeller are discussed,which provides the theoretical basis for the simulation experiment.Secondly,the essence and mathematical description of thrust allocation are introduced.The main factors that need to be considered in the target function of thrust allocation are analyzed,including energy consumption,thrust allocation error,propeller wear and noise,the bow thrust,etc.In addition,the thrust balance and the limitation of thrust magnitude and angle are considered.This paper summarizes the search steps of the main propeller-rudder in different situations,and proposes a search strategy for the case of too large thrust allocation error.Then,the principle and basic steps of the basic genetic algorithm are introduced.In view of the shortcomings of the basic genetic algorithm and the needs of the thrust allocation problem,this paper improves several aspects of the basic genetic algorithm,and proposes an adaptive genetic algorithm based on M-elite strategy.In the new algorithm,elite strategy is added to protect individuals with high adaptability.A new mutation method is used.According to the different algebra of the algorithm,the position of mutation operation is different,which makes the evolutionary operation more targeted and greatly improves the efficiency of the algorithm.The termination condition of the algorithm is optimized,and the solution of the previous time is introduced into the initial population,so as to improve the search speed.At the same time,a hybrid optimization method is proposed to solve the problem that genetic algorithm lacks the ability of precise search.Bat algorithm with simple structure and strong ability of local optimization is used to assist the optimization of genetic algorithm.By adjusting the parameters r₀ and f_max of bat algorithm,the local optimization ability of bat algorithm is emphasized,and the search results of genetic algorithm are optimized by using bat algorithm,so as to improve the overall accurate search ability of the algorithm.The performance of the three algorithms is tested by different test functions.Finally,a simple PID control system is designed for the simulation ship,and the closed-loop experiment is carried out for the ship.Then the three algorithms are verified by the simulation experiment,and the simulation results are analyzed and compared.It is proved that the improved algorithm can solve the problem of ship thrust allocation with main propeller-rudder,and the algorithm has good global optimization ability and robustness.