| With the increasing intelligence of ship microgrids and the complexity of their internal structures,the degree of electrification is getting higher and higher,and the control problems of ship microgrids are becoming more and more prominent.In order to better control the microgrid of complex ships,among the many methods,model predictive control methods have received widespread attention.Considering that the ship microgrid itself has complex characteristics such as multivariable,strong coupling,and nonlinearity,it is difficult to model the ship microgrid,so the system can be identified from the data of the system.The statespace model of the ship microgrid is obtained through the subspace identification technology,and the performance indicators of the system are optimized by using the model prediction control method,so as to obtain a good control effect.At the same time,it is considered that the factors that cause the pulse waveform to change during the driving process of the ship microgrid will cause the power fluctuation of the DC link end,which in turn will affect the voltage stability of the DC bus.Starting from the algorithm and characteristics of subspace identification,combined with the model prediction control algorithm,this paper conducts indepth research and discussion on the problems of constraints and disturbances in the ship microgrid.The main research work of this paper is as follows:(1)For cases where the system parameters are unknown,the multivariate output error state space algorithm in the subspace identification method is studied,and the theoretical derivation process of the identification method is explained,that is,the Hankel matrix is established through the input and output data information of the system and decomposed to obtain the extended observation matrix,and then the state space coefficient of the system is obtained according to the expansion matrix to establish the model of the system.On the basis of analyzing the modeling process of ship microgrid,the state-space model of ship microgrid is obtained by subspace identification method and the situation of ship pulse load is considered.This identification method has the advantages of simple calculation and accurate model identification.(2)This paper introduces the predictive control method of ship microgrid model based on subspace identification.The multivariate output error state space algorithm is used to obtain the state space model of the ship microgrid and use it as the prediction model of the model prediction control method.The cost function of the model predictive control method is defined,considering the optimal value of the solution in the presence of constraints in the system.Aiming at the problem that the model does not match due to the change of system parameters,the online subspace identification is implemented,the updated system model is obtained,and the stable operation of the system is maintained through the model prediction control method.In this paper,this method is applied to the simulation of ship microgrid,and the final results show that the method has a good control effect on the basis of accurate identification and prediction model.(3)Considering the disturbance factors that occur during the actual operation of the ship microgrid,the robust model prediction controller of the ship microgrid based on subspace identification is designed for the situation that the bus voltage fluctuation caused by the system due to interference.In this paper,the H∞ robust model prediction control method is adopted,and after the object model is switched,the online data is collected,and the model is updated in real time by the subspace identification method to solve the constraints and interference problems of the system in time,and the method can realize the dynamic optimal control of the system operation process.In this paper,this method is applied to the situation of pulse disturbance and ship load change in the ship microgrid,and the final simulation results show that under the control of the method,the ship microgrid can obtain an accurate statespace model and maintain stable operation. |