Research On Design And Real-time Optimization Control Of Power Storage Power Supply Syste

With the rapid development of the economy,transportation tools such as cars,trains,and ships play an important role in people’s daily life,which also directly leads to a great increase in demand for fossil energy such as coal and oil,bringing problems such as energy exhaustion and environmental pollution.In September 2020,the country clearly proposed carbon-to-peak and carbon neutrality,and vigorously develop clean energy such as wind,solar energy,and ocean energy to achieve green improvement.On this background,transportation tools have transformed from a fossil fuel power system with internal combustion engines to a new energy power system with electrical energy.At present,the new energy power system mostly adopts the structure of dual energy sources or three energy source systems,and the comprehensive energy and independent power control strategy is used.However,it is difficult to cope with the "multi-energy complement,multi-time scale,multi-subject collaboration" under the system structure.Therefore,it is urgent to design a power storage system that meets the above requirements.In this paper,a hybrid power system configuration is proposed,which uses fuel cells,lithiumion batteries and super-capacitors(which are called three power sources below)as power sources,and a two-layer model predictive control based on event-driven mechanism optimization control strategy is designed.The main work completed is as follows:Firstly,the unified dynamic model of hybrid power system is established.The mechanistic model and data-driven method are used to analyze the three energy sources of the system,and then dynamic simulating models are built based on Matlab/Simulink platform.A state of charge(SOC)estimation method for lithium battery and super-capacitor based on Extended Kalman Filter(EKF)algorithm is established and verified by simulation.The structure of three-power converter(DC/DC converter)is analyzed under different modes and circuit parameters are optimized.Based on the data-mechanism collaborative driving model of hybrid energy power system,this paper solves the problem that the energy balancing model is mostly established from the steady-state perspective,and establishes a multi-energy coupling dynamic model,which can improve the performance and energy utilization efficiency of the system.Secondly,a two-layer MPC based on event-driven mechanism optimization control strategy is designed,and the multi-objective joint optimization cost function is established and solved efficiently.By calculating the driving function,the event-driven mechanism updates the system control input in real time,which solves the problem of multi-time scale dynamic response(minute response of fuel cell,second response of lithium battery,millisecond response of super-capacitor)characteristics of the three power sources under external excitation,in order to respond quickly.The two-layer MPC integrates steady-state target calculation and dynamic optimization control to solve the problem of performance degradation caused by fast response of the system.Real-time power scheduling is carried out under hierarchical decision-making mechanism,so the fast response system can realize highprecision real-time optimization control under multi-energy coupling constraints.Finally,the optimization effects of model predictive control,two-layer MPC and twolayer MPC based on event-driven mechanism are compared and analyzed.It can be concluded that,compared with other control strategies,the two-layer MPC based on event-driven mechanism optimization control strategy designed in this paper has advantages of high stability,fast response speed and high system efficiency.