| Rail transit has gradually become the mainstream of people’s travel,and tram as the main means of transportation in cities,promoting its development will improve people’s travel efficiency.Among them,the hybrid energy storage system(HESS)power supply method improves the urban landscape and has the characteristics of high power density and high energy density.In the case of a certain energy storage configuration,using traction calculation simulation to simulate the running process of the tram is a prerequisite for its safe operation.Similarly,after the operation line is determined,it is also very important to optimize the capacity configuration and energy management strategy of the HESS to achieve its high-efficiency and energy-saving operation.Based on the content introduced above,this thesis studies the development of tram traction calculation software,on-board energy storage system capacity configuration methods and software development,and online optimization of energy management strategies.In order to quickly realize the calculation and simulation of tram traction,this thesis uses Java and Swing framework to develop an energy storage tram traction calculation software.First,the mathematical modeling of hybrid energy storage system and traction calculation are established,and when the speed limit conditions are more complicated,based on the shortcomings of the current forward and reverse traction calculation methods,a traction calculation method with adaptive merging interval is developed.Finally,the traction calculation and energy management strategy are calculated in parallel,and verified the calculation accuracy of the software.In order to comprehensively consider the coupling between the installation weight,energy loss,life loss,energy management strategy and capacity configuration of HESS,a fuzzy control-based multi-objective collaborative optimization method for capacity configuration is proposed.Fuzzy rules are formulated according to line conditions and HESS characteristics.In order to reduce the subjectivity of the membership function,the improved NSGA-II is used to collaboratively optimize the dual-source configuration and fuzzy control parameters.After optimization and comparison,the hybrid energy storage battery semi-active topology is more suitable for the operation requirements of trams on this line.On this basis,the optimization effects under different energy management strategies are compared,the simulation results verify the effectiveness of the method proposed in this thesis.Finally,using Java and Swing framework to develop parameter matching and control strategy simulation software,which is used to quickly calculate the optimal value of the hybrid power system configuration and simulate the actual operating conditions of the system.In order to improve the operating performance and economic benefits of trams,it is necessary to optimize the design of energy management strategies.At present,most energy management strategies are offline optimization.In order to achieve online optimization control and avoid the impact on the strategy when the driving conditions change greatly,this thesis proposes a reinforcement learning energy management strategy based on working condition recognition.Firstly,the typical driving conditions of trams are constructed by historical driving data and short-stroke analysis method,and obtained the Markov power state transition matrix under different operating conditions.Then,with the goal of minimizing HESS energy consumption,an energy management strategy optimization model was constructed,and the power allocation strategy under different working conditions was obtained through the reinforcement learning algorithm.Finally,the improved LVQ neural network was used to identify driving conditions online,and the control system makes real-time decisions based on the current identified operating conditions and train status.Through simulation comparison,the optimality and adaptability of the proposed strategy were verified.Finally,based on the hybrid energy storage system experiment platform,the energy management strategy involved in this thesis was tested and verified.The upper computer sends the power distribution control command to the controller in real time to verify the validity and feasibility of the strategy proposed in this thesis. |
