Research On Adaptive Energy Management Strategy Of Modern Tram Based On Condition Recognition

With the continuous advancement of China’s urbanization process and the rapid development of economy and society,the city size and population have been growing rapidly,resulting in increasingly serious urban problems caused by environmental pollution,traffic congestion and energy crisis.In recent years,the development of urban rail transit has become the key to solve the above problems.Among them,modern tram is gradually becoming one of the important public transportation means to realize environmental friendliness and improve urban traffic conditions.At present,the on-board energy storage modern tram has been widely used,and its energy management strategy needs to be further studied.In this paper,Hybrid Energy Storage System(HESS)online Energy Management Strategy(EMS)and bidirectional DC/DC converter control Strategy are studied for Hybrid Energy Storage System(HESS)Hybrid Energy Management System(EMS)and lithium titanate battery as power source trams.Firstly,the common topologies of hybrid energy storage systems are compared and analyzed,and the semi-active battery topologies with good control effect and relatively low cost are selected.The lithium titanate battery and supercapacitor were modeled and analyzed.The Arbin monomer tester was used to identify the parameters of the Rint model of SCIB lithium titanate battery,and the transient power characteristics of the energy storage element were analyzed.The working principle of bidirectional DC/DC converter is analyzed,which lays a foundation for energy management strategy and control strategy of the converter.Secondly,an adaptive energy management strategy based on condition recognition is proposed.In order to avoid the adverse effects of frequent charge and discharge,transient power and peak current shock on the life of lithium titanate batteries,a power signal filtering method based on real-time wavelet transform was proposed.Based on the real train power signal data of wuhan tram line T1,the tram operating condition was constructed by means of clustering analysis.The Back Propagation(BP)neural network was trained to identify the operating conditions.On this basis,combined with the real-time wavelet transform results and the overcharge and overdischarge protection strategy,the online power distribution is carried out,and the adaptive strategy based on the condition identification is formed.Matlab/Simulink was used for comparative simulation analysis to verify the effectiveness of the adaptive energy management strategy based on condition recognition.Thirdly,an improved model predictive control method for bidirectional DC/DC converter is proposed.Starting from the shortcomings of PI controller and traditional predictive current control method,the objective function in the model predictive control method is improved by combining the results of condition identification.By adjusting the weight coefficient,the flexibility of the objective function and the dynamic performance of the system are improved,and the calculation time is reduced.Matlab/Simulink was used for comparative simulation analysis to verify the effectiveness of the proposed control strategy.Finally,this paper builds the prototype platform of 90 k W tram hybrid energy storage system.The system architecture,software and hardware design of the hybrid energy storage platform are introduced,and the charge and discharge experiments and the verification experiments based on the rule energy management strategy are carried out with the prototype.