Battery Identification Modeling And Energy Management Of Locomotive

With develpoment of China's railway electrification,the Internal locomotive has been gradually replaced by the Electric locomotive,which has become the main body of China's railway transport.The Electric locomotive has its own clean pollution,high energy efficiency,high transport efficiency.Nowadays,with the rapid deveploment of China's railway industry,the igh Speed Electric Multiple Unitswas play an increasingly important role in the short-distance railway passenger transport.The industry of China's rail transport will develop toward high speed and efficient in the future.The battery plays more and more important role in the train power supply system as chemical energy storage device.When the train is powered off,for example the temporary parking and sudden failure,the battery starts to work as an auxiliary power supply to guarantee that important electrical equipment of the train emergency electricity,safe,efficient operation of the train and passenger comfort.Battery,as the main entire auxiliary power supply system,its core and basis of state of charge(SOC)estimation and energy management of battery and energy management of the system is based on study working characteristics and accurate Model System Identification in the reasonable and effective experimental program.ence,it is the primary theoretical basis on energy management for the entire train power supply system that make the battery charging and discharging,experimental platform and simulation of different working conditions,equivalent circuit model of battery based on static,dynamic characteristics of battery charging and discharging work.From the system modeling identification and energy management,the battery charging and discharging experimental platform is built to collect the corresponding experimental data,which includes two kinds of battery working mode experimental programs,constant and pulse current.Battery operating characteristics is studied based on experimental data.Battery polarization effect is found in the process of battery pulse test.Moreover,the static work experiment of battery is used to study the effect of polarization on the battery operating characteristics.The previous work provides a certain experimental and theoretical basis for future reseach,including model establishment,state of chargeing estimation,energy management and so on.For polarization,hysteresis and other non-linear dynamic characteristics of battery,the RC circuit equation is introduced into the battery model to establish the first-order dynamic battery model.Laplace transform and forward Euler transformation algorithm is used to deduce and obtain the recursive model,which is suitable for the least squares identification algorithm,which is based on the first order model.For the sake of the mutual influence of the adjacent data and data saturation phenomenon in the process of identification,the forgetting factor is added into the recursive least squaresalgorithm.The model parameters of battery is solved by identification based on experimental data,which is obtained through simulating the actual battery working conditions in battery experiment.The accuracy of the battery model is verified based on MATLAB/Simulink simulation experiment platform.The working simulation model of the Super-capacitor is established based on its Second-order Equivalent Circuit Space State Equation and simulated by MATLAB/Simulink simulation experiment platform.Its working characteristics is studied based on the simulation results.The previous work has provided a certain experimental and theoretical basis for the future reseach,including the state of charge estimation and energy management.The reasonable and effective experimental program is developed to eliminate the influence of battery polarization on state of charge and the relationship between battery state of charge and its true open circuit voltage is established.The model parameters of battery is obtained by identification based on experimental data.The space state equation,which is suitable for the Unscented Kalman Filtering Algorithm,is derived from the battery model.The Unscented Kalman Filtering Algorithm is used to estimate battery state of charge.After study on working characterwastics and polarization effects of battery,the Pulse Three-stage Fast Charge Control Strategy is developed based on traditional Three-stage Fast Charge Control Strategy.Based on the establishment of the accurate model of the battery,the effectted result of battery state on the balance of the battery pack is obtained by simulating.Reasonable and effective battery working deadline voltage is developed by combinating equilibrium simulation and battery pack based on energy storage characteristics of battery and super-capacitor.The battery-super capacitor hybrid auxiliary power supply system is developed.Based on actual operating conditions of train power supply system,the reasonable and effective energy management strategy is invented.The result of system energy allocation is obtained by simulating actual conditions.The energy output result of battery-super capacitor hybrid auxiliary power supply system is analyzed with digital analyswas method to verify the effectiveness of the hybrid system work.The study shows that battery dynamic model meets the targets expected in both model selection and algorithm selection.The battery state charge estimation has achieved high accuracy.It is satwasfied that the proposed battery and system energy management strategy got ideal results.