Research On State Estimation And Equalization Control Algorithms For Automotive Lithium-Ion Battery Pack

In today’s era of surging energy demand and increasingly severe environmental pollution,new energy electric vehicles,as an alternative to traditional fuel vehicles,have received strong support from the state due to their advantages of low emissions,low pollution,and low cost.As the energy source of new energy electric vehicles,lithium-ion batteries have been widely used in electric vehicles and hybrid vehicles of oil and electricity(Hydrogen fuel batteries and lithium-ion batteries)due to their high working voltage,high energy density,long cycle life,low self discharge rate and other advantages.Correct estimation of battery cell state can avoid overcharging and discharging of the battery pack,thereby avoiding safety accidents and providing the possibility for precise balance control of the battery pack;Long term use of battery packs with significant differences between individual battery cells can also cause serious safety accidents.To avoid accidents and increase the efficiency of battery pack usage,it is necessary to balance and control each cell in the battery pack.This paper takes the vehicle lithium-ion Battery management system as the research object,and conducts corresponding research on its two important functions at the control algorithm level,namely,lithium-ion battery state estimation(state of charge and state of health)and equalization control.The main research work of this article is as follows:(1)The dynamic characteristics of the battery cell are described based on a second-order RC equivalent circuit model,and an adaptive algorithm is designed based on the equivalent circuit model to improve estimation accuracy.Then,the Radial basis function neural network algorithm is used to compensate the uncertainty bias and interference in the lithium-ion battery model online.Finally,the extended Kalman filter nonlinear observer is used to update the state variables,covariance and Kalman gain continuously and adaptively,so as to realize the accurate estimation of the state of charge.The simulation results show that the state estimation algorithm designed in this paper can not only achieve accurate state of health estimation and state of charge estimation at the level of battery packs composed of different cells(18650 lithium-ion batteries and lithium-iron phosphate battery)or even multiple cells.(2)Based on port Hamiltonian theory,port Hamiltonian modeling is performed on the open circuit voltage of lithium-ion cells in the battery pack,as well as the capacitor voltage and inductor current in the Cuk converter.Corresponding IDA-PBC algorithms are designed according to two different dynamic modeling methods.Afterwards,a SIMULINK/PLECS simulation modeling was conducted on the equalization system composed of n series cells and n-1 Cuk converter,verifying the effectiveness of the IDA-PBC algorithm applied to the equalization control of lithium-ion battery packs from a simulation perspective.The simulation results showed that the final state of charge of each cell in the balance system of the two modeling methods can converge to a certain range,and the accuracy is maintained within 0.2%,In addition,the inductor current and capacitor voltage in each Cuk converter in the equalization system,as well as the open circuit voltage of each cell in the battery pack,will eventually approach a certain demand value,achieving better trajectory tracking control.The simulation results of both models also indicate that an equilibrium system that considers more state variables not only reduces the time required for equalization,but also improves the equalization speed.It also reduces the duty cycle oscillation of MOSFETs in the Cuk converter,thereby protecting the switching devices.(3)A two-layer balanced topology consisting of a DC-DC converter and multiple cells was designed based on the concept of "power conservation".Multiple battery cells are connected in series to form a multi-module system,and the bottom layer of the balance system is composed of multiple battery cell systems connected in series.Each battery cell in the bottom layer of the multi-module system is connected in series with a direct switch,and connected in parallel with a bypass switch to form a corresponding balance topology;The top layer is composed of several DC-DC converters connected in series with the output terminals of a multi-core system.Then,proportional control algorithms and optimal control algorithms were designed for module to module balance and intra module balance,respectively,in order to achieve the goal of quickly and accurately converging the SOC of each cell to a certain range of balance.Afterwards,the impact of different equalization cycles on equalization accuracy was compared,and the impact of different individual cell numbers in the same equalization cycle connected circuit on the intra group equalization speed was analyzed and compared,as well as the impact of different bus currents on inter group and intra group equalization speed.