Control Methods Of Modular Multilevel Converter Based Battery Energy Storage System

Due to the energy crsis and environment pollution,the penetration rate of renewable energy in the grid is increasing during the last decade years.Compared to the fossil fuels,the renewable energy features the merits of regeneration capability,little pollution,etc,which is preferable to the sustainable development of society.Among them,the most widely utilized are photovoltaic and wind power.However,they are strongly affected by the natural conditions and the output of them are uncertain and intermittent,which threatens the power quality,stability and reliability of utility grid.To deal with this issue,energy storage system is indispensable due to its ability to decouple the demand of utility system and output of renewable energy.The energy storage system can be classified as three types:physical energy storage system,electromagnetic energy storage system and electrochemical energy storage system.Among them,the most widely used is electrochemical energy storage system,like battery packs,which takes the advantages of large energy density,high reliability and long powering time.Battery management system(BMS)is responsible for detection of battery status,governing its normal operation,and the most important,balancing the state of charge(SOC)to maximize the output power of whole storage system.In order to satisfy the grid connection demand,several battery cells need to be series and parallel connected,boost by the DC/DC converter,inverted by the DC/AC converter and connect to the utility grid with a step-up transformer employed.It is a simple structure with easy control method,but the regulation of large number of battery cells will worsen the performance of BMS.Besides,the voltage and current levels of switching devices are relative high and large filter devices are needed due to its limited output step levels(two or three).Modular multilevel converter(MMC)takes the advantages of modular design,scale ability,etc,which makes it recongised by both industry and academic.Comparing to the two-level converter,MMC features the merits of improved output performance,higher power level,better scale ability and redundancy,etc.The MMC offers the capability of embedding BESS in a split manner,which makes a more accurate regulation of battery cells.In recent years,more researches have been done on modular multilevel converter based battery energy storage system(MMC-BESS).Beyond the application in grid connection,BESS is also widely adopted in the EVs.In EV,the major electronic circuits are motor propulsion circuit,battery charger and BMS.Traditionally,the three circuits are installed independently,which occupies the limited installation space.Then,less room is left for BESS and the maximum mileage is reduced.Besides,the SOC balancing ability of BMS is also reduced due to the control complexity and cost.The multi-function MMC-BESS for EV can solely provide the motor driving function,as well as AC and DC battery charging function,which makes it an attractive alternate with improved space utilization.Additionally,based on the flexible control ability of MMC-BESS,the SOC balancing performance of BMS is also improved.Along with the rapid popularization of EV,more retired battery cells will be produced in the near future.To reduce the cost of battery production and expand battery market,the research on second utilization of retired batteries is necessary.In EV,the batteries are retired with 70%-80%energies left,which is still sufficient for energy storage systems.The state of health(SOH)defines the healthy status of battery in terms of capacity degradation and it is vaired with different production engineering and working conditions.Until now,most of the control strategies for BESS set SOC balancing as the optimal control object.However,when the SOHs of batteries are quite different,the ones with lower SOHs will decay to the retired level first under traditional SOC balancing method and the life-span of BESS is also reduced.Thus,a novel control strategy is preferred to prolong the life-span of BESS with consideration of different SOHs.To deal with these issues illustrated above,this papar proposes an in depth analysis of the operational principle and control strategies for MMC-BESS by theoretical analysis,simulation verification and experiments tests.A general SOC balancing control method is proposed for MMC-BESS,both under balance and unbalance grid contidions.Avoiding the employment of so many PI controllers,a hybrid model predictive control(H-MPC)is proposed for MMC-BESS with reduced computation burden.An integrated power conversion system is proposed for EV with the application of MMC-BESS.The operational principles under differernt working modes are also analyzed.An SOH balancing control method is proposed for MMC-BESS with retired batteries utilized,which could prolong its life-span.The key technologies are summarized as follws:(1)A general SOC balancing control method is proposed for MMC-BESS.Based on the mathematical analysis of phase output power,specific components of circulating currents can be utilized to achieve SOC balancing.The proposed SOC balancing control method can be classified as three parts:phase SOC balancing,arm SOC balancing and individual SOC balancing.Different control objects are set for three control layers.A general SOC balancing method is proposed for MMC-BESS,both under balance and unbalance grid conditions,which makes it applicable for the practical engineering.(2)H-MPC is proposed for MMC-BESS.Firstly,the widely utilized MPC methods for MMC is introduced,which indicates the reduction of computation burden is one of the furture development trends.Traditionally,the SOC balancing in MMC-BESS is implemented based on many PI controllers,which might worsen the system performance consequently.To deal with this issue,an H-MPC is proposed.The H-MPC is comprised of two steps:PI control step and MPC control step.Setting different control objects for each step,the proposed H-MPC could achieve favorable system performance with reduced computation burden,making it possible to be employed in multi-level and high power applications.(3)An integrated power conversion system for EV.A multi-function modular multilevel conversion system for EV is proposed,which can solely provide the motor driving function,as well as AC and DC battery charging functions.The SOC balancing control strategies under different working modes(motor driving mode、DC charging mode and single-phase AC charging mode)are proposed based on the mathematical analysis of system circuits.Besides,an optimal group charging method is also presented,which could improve the system efficiency effectively.The capability of MMC-BESS to fully utilize the battery energies is analyzed by the evaluation of minimum SOH in terms of probability,which shows superior performance than the centralized connection case with two-level converters utilized.(4)An SOH balancing control method is proposed for MMC-BESS with retired batteries utilized.Taking full consideration of SOH difference effects on system’s life-span,a detailed quantitative analysis is made between SOH,depth of discharge(DOD)and life-cycles.As the on-line evaluation of SOH is hardly to be achieved,a relative SOH estimation method is proposed based on the detection of SOC.Furthermore,a SOH balancing method is proposed,which will prolong the life-span of MMC-BESS effectively.