Based　on　the　rain　flow　algorithm,　a　modeling　method　to　predictive　the　state　of　health　(SOH)　of　the　LiCoO2　lithium-ion　battery　was　proposed　in　terms　of　the　complex　driving　conditions　of　the　electric　vehicles　(EV).　First,　a　real-time　rain　flow　algorithm　was　developed　to　count　the　cycling　numbers　according　to　different　depth　of　discharge　(DOD).　Based　on　the　enveloped　area　surrounding　by　the　equivalent　charging/discharging　currents　and　the　state　of　charge　(SOC),　a　degradation　factor　via　integrating　the　enveloped　area　was　proposed,　and　a　way　to　describe　the　battery　SOH　was　given　by　accumulating　the　cycling　numbers　of　different　degradation　factors.　Finally,　the　performance　of　the　proposed　battery　SOH　model　was　evaluated　through　experiments,　which　are　able　to　on-line　estimate　the　battery　SOH　even　under　the　cases　of　arbitrary　charging/discharging　currents,　DOD　and　cycle　numbers.　©　2017　Chin.　Soc.　for　Elec.　Eng.