In　view　of　affection　of　load　on　power　consumption　rate　in　electric　vehicle　distribution　process,　a　vehicle　routing　optimization　model　with　soft　time　window　was　formulated　to　minimize　the　comprehensive　cost,　such　as　fixed　cost,　driving　cost,　power　charging　cost　and　time　window　penalty　cost.　An　improved　hybrid　genetic　annealing　algorithm　was　designed　to　solve　the　problem,　and　then　the　power　consumption　rate　and　incomplete　charging　strategy　of　electric　vehicles　under　dynamic　load　were　discussed.　The　validity　of　the　model　and　algorithm　was　verified　by　taking　electric　vehicle　distribution　service　of　fresh　enterprise　as　an　example.　The　results　showed　that　an　incomplete　charging　strategy　had　significant　advantages　over　the　complete　charging　strategy　in　case　of　charging　time,　driving　distance　and　distribution　cost　under　dynamic　load　of　vehicles.　Compared　with　the　classical　genetic　algorithm,　the　improved　hybrid　genetic　annealing　algorithm　could　converge　to　optimal　solutions　quickly　and　effectively.　©　2023　CIMS.　All　rights　reserved.