The performance of fixed ejector is limited by working conditions, so it is difficult to meet the operating requirements of electric vehicle heat pump system under varying conditions and wide temperature range. In this paper, a transcritical CO2 refrigeration cycle with dual ejectors in parallel (DEP) for electric vehicle heat pump system is established and its operation condition weight statistics method in all climate is proposed. On this basis, a fixed ejector optimization method based on the genetic algorithm is proposed, which takes the increment of integrated part load value (ΔIPLV) compared to the conventional transcritical CO2 refrigeration cycle (CON) system as the evaluation index. The performance analysis results show that under the cooling conditions, the coefficient of performance (COP) of DEP system with the optimized ejector for cooling is increased by 17.32%–23.42% compared to the CON system, and the COP of the transcritical CO2 refrigeration cycle with a single ejector (SEJ) system is increased by 7.31%–9.47%. In the heating mode, the COP of DEP system with the optimized ejector for heating is increased by 18%–19.79%, while the COP of SEJ system with the unoptimized ejector is decreased by 0.07%–2.43%. •An electric vehicle CO2 heat pump system with dual ejectors in parallel is investigated.•The system operation condition weight statistics method in all climate is proposed.•A fixed ejector optimization method based on the genetic algorithm is proposed.•The performance of the system with optimized ejector is effectively improved.