Abstract: When the Monte Carlo method is used to simulate a large-scale fission reactor, the power density of different regions varies greatly. Compared with the high-power region, the number of particles in the low-power region is relatively insufficient, and the statistical error is also larger. For this kind of Monte Carlo transport calculation problem with global characteristics, in order to obtain accurate statistics of neutron flux in the whole space, the global weight window (GWW) method and the uniform fission site (UFS) method were studied in this paper, and the flux-based GWW function was developed in RMC. The GWW function and two UFS functions were applied to the Hoogenboom-Martin PWR whole core benchmark, which achieves a better global variance reduction effect, reduces the asymmetry of the reactor core calculation and greatly improves the calculation efficiency. In order to combine the advantages of the two kinds of variance reduction algorithm, the combined algorithm of GWW and UFS was developed in the Monte Carlo code RMC and applied to the critical calculation of the PWR. The results show that under the same calculation conditions, with adoption of the combined algorithm, the variance of the periphery of reactor core is further reduced and the calculation efficiency is greatly improved, which verifies the effectiveness of the method in the critical calculation of the whole core.