The construction of the burnup lib determines the accuracy of burnup and decay heat calculations. The evaluation of burnup information in the nuclear lib is complex, leading to a large, rigid, and inefficient burnup matrix. This paper begins with the basic composition of the burnup lib, considering the impact of each nuclide and its transformation relationships on the accuracy of neutronics calculations and target nuclide nuclear density calculations, which serves as the basis for the compression of the burnup lib. To address the decay heat calculation accuracy loss caused by the compression of fission products, a nonlinear least squares optimization algorithm is used to fit the decay heat release function, and pseudo-decay nuclides are constructed to replace the fission product decay heat calculation, thereby maintaining the accuracy of decay heat calculations. Verification results show that the original detailed burnup lib contains more than 1 500 nuclides, which are reduced to fewer than 200 nuclides after compression. The compressed burnup lib does not introduce significant deviations in the calculation of the effective multiplication factor and nuclear density. In terms of decay heat calculations, the pseudo-decay nuclides significantly restore the decay heat calculation accuracy, with the contribution of decay heat to total power having a calculation deviation of less than 0.5%, meeting the required accuracy for decay heat calculations.