Objective:To explore the optimization of tritium internal exposure monitoring and dose estimation for radiation workers at pressurized heavy water reactor nuclear power plants (PHWR NPPs).Methods:A total of 92 radiation workers at Candu-6 PHWR NPPs were observed to calculate the distribution of quenching index (tSIE). Ten samples with tSIE values ranging from 160 to 460 were selected to make the blank urine quenching curve. By comparing with the quenching curve accompined with the equipment, the optimization of correlationship of urine tritium measurement system was carried out. In estimating tritium internal radiation dose, the dose coefficient was optimized and adjusted by comparing the Canadian algorithm with the national standard algorithm. Dead time correction curve was drawn by means of heavy water simulation experiment diluting main heat transfer system.Results:Through the analysis and optimization of tritium internal dose estimation, the optimal dose coefficient was determined to be 4.8 and the quenching curve of blank urine sample was made. By comparison, it was found that the deviation was larger when tSIE<200, which confirmed the necessity of blank curve. When the tritium concentration in urine was higher than 2.4×10 4 Bq/ml, the urine sample needed to be diluted and then measured to correct the effect of death time. Conclusions:By optimizing both tritium internal exposure monitoring and dose estimation, not only the collective dose management can be improved, but also the unplanned tritium intake can be detected and treated in time, which plays an important role in ensuring the protection and safety of the radiation workers at PHWR NPPs.