The phase equilibrium of gas hydrate is the important foundation for hydrate research and accurate equilibrium data are obtained mainly by experiments. In order to improve the accuracy of hydrate equilibrium, the research effort focused on measurement technology itself. Firstly, intensive variables under three-phase equilibrium were analyzed in terms of the Gibbs phase rule. Then the multi-step heating dissociation method was investigated in detail after a brief review of measurement technologies. No leakage and gas saturation must be ensured in sample preparation stage. During data processing, the endpoint of complete hydrate dissociation usually is regarded as phase equilibrium condition, which corresponds to the largest pores in sediment, initial concentration of solution and initial gas composition. The experimental results indicate that CH4 hydrate phase equilibrium in 0.0003 mass fraction Sodium dodecyl sulfate (SDS) solution is not changed, but the "climbing wall effect" changes CH4 hydrate equilibrium in silica sand saturated by SDS solution. Hydrate phase equilibrium of 0.180 mol fraction CO2-N-2 mixture in 0.004 mol fraction tetrahydrofuran (THF) solution is changed obviously. Additionally, the accuracy of CO2 hydrate phase equilibrium using heating rate of 0.2 K/h is higher than using 0.6 K/20 min which shows that a low heating rate can improve the accuracy. The study also indicates that isolated CO2 hydrate in 0.009 mol fraction NaCl solution is different from that of bulk hydrate. Suggestions to control heating rate and judge isolated hydrate are also proposed. (C) 2015 Elsevier B.V. All rights reserved.