Introduction (problem statement and relevance). At present, the majority of automobile manufacturers have organized mass production of plug-in hybrid vehicles, i.e. PHEV’s. This solution allows driving using electric traction only for a long time due to an increased capacity battery which can be charged from both an external source and the internal combustion engine (ICE) as well as recuperate the brake energy. In the near future, the demand for such power plants will show significant increase, especially in urban environment. During electric driving, the ICE may not be started for a long time, hence the canister will not be purged. In this regard, studying of evaporation generation processes and finding solutions for their minimization are relevant.The purpose of the study is to develop approaches for the fuel tank isolation valve (FTIV) control algorithm depending on the fuel volume in the vehicle fuel tank.Methodology and research methods. The fuel vapor generation in the fuel tank vapor space has been analyzed for an open evaporation system, in which the vapor volume is constantly connected to the canister, and for a closed system, where the vapor volume and canister are separated by the fuel tank isolation valve. The influence of change in temperature, composition and amount of the air-fuel mixture in the vapor space on the fuel tank pressure has been determined.Scientific novelty and results. The equations for determination of the vapor mass in the vapor space with the closed fuel tank evaporation system (Mv), for the vapor mass during the canister feed (ΔMcv), total heating of the mixture up to a certain temperature depending on the number of canister feed steps (Δt), for the air mass in the vapor space of the closed fuel tank evaporation system depending on the number of canister feed steps (Man) have been obtained.Practical significance. The equations have been obtained allowing determination of the individual and total canister feed during fuel heating within the given temperature interval. The dependencies obtained allow forming optimal algorithms to control the canister feed by varying the pressure release depth depending on the fuel properties and vapor space volume with successive steps of the canister feed and temperature intervals between them.