Continuous, safe and reliable operation of the hydrogenerator is possible only within the defined P-Q diagram, which is the key document of each hydrogenerator. Limits are defined within P-Q diagram, and some of them are isotherms that represent the maximum allowed heating of different parts of the hydrogenerator, such as the stator winding, rotor winding, end region of the stator core in capacitative mode of operation, etc. Reliable information about the temperature of the rotor winding is of great importance, but traditionally, no sensors for temperature measurements are placed on the rotor due to various problems such as: large centrifugal forces, strong electromagnetic fields, problems with the power supply of the measuring system, etc. Instead, monitoring of the average value of the rotor winding temperature is usually performed using the indirect U-I method. The paper presents two independent methods for determining the rotor temperature of the hydrogenerator, an indirect and the direct one, and describes their advantages and disadvantages. Models are presented for their mutual comparison in order to analyze the possibilities of each individual method and to improve the temperature monitoring of the hydrogenerator rotor in order to increase the reliability of the whole hydroaggregate.