Reconstruction of core inlet temperature distribution by cold leg temperature measurements

The reduced core of Loviisa NPP contains 33 thermocouple measurements measuring the core inlet temperature. Currently, these thermocouple measurements are not used in determining the inlet temperature distribution. The average of cold leg temperature measurements is used as inlet temperature for each fuel assembly. In practice, the inlet temperature distribution is not constant. Thus, using a constant inlet temperature distribution induces asymmetries in the measured core power distribution. Using a more realistic inlet temperature distribution would help us to reduce virtual asymmetries of the core power distribution and increase the thermal margins of the core. The thermocouples at the inlet cannot be used directly to measure the inlet temperature accurately because the calibration of the thermocouples that is done at hot zero power conditions is no longer valid at full power, when there is temperature change across the core region. This is due to the effect of neutron irradiation on the Seebeck coefficient of the thermocouple wires. Therefore, we investigate in this paper a method to determine the inlet temperature distribution based on the cold leg temperature measurements. With this method we rely on the assumption that although the core inlet thermocouple measurements do not measure the absolute temperature accurately they do measure temperature changes with sufficient accuracy particularly in big disturbances. During the yearly testing of steam generator safety valves we observe a large temperature increase up to 12 degrees in the cold leg temperature. The change in the temperature of one of the cold legs causes a local disturbance in the core inlet temperature distribution. Using the temperature changes observed in the inlet thermocouple measurements we are able to fit six core inlet temperature response functions, one for each cold leg. The value of a function at an assembly inlet is determined only by the corresponding cold leg temperature disturbance. The linear combination of the six response functions gives the inlet temperature distribution. In this paper the reconstruction method is described and the effect of inlet temperature distribution on core power distribution is studied. (Authors)