PSI - Issue 57

Malik Spahic et al. / Procedia Structural Integrity 57 (2024) 833–847 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

836

4

Besides creep-fatigue damage due to cyclic operation, creep damage is accumulated as well during nominal operation. At nominal operation, the rotor is subject to mechanical stress due to centrifugal, torsional and bending loading. It should be noted that in practice these are much lower than the thermal stresses experienced during a start up or shut-down. These mechanical stresses cause primary creep damage which at least for the rotors within the ENGIE fleet is typically negligible at locations where high thermal stresses are present. Nevertheless, potential creep fatigue interaction should be accounted for. Power plants can have many different operating profiles, from base load with only a few start-stops a year over a few cold starts and weekend stops, to plants that have daily two start-stop cycles covering the morning and evening energy demand peaks. The number of start-stops as well as important load variations are increasing due to the energy transition and hence a proper risk assessment and monitoring is needed. One means of protecting the steam turbine rotor against excessive thermal stresses is the so-called Rotor Stress Evaluator (RSE). The RSE is typically installed within the Digital Control System (DCS) and continuously monitors the thermal stress of the rotor with a simplified one-dimensional model of the rotor. It then compares these stresses with the imposed material stress limits. If the margin with respect to the material stress limits becomes too small, the stress computer intervenes by reducing the load gradient, thereby reducing the rate at which the rotor is heated and hence as well the thermal stresses. A schematic overview of the RSE functioning is given in Figure 4. Many different types exist depending on the Original Equipment Manufacturer (OEM) of the steam turbine. Most types use a temperature probe measurement in the stator casing as a reference for the rotor temperature, although this is not always very representative. Sometimes corrections are being implemented to address this issue when using the probe as an input. Another solution, adopted as well by ENGIE Laborelec on different units, is to use a thermodynamic calculation of the rotor surface temperature based on the steam pressure and temperature at inlet and outlet of the steam turbine along with the control valves’ position and rotational speed of the rotor.

Figure 4: Schematic overview of the RSE functioning based on a simplified 1D model of the rotor