PSI - Issue 57

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

834

2

The energy transition has forced the conventional gas and steam turbine power plants into more flexible operation with an increased number of starts and stops, which is not what they were initially designed for. This cyclic operation has accelerated the lifetime consumption of the steam turbine rotors, causing thermo-mechanical fatigue damage of the material leading in the end to rotor cracking [1]. In the last couple of years, an increasing number of steam turbine rotors has been found cracked [1]. A major incident also occurred within the ENGIE fleet in Europe where a crack half-way through a steam turbine rotor was found. Fortunately, the vibration protection system forced a trip of the unit before a catastrophic failure could take place. The main damage was however done causing unavailability of the unit for a full year. The associated costs for business interruption and repair in this case easily exceeded 10M€. They can rise however beyond 100M€, mainly due to business interruption when larger delays are present if repair is not deemed possible and a new rotor is needed. More recently, a new incident occurred on an ENGIE unit in South-America enforcing again unexpected downtime and repair. This unit was of a completely different design which shows that this type of cracking is not design specific, but a more general concern for steam turbine rotors. Clearly, a more wide fleet risk management approach is required as many assets have been accumulating a large number of starts, many of them reaching or having exceeded 1000 starts. Different parameters have been considered to obtain a first high-level risk assessment. For units considered at risk, custom finite element models of the steam turbine rotors were developed to identify the critical locations on the rotor and their lifetime consumption. The underlying damage assessment is based on the experience with the aforementioned cracked steam turbine rotors, which presented a unique opportunity for the ENGIE Laborelec experts to calibrate their damage prediction tool. Overall, the ENGIE fleet within Europe with around 25 units of many different designs and operating conditions provides an excellent benchmark for this matter. Based on the lifetime consumption, further actions are recommended to mitigate the risk for rotor cracking. In the remainder of this paper, a theoretical background firstly describes all the concepts used further in the paper. In a second part, both cracked steam turbine rotors are further detailed. Then the main parameters of the fleet risk management approach are presented. Finally, some risk mitigation actions are given which include assessment of the lifetime consumption with the damage prediction tool, reduction of the operational gradients and machining of the rotor surface.

Nomenclature RSE

R otor Stress Evaluator

DCS Digital Control System OEM Original Equipment Manufacturer CCGT Combined Cycle Gas Turbine HP High Pressure HIP High and Intermediate Pressure ST Steam Turbine EOH Equivalent Operational Hours HTC Heat Transfer Coefficient LCF Low Cycle Fatigue

2. Theoretical background During a start-up, as steam enters the steam turbine, the outer surface of the rotor is heated up. Due to the thickness of the rotor and the corresponding thermal inertia, the steam’s heat is only slowly transferred in the radial direction towards the rotor center. Consequently, temperature differences are created within the rotor body as shown in Figure 1. These temperature differences cause compressive stresses on the rotor surface and tensile stresses in the rotor center. During shut-down, when cooling down the rotor, tensile stresses are created at the rotor ’s surface, while the rotor center is in compression.