(113b) Fouling early detection in ethylene plant Steam Turbines

The petrochemical industry is facing more and more stringent requirements in terms of process reliability and maintenance costs. Thus, companies are in the need for advanced technologies to monitor the operation of critical process equipment.

The phenomenon of fouling builds up gradually with operation and represents one of the major issues affecting turbomachinery in charge gas and refrigerant trains. Steam Turbines fouling, in particular, is caused by poor steam purity and typically results in a reduced steam swallowing capacity due to the increase of pressure in machine section at constant steam flow.

To enable an early detection of this phenomenon the variation of the Stodola constant (K_ST) calculated across machine wheel chamber and exhaust sections is compared with K_ST resulting from the machine design tool calculation at same steam conditions. The long term monitoring of this parameter can be applied to prevent severe machine damage and to suggest the timing for machine cleaning procedures.

The results of this comparison on a specific use-case will be presented in detail in this paper.