Most of the world’s power is produced by large steam turbines using fossil fuel, nuclear and geothermal energy.
The LP exhaust hoods of these turbines are known to contribute significantly to the losses within the turbine, hence a minor improvement in their performance, which results in a lower backpressure and thus higher enthalpy drop for the steam turbine, will give a considerable benefit in terms of fuel efficiency.
Understanding the flow field and the loss mechanisms within the exhaust hood of LP steam turbines is key to developing better optimized exhaust hood systems. A detailed analysis of loss generation within the exhaust hood was done by the authors . It was found that most losses occur at the upper hood and are caused by the swirling flows, which mostly start at the diffuser outlet, especially for the top diffuser inlet sector flows that have a complex path to the condenser. The authors further numerically investigated the influence of hood height variation on performance of an LP turbine exhaust hood , which further contributed to the knowledge of the loss mechanisms.
With the loss mechanisms in exhaust hoods reasonably well understood, flow deflection at the upper hood is investigated in the current paper. The deflection is aimed at minimizing the intensity of the vortices formed thus reducing the exhaust losses. The deflector configurations analyzed are modifications of the walls of the reference configuration’s outer casing. The numerical models of the reference configuration which are based on a scaled axial-radial diffuser test rig operated by ITSM have already been validated by the authors at design and overload operating conditions and three tip jet Mach numbers (0, 0.4 and 1.2).
Deflector configurations investigated are found to re-direct the flow at the upper hood and minimize the intensity of the swirling flows hence leading to improvement in performance of LP steam turbine exhaust hoods. The best performing deflector configuration is found to give a considerable improvement in performance of 20% at design load and 40% at overload both at tip jet Mach number of 0.4 (corresponding to shrouded last stage blades). At design load and tip jet Mach number of 1.2 (corresponding to unshrouded last stage blades), the improvement is found to be moderate. About 7% performance increase is observed.