Overspray and Interstage Fog Cooling in Compressor Using Stage-Stacking Scheme – Part 2: Case Study

A stage-by-stage wet-compression theory and algorithm have been developed for overspray and interstage fogging in the compressor. This theory and algorithm are used to calculate the performance of an 8-stage compressor under both dry and wet compressions. A 2D compressor airfoil geometry and stage setting at the mean radius are employed. Six different cases with and without overspray have been investigated and compared. The stage pressure ratio enhances during all fogging cases as does the overall pressure ratio, with saturated fogging (no overspray) achieving the highest pressure ratio. Saturated fogging reduces specific compressor work, but increases the total compressor power due to increased mass flow rate. The results of overspray and interstage spray unexpectedly show that both the specific and overall compressor power do not reduce but actually increase. Analysis shows this increased power is contributed by increased pressure ratio and, for interstage overspray, "recompression" contributes to more power consumption. Also it is unexpected to see that air density actually decreases, instead of increases, inside the compressor with overspray. Analysis shows that overspray induces an excessive reduction of temperature that leads to an appreciable reduction of pressure, so the increment of density due to reduced temperature is less than decrement of air density affected by reduced pressure as air follows the polytropic relationship. In contrast, saturated fogging results in increased density as expected. After the interstage spray, the local blade loading immediately showed a significantly increase. Fogging increases axial velocity, flow coefficient, blade inlet velocity, incidence angle, and tangential component of velocity. The analysis also assesses the use of an average shape factor in the generalized compressor stage performance curve when the compressor stage information and performance map are not available. The result indicates that using a constant shape factor might not be adequate because the compressor performance map may have changed with wet compression. The results of non-stagestacking simulation are shown to underpredict the compressor power by about 6% and net GT output by about 2% in the studied cases. NOMENCLATURE CET Compressor Exit Temperature (K) P Pressure (kPa) Pc Compressor Power (kW) Pnet Net GT Output Power (kW) Pt Turbine Power (kW) R Gas constant (kJ/kg-K) RH Relative humidity rp Compressor pressure ratio SF Shape factor T Temperature (K) U Tangential velocity V Inlet velocity Va Axial velocity W Relative velocity Wc Compressor Specific Work (kJ/kg) WNet Net Specific Work (kJ/kg) Wt Turbine Specific Work (kJ/kg) Greek φ Flow coefficient (≡Va/U) γ Specific heat ratio ρ Density (kg/m) ηC Overall compressor efficiency (by Eq. 3) ηp Polytropic (or small-stage) efficiency ηCS Overall isentropic compressor efficiency (by Eq. 4) ψ Rotor work coefficient (≡ Δh0/1⁄2U) Subscripts 1, 2,.. Stage numbers a Dry air fraction c Compressor i Rotor Stage i+0.5 Stator Stage t Turbine Superscripts * Ratio of the off-design value over the design value INTRODUCTION When dealing with gas turbine inlet fogging, previous researchers [e.g. 1-3] treated compressor as a single unit and developed the wet compression theory employing only thermodynamic analysis. Under this approach, the interstage fogging cannot be included in the analysis. More complex analyses was then undertaken by researchers [4-11] employing both thermodynamic, heat transfer, and aerodynamic theories through each compressor stage with the help of general performance curves to estimate the compressor aerodynamic performance. Wanting to avoid the uncertainty involved in determining the shape factor when using the general compressor performance curve, a stage-stacking scheme for wet-compression theory has been developed in Part 1 [12] to analyze both inlet overspray and interstage fogging in the compressor. The associated algorithm is integrated into the in-house computational code FogGT [13] to calculate the stage-by-stage compressor performance and the overall gas turbine system performance. In this paper, a case study is