Total Pressure Ratio Discrepancy (NASA Stage 35 2-D)

[safdarianebi]

Hi folks,

I am simulating flow in span 50% of NASA Stage 35 compressor. My simulation is 2 dimensional. My configuration file details can be seen below. I have tried different meshes and regardless of the mesh details I get similar results when solution converges. Everything is reasonable for me, but the calculated total pressure ratio is 1.35, which is very smaller than that for 3D analysis, which is 1.68.

Do you think its ok or something is wrong?

I am using unstructured mesh and if you think it is required I can give more info about mesh details.

            # ------------- DIRECT, ADJOINT, AND LINEARIZED PROBLEM DEFINITION ------------%
            
            "SOLVER": "RANS",
            "KIND_TURB_MODEL": "SST",
            "MATH_PROBLEM": "DIRECT",
            "RESTART_SOL": "NO",
            "SYSTEM_MEASUREMENTS": "SI",
            
            # -------------------- COMPRESSIBLE FREE-STREAM DEFINITION --------------------%
            
            "MACH_NUMBER": params.inlet_total_state.mach_number,
            "AOA": params.angle_of_attack,
            "SIDESLIP_ANGLE": 0.0,
            "FREESTREAM_PRESSURE": params.inlet_total_state.P,
            "FREESTREAM_TEMPERATURE": params.inlet_total_state.T,
            "FREESTREAM_DENSITY": params.inlet_total_state.rho_mass(),
            "INIT_OPTION": "TD_CONDITIONS",
            "FREESTREAM_OPTION": "TEMPERATURE_FS",
            "FREESTREAM_TURBULENCEINTENSITY": 0.05,                          # standard for turbomachinery
            "FREESTREAM_TURB2LAMVISCRATIO": 100.0,                           # standard for turbomachinery
            
            # ---------------------- REFERENCE VALUE DEFINITION ---------------------------%
            
            "REF_ORIGIN_MOMENT_X": 0.00,
            "REF_ORIGIN_MOMENT_Y": 0.00,
            "REF_ORIGIN_MOMENT_Z": 0.00,
            "REF_LENGTH": 1.0,
            "REF_AREA": 1.0,
            "REF_DIMENSIONALIZATION": "DIMENSIONAL",
            
            # ------------------------------ EQUATION OF STATE ----------------------------%
            
            "FLUID_MODEL": "IDEAL_GAS",
            "GAMMA_VALUE": params.inlet_total_state.gamma,
            "GAS_CONSTANT": params.inlet_total_state.gas_constant,
            
            # ---------------------- VISCOSITY AND CONDUCTIVITY MODEL ---------------------%
            
            "VISCOSITY_MODEL": "SUTHERLAND",
            "MU_REF": 1.716E-5,
            "MU_T_REF": 273.15,
            "SUTHERLAND_CONSTANT": 110.4,
            "CONDUCTIVITY_MODEL": "CONSTANT_PRANDTL",
            
            # ---------------------------------- MIXING PLANE -----------------------------%
            
            "AVERAGE_PROCESS_KIND": "MIXEDOUT",
            "TURBULENT_MIXINGPLANE": "YES",
            "PERFORMANCE_AVERAGE_PROCESS_KIND": "MIXEDOUT",
            "MIXEDOUT_COEFF": "(1.0, 1.0E-05, 15)",
            "AVERAGE_MACH_LIMIT": 0.05,
            
            # ------------------------ LINEAR SOLVER DEFINITION ---------------------------%

            "LINEAR_SOLVER": "FGMRES",
            "LINEAR_SOLVER_PREC": "LU_SGS",
            "LINEAR_SOLVER_ERROR": 1E-6,
            "LINEAR_SOLVER_ITER": 10,
            
            # ----------------------- SLOPE LIMITER DEFINITION ----------------------------%
            
            "VENKAT_LIMITER_COEFF": 0.05,
            "LIMITER_ITER": 999999,
            
             # -------------------- TURBULENT NUMERICAL METHOD DEFINITION ------------------%
            
            "CONV_NUM_METHOD_TURB": "SCALAR_UPWIND",
            "MUSCL_TURB": "NO",
            "SLOPE_LIMITER_TURB": "VENKATAKRISHNAN",
            "TIME_DISCRE_TURB": "EULER_IMPLICIT",
            "CFL_REDUCTION_TURB": 1.0,
            
            # -------------------- FLOW NUMERICAL METHOD DEFINITION -----------------------%
        
            "CONV_NUM_METHOD_FLOW": "JST",                       # Convective numerical method
            "MUSCL_FLOW": "NO",                                  # Monotone Upstream-centered Schemes for Conservation Laws
            "SLOPE_LIMITER_FLOW": "VAN_ALBADA_EDGE",
            "ENTROPY_FIX_COEFF": 0.001,                          # higher this the less accurate, 0.1 already pretty high, usually do 0.05
            "JST_SENSOR_COEFF": (0.5, 0.02),
            "TIME_DISCRE_FLOW": "EULER_IMPLICIT",
            
            # ------------- COMMON PARAMETERS DEFINING THE NUMERICAL METHOD ---------------%
            
            "NUM_METHOD_GRAD": "WEIGHTED_LEAST_SQUARES",
            "CFL_NUMBER": 10.0,
            "CFL_ADAPT": "NO",
            "CFL_ADAPT_PARAM": "( 0.3, 0.5, 1.0, 1000.0)",
            
            # --------------------------- CONVERGENCE PARAMETERS --------------------------%
            
            "OUTER_ITER": 1000,
            "CONV_RESIDUAL_MINVAL": -10,
            "CONV_STARTITER": 10,
            "CONV_CAUCHY_ELEMS": 100,
            "CONV_CAUCHY_EPS": 1E-6,

            # --------------------------- BOUNDARY CONDITIONS -----------------------------%
            
            "SPATIAL_FOURIER": "NO",
            "MULTIZONE": "YES",
            "CONFIG_LIST": CONFIG_LIST,
            "MARKER_INLET": MARKER_INLET,
            "MARKER_HEATFLUX": MARKER_HEATFLUX,
            "MARKER_PERIODIC": MARKER_PERIODIC,
            "MARKER_TURBOMACHINERY": MARKER_TURBOMACHINERY,
            "MARKER_MIXINGPLANE_INTERFACE": MARKER_MIXINGPLANE_INTERFACE,
            "MARKER_ZONE_INTERFACE": MARKER_ZONE_INTERFACE,
            "MARKER_GILES": MARKER_GILES,
            "TURBOMACHINERY_KIND": TURBOMACHINERY_KIND,
            "TURBO_PERF_KIND": TURBO_PERF_KIND,
            "MARKER_PLOTTING": MARKER_PLOTTING,
            "MARKER_MONITORING": MARKER_PLOTTING,
            "GILES_EXTRA_RELAXFACTOR": "( 1.0, 1.0)",
            "MARKER_ANALYZE": f"( {self.Stages[-1].outflow_passage.mesh_params.outlet_label} )",
            "RAMP_OUTLET_PRESSURE": "NO", # YES can help with convergence
            "RAMP_OUTLET_PRESSURE_COEFF": f"(80000, 10.0, 4000)",

[bigfooted]

So you did 2D and 3D simulations with SU2 and got 1.35 and 1.68 for the pressure ratios, using the same physics parameters? How does it compare to measurements? It does not sound very wrong to me, in 2D you neglect any changes in the 3rd direction, and the presence of walls. so it's normal to have a lower pressure in 2D I think. But a comparison with measurements and possibly CFD results from other groups with other codes can show if your SU2 results are what is to be expected.

[bigfooted]

So if you have a picture showing this comparison, please post it, that would give a lot of insights...