Hi – you need to specify the total pressure and temperature at the inlet, and from what you are saying this has been done correctly, there could be a number of reason your outlet and inlet mass is to matching, first of all you need to monitor your convergence has your flow residual dropped by at least 2 or 3 order of magnitude, if not, where is the maximum change is there a meshing issue there, or any separation occurring at the outlet etc,.. Also please quote what percentage change in mass you have observed. finally you should besetting your BC in the Relative frame of the fan, not absloute. If you have access to CFX also try it more suitable for blade flows.

Shahrokh Shahpar Thanks for your answer! My case has no fan in the flowfield. I just add two boundary surfaces at the inlet and outlet of the fan. The main flowfield is the flow over the aircraft.I give the static pressure and target mass flow at the fan inlet,as it is the out let of the external flowfield. Then I give the total pressure and the total temperature at the fan-outlet, as it is the inlet of the external flowfield. Other boundaries of the external flowfield are wall and pressure farfield. my case has converged. but the fan-inlet mass flow rate is 30% less than the fan-outlet. The discrepancy of this part mass flow rate is balanced by the far field mass flow rate.So, the result is not real in physical. By the way, I can't set BC in the Relative frame, because there is no fan-zone(inner flow) in the flowfield.

OK, understood, I think you are trying to represent the fan by an actuator disk type (or body forces) model where you have no fan geometry but the total pressure is raised by the work done on the fluid (turning or not turning downstream as you can also assume some OGV take the swirl out). In this model to be realistic you also need to include a loss model to account for the Total Pressure at the inlet and outlet may not be the same, but this is not the reason for your mass discrepancy – I agree 30% is a very large value, have you tried running your set up with a back pressure which is not far from the inlet mass pressure at the face of the engine, this one should run with almost zero perturbation to the flow, going through your AD.ie no pressure rise or VJet either.

Sorry, I can't understand the operation advise about the "back pressure", I upload a section of my mesh, and point out the BC's location. I think my problem is that I set a target mass flow rate for the fan-inlet, but there is no constraint for the fan-outlet's mass flow rate(there is no blank for this variable in the "pressure inlet" BC type of fluent)

OK, I understand better, it seems you have to do two things, either:

1) change the Inlet to Exit BC and Exit to Inlet BC - as these are your Exit (on the LHS) to your computational mesh you are showing and Vice versa. ie on the LHS this beomes an Exit BC with a Specified mass and on the RHS it becomes the Inlet with BC with the same mass - hence I am sure your mass would be correct then, and the aformentioend BCs will ensure the surroundijng stream lines are drawn to the Engine, almost like a real engine.

2) If you still want a jump in Pressure (that happens in a real engine) then in Fluent you can creat a Zone 2 inside your square box where your engine is to produce a Delta P (you need to independantly mesh this).

Hope this is usefil, Please try and let me know if it works for you?!