In order to explore the effects of non-axisymmetric endwall on counter-rotating compressor

taking the rotors of a counter-rotating compressor as research object

based on artificial neural network and genetic algorithm

the optimization of non-axisymmetric endwall contouring was carried out successively on rotor 1(R1) and rotor 2(R2) in multistage environment. The changes of flow field and performance before and after optimization were compared and analyzed. The purpose was to further improve the performance of the counter-rotating compressor with implementation of non-axisymmetric endwall contouring. The results show that on the optimization condition point

the non-axisymmetric endwall changes the static pressure distribution near rotors' hub endwall

decreases the circumferential static pressure gradient

and reduces the intensity of secondary flow;meanwhile it also changes the flow capacity along the radial span of rotors

and improves the efficiency of upper area;finally the efficiency of the counter-rotating compressor is increased by 0.78%.