In　this　study,　a　simulation-based　multi-objective　optimization　method　is　developed　for　optimizing　the　structural　design　of　hydrodynamic　cavitation　(HC)　reactor　and　improving　the　cavitation　effect　of　HC　reactor.　The　developed　method　integrates　simulation　technique　of　computational　fluid　dynamics　(CFD)　and　optimization　techniques　of　surrogate　model　and　nondominated　sorting　genetic　algorithm　II　(NSGA-II)　into　a　general　framework.　The　effect　of　structure　parameters　and　their　interactions　on　the　cavitation　effect　of　the　self-excited　oscillation　cavity　(SEOC)　are　analyzed.　Results　demonstrate　that　optimization　techniques　of　surrogate　model　and　NSGA-II　can　effectively　improve　the　structure　and　the　capacity　of　SEOC.　Simulation　results　show　that　the　internal　vapor　volume　fraction　and　outlet　vapor　volume　fraction　of　SEOC　(based　on　the　optimized　structure)　increase　by　13.46　and　38.01%,　respectively.　The　optimized　structure　of　SEOC　is　also　verified　experimentally　through　the　degradation　experiment　of　methylene　blue　solution.　The　degrees　of　degradation　before　and　after　optimization　respectively　are　10.12　and　16.14%,　and　the　degradation　capacity　increases　by　59.5%.　This　study　will　play　a　significantly　guiding　role　on　the　optimization　design　of　HC　reactor　for　advanced　oxidation　processes　(AOPs)　to　obtain　the　preferable　cavitation　effect.