Focusing　on　performance　and　emissions　optimization,　a　novel　parallel　computing　optimization　platform　was　implemented　to　optimize　the　intake　and　exhaust　phases　of　a　hydrogen　Wankel　rotary　engine　(WRE).　An　improved　multi-objective　particle　swarm　algorithm　implemented　with　the　Sobol　sequence　was　introduced　in　this　study,　which　makes　it　superior　in　global　search.　A　one-dimensional　model　integrating　the　leakage　models　was　built　and　validated　under　various　excess　air　ratios.　The　parametric　control　variables　of　the　intake　and　exhaust　phases　were　defined　as　rise　stage,　main　stage,　and　decline　stage.　The　indicated　mean　effective　pressure　(IMEP),　indicated　specific　fuel　consumption　(ISFC),　and　nitrogen　oxide　(NOx)　were　used　as　evaluation　objectives.　The　optimization　results　showed　that　there　was　a　quadratic　relationship　between　ISFC　and　IMEP,　and　the　ISFC　decreased　with　increasing　IMEP.　The　relationship　between　NOx　and　IMEP　was　closer　to　linear,　and　the　NOx　increased　with　the　increase　of　IMEP.　The　timing　of　intake　port　full　closing　(IPFC)　contributed　the　most　influence　to　IMEP　and　NOx,　and　a　delayed　IPFC　resulted　in　a　lower　IMEP.　The　timing　of　exhaust　port　start　opening　(EPSO)　significantly　affected　the　ISFC,　and　an　earlier　EPSO　resulted　in　a　higher　ISFC.　In　the　optimal　case,　the　IMEP　was　increased　by　2.0%,　ISFC　was　reduced　by　1.1%,　and　NOx　was　only　increased　by　0.1%.　It　is　a　prospective　approach　to　further　improve　performance　and　emissions　simultaneously　using　parallel　computing　optimization　platform.