To　improve　the　dynamic　response　characteristics　of　permanent　magnet　synchronous　motor　(PMSM)　servo　system,　an　inertia　identification　method　based　on　the　theory　of　model　reference　adaptive　system　(MRAS)　has　been　researched.　A　novel　inertia　identification　method　using　genetic　algorithm　(GA)　is　proposed　for　requirements　of　rapid　convergence　and　high　precision.　This　method　takes　advantage　of　the　global　search　capability　of　GA,　taking　deviation　between　actual　angular　velocity　and　estimated　angular　velocity　as　control　error,　utilizing　integral　of　time　multiplied　absolute　value　of　error　(ITAE)　as　optimized　target,　dynamically　adjusting　adaptive　gain　of　MRAS,　to　achieve　optimization　of　control　parameters　on-line.　Experimental　results　proved　that　this　method　has　a　faster　convergence　rate　and　a　higher　precision　in　inertia　identification,　confirmed　effectiveness　and　feasibility　of　this　method.