Transmission　error　is　a　key　indicator　to　characterize　the　performance　of　precision　reducer.　The　test　result　of　transmission　error　will　be　influenced　by　input　speed　and　load　torque.　In　this　study,　a　morphological　construction　method　of　transmission　error　of　precision　reducer　based　on　BP　neural　network　was　proposed　to　obtain　the　morphology　of　transmission　error　under　different　input　speeds　and　load　torques.　Then　the　transmission　accuracy　of　precision　reducer　could　be　accurately　evaluated.　With　RV-40E　reducer　as　an　example,　the　morphological　structure　of　its　transmission　error　was　obtained.　The　experimental　results　demonstrate　that　the　model　established　by　the　proposed　method　can　reflect　the　morphology　of　transmission　error.　©　2022　IEEE.