In　order　to　better　deal　with　the　current　unstable　network　security　situation　and　improve　the　accuracy　and　generalization　ability　of　the　intrusion　detection　model,　this　paper　proposes　a　feature　fusion　technique　based　on　gradient　importance　enhancement,　which　combines　feature　fusion　and　feature　enhancement　to　increase　the　diversity　of　sample　features,　so　that　the　model　can　focus　more　on　the　sample　features　related　to　classification,　making　the　model　more　generalizable　and　improving　the　accuracy　of　the　model.　The　final　model　was　experimented　on　two　datasets,　NSL-KDD　and　CICIDS2017,　and　its　accuracy　reached　99.84%　and　99.78%,　respectively.