A　kind　of　dissolved　oxygen(DO)　control　method　based　on　growing　neural　network(GNN)　was　proposed　due　to　the　wastewater　treatment　process　is　nonlinear,　strong　coupling,　great　legging　and　so　on.　This　controller　can　control　the　change　of　object　character　and　adjust　the　structure　of　neural　network　automatically.　Based　on　the　growing　conditions,　the　neural　network　can　increase　the　hidden　layer　nodes　to　improve　the　ability　of　the　controller.　The　results　show　that　the　proposed　method　is　more　effective　on　the　overshoot,　the　adjusting　time　and　the　self-adoptive　ability.