The　accumulative　generation　operator　of　traditional　DGM(1,　1)　model　does　not　consider　that　data　volatility　will　affect　the　data　development　trend.　This　limitation　makes　a　homogenous　exponential　growth　prediction　model　difficult　to　simulate　stochastic　oscillation　sequence.　To　solve　this　problem,　this　paper　proposes　a　stochastic　oscillation　analysis　method　based　on　the　mean　value　sequence　of　raw　data,　and　design　a　new　accumulative　operator　using　trigonometric　function　values　during　its　weight　generation.　Based　on　this　accumulative　operator,　a　new　DGM(1,　1)atan　model　with　variable　weights　is　provided.　Finally,　this　model　is　used　to　forecast　data　in　field　of　marine　water　quality　monitoring.　Result　shows　the　effectiveness　and　practicability　of　the　proposed　model.　©　2017,　Editorial　Board　of　Journal　of　Systems　Engineering　Society　of　China.　All　right　reserved.