Accurate　and　real-time　prediction　of　water　quality　not　only　helps　to　assess　the　environ-mental　quality　of　water,　but　also　effectively　prevents　and　controls　water　quality　emergen-cies.　In　recent　years,　neural　networks　represented　by　Bidirectional　Long　Short-Term　Memory　(BiLSTM)　and　Encoder-Decoder　(ED)　frameworks　have　been　shown　to　be　suitable　for　prediction　of　time　series　data.　However,　traditional　statistical　methods　cannot　capture　nonlinear　characteristics　of　the　water　quality,　and　deep　learning　models　often　suffer　from　gradient　disappearance　and　gradient　explosion　problems.　This　work　proposes　a　hybrid　water　quality　prediction　method　called　SVABEG,　which　combines　a　Savitzky-Golay　(SG)　fil-ter,　Variational　Mode　Decomposition　(VMD),　an　Attention　mechanism,　BiLSTM,　an　ED　structure,　and　a　hybrid　algorithm　called　Genetic　Simulated　annealing-based　Particle　Swarm　Optimization　(GSPSO).　SVABEG　first　adopts　the　SG　filter　and　VMD　to　remove　noise　and　deal　with　nonlinear　features　in　the　original　time　series,　respectively.　Then,　SVABEG　combines　BiLSTM,　the　ED　structure　and　the　attention　mechanism　to　capture　bi-directional　long-term　correlations,　realize　dimensionality　reduction　and　extract　key　infor-mation,　respectively.　Furthermore,　SVABEG　adopts　GSPSO　to　optimize　its　hyperparameters.　Experimental　results　with　real-life　datasets　demonstrate　that　the　proposed　SVABEG　out-performs　current　state-of-the-art　algorithms　in　terms　of　prediction　accuracy.　(c)　2023　Elsevier　Inc.　All　rights　reserved.