Waste　water　treatment　is　related　to　environmental　protection　and　human　health.　Aiming　at　the　difficulty　of　accurate　prediction　of　effluent　TP　in　wastewater　treatment　plants　(WWTPs),　a　multi-scale　data-driven　method　for　effluent　total　phosphate　(TP)　prediction　is　proposed　in　this　paper.　Firstly,　the　time-scale　characteristics　of　water　quality　data　were　analyzed.　Secondly,　a　weighted　average　data　processing　(WAP)　method　is　designed　to　reconstruct　the　data　set　for　multi-scale　data.　By　matching　the　time　scales　of　different　variables,　the　effective　data　quantity　is　expanded　and　the　prediction　frequency　is　improved.　Then,　the　prediction　model　of　effluent　TP　based　on　echo　state　network　(ESN)　was　constructed　to　achieve　accurate　prediction　of　effluent　TP.　Finally,　the　proposed　method　is　applied　to　the　actual　effluent　data　set.　Experimental　results　show　that　the　proposed　method　can　accurately　predict　effluent　TP.　©　2023,　The　Author(s),　under　exclusive　license　to　Springer　Nature　Singapore　Pte　Ltd.