Due　to　the　multi-variable,　nonlinear,　large　time　delay　and　strong　coupling　features　of　the　wastewater　treatment　process,　a　recurrent　high-order　neural　network　is　used　to　model　the　key　water　quality　parameters(Chemical　Oxygen　Demand,　Biological　Oxygen　Demand,　Suspended　Solid　and　Ammonia　Nitrogen)　for　the　wastewater　treatment　process,　and　the　neural　network　is　trained　by　an　filtering　algorithm.　Operational　data　of　a　wastewater　treatment　plant　is　employed　to　illustrate　the　efficacy　of　the　proposed　modeling　method.　Meanwhile,　the　results　are　compared　with　feed-forward　neural　network　and　the　general　recurrent　neural　network　to　indicate　the　modeling　accuracy　of　the　recurrent　high-order　neural　network.　©　2011　ACADEMY　PUBLISHER.