The　activated　sludge　wastewater　treatment　process　is　a　high　nonlinear　and　time-varying　system.　The　water　quality　changes　violently,　and　the　states　have　strong　coupling　between　each　other.　As　these　problems,　the　paper　proposes　an　adaptive　fuzzy　neural　network　controller.　The　results　of　simulation　show　this　controller　can　adjust　membership　function　on-line,　optimize　control　rules,　remove　contamination　quickly　and　has　great　robustness　when　applied　to　activated　sludge　system.　This　study　provides　a　new　approach　to　the　research　of　wastewater　treatment　process.