The　increasing　municipal　solid　waste　has　become　one　of　the　stubborn　diseases　of　modern　society.　Incineration　technology　with　the　advantage　of　reduction　and　resource　utilization　is　an　important　avenue　to　solve　this　problem.　Toxic　gases　emitted　from　incineration　not　only　pollute　the　environment　but　also　pose　a　serious　threat　to　human　health.　However,　the　complexity　of　incineration　process　brings　great　difficulty　to　the　detection　of　pollutants　concentration　in　the　exhaust　gases.　In　order　to　address　the　problem　that　the　concentration　of　pollutants　is　difficult　to　predict,　a　prediction　model　of　dioxin　emission　concentration　based　on　stochastic　configuration　network　is　developed　in　this　paper.　The　prediction　model　takes　the　online　operation　data　of　municipal　solid　waste　incineration　power　plant　as　input　parameters　and　realizes　the　online　prediction　of　dioxin　emission　concentration.　It　provides　a　scientific　basis　for　choosing　the　appropriate　control　strategy.　When　compared　with　BP　neural　network,　RBF　neural　network　and　support　vector　machine　prediction　model,　it　is　easily　found　that　the　proposed　prediction　model　can　effectively　improve　the　prediction　accuracy　of　dioxin　emission　concentration　during　the　municipal　solid　waste　incineration　process.　©　2020　IEEE.