To　accurately　predict　the　key　parameters　of　the　municipal　solid　waste　incineration　(MSWI)　process,　this　paper　proposes　an　improved　case-based　reasoning　(CBR)　predictive　modeling　method　based　on　a　deep　Q　network　to　realize　the　case　adaptation　process.　First,　the　MSWI　operation　process　is　analyzed　to　screen　out　the　relevant　feature　variables　and　build　the　corresponding　case　base.　Second,　the　K-nearest　neighbor　(KNN)　algorithm　is　used　to　realize　the　case　retrieval　process　of　the　parameter　prediction,　and　cases　similar　to　the　current　incineration　state　are　obtained.　Then,　based　on　the　＇Learning-Evaluation-Revision＇idea,　the　case　difference　adaptation　knowledge　between　similar　cases　and　the　feature　variables　of　the　current　state　is　learned　through　the　deep　Q　network　to　realize　key　parameter　prediction.　Finally,　the　actual　data　of　a　solid　waste　incineration　plant　are　used　to　predict　the　key　parameters　of　the　furnace　temperature　and　flue　gas　oxygen　content.　The　results　show　that　the　proposed　method　can　accurately　predict　the　MSWI　process　parameters.　©　2024　IEEE.