The　operation　of　coal-fired　power　plants　emits　a　significant　amount　of　SO2,　causing　a　considerable　environmental　challenge.　Wet　flue　gas　desulfurization　(WFGD)　is　the　mostly　used　desulfurization　method　in　current　practice　for　address　this　issue,　however,　the　simultaneous　optimization　of　desulfurization　efficiency　and　cost　remains　challenging.　Aiming　to　achieve　optimal　performance　and　reduce　the　computational　complexity,　this　paper　proposes　a　novel　multi-objective　non-dominated　sorting　beluga　whale　optimization　(NSBWO)　algorithm　to　concurrently　optimize　desulfurization　efficiency　and　cost.　Derived　from　the　NSGA-II　framework,　the　NSBWO　algorithm　adopts　the　beluga　whale　algorithm　to　generate　offspring　instead　of　the　conventional　genetic　algorithm　used　in　the　original　algorithm.　This　modification　alleviates　the　concerns　about　performance　dependence　on　user-defined　parameters,　enhancing　its　performance　in　multi-objective　optimization.　Experimental　results　demonstrate　that,　compared　to　classical　algorithms,　NSBWO　exhibits　superior　overall　performance　and　faster　convergence　speed.　Finally,　the　algorithm　is　applied　to　real-world　data　sourced　from　the　flue　gas　desulfurization　system　of　a　1000MW　coal-fired　power　plant　unit.　The　obtained　results　validate　the　algorithm＇s　effectiveness　in　optimizing　both　the　efficiency　and　cost　of　the　wet　flue　gas　desulfurization　system.　　©　2024　IEEE.