Swarm-diversity　is　an　important　factor　influencing　the　global　convergence　of　particle　swarm　optimization　(PSO).　In　order　to　overcome　the　premature　convergence,　the　paper　introduced　a　negative　feedback　mechanism　into　particle　swarm　optimization　and　developed　an　adaptive　PSO.　The　improved　method　takes　advantage　of　the　swarm-diversity　to　control　the　tuning　of　the　inertia　weight　(PSO-DCIW),　which　in　turn　can　adjust　the　swarm-diversity　adaptively　and　contribute　to　a　successful　global　search.　The　proposed　PSO-DCIW　was　applied　to　some　well-known　benchmarks　and　compared　with　the　other　notable　improved　methods　for　PSO.　The　relative　experimental　results　show　PSO-DCIW　is　a　robust　global　optimization　method　for　the　complex　multimodal　functions,　which　can　improve　the　performance　of　the　standard　PSO　and　alleviate　the　premature　convergence　validly.