Revealing　the　relations　among　parallel　mechanism　and　robot　comprehensive　performance,　topological　structure　and　dimension　is　the　basis　to　optimize　mechanism.　Due　to　the　correlation　and　diversity　of　the　single　performance　indexes,　statistical　principles　of　linear　dimension　reduction　and　nonlinear　dimension　reduction　were　introduced　into　comprehensive　performance　analysis　and　evaluation　for　typical　parallel　mechanisms　and　robots.　Then　the　mechanism＇s　topological　structure　and　dimension　with　the　best　comprehensive　performance　could　be　selected　based　on　principal　component　analysis　(PCA)　and　kernel　principal　component　analysis　(KPCA)　respectively.　Through　comparing　the　results,　KPCA　could　reveal　the　nonlinear　relationship　among　different　single　performance　indexes　to　provide　more　comprehensive　performance　evaluation　information　than　PCA,　and　indicate　the　numerical　calculation　relations　among　comprehensive　performance,　topological　structure　and　dimension　validly.