A　newly　introduced　Self-adaptive　Differential　Evolution　algorithm　via　Generalized　Opposition-Based　Learning　(SDE-GOBL)　is　applied　to　optimal　design　of　two　sewer　networks.　Every　chromosome　consists　of　the　information　of　network　layout.　Select　a　feasible　design　which　satisfies　the　constraints　of　velocity,　slope　and　proportional　water　depth　to　get　optimal　cost　through　the　algorithm.　Two　sewer　optimization　problems　in　which　the　pipe　diameters　are　considered　as　the　decision　variables　are　solved　by　the　SDE-GOBL　algorithm.　Comparisons　with　the　previous　works　are　made　and　the　results　show　that　the　proposed　algorithm　performs　better　in　terms　of　solution　quality　and　efficiency.