In　view　of　the　collaborative　optimization　and　control　of　various　kinds　of　energy　in　the　Energy　Internet,　a　collaborative　optimal　control　strategy　is　proposed　to　realize　large-scale　utilization　of　distributed　renewable　energy.　The　dynamic　multi-agent　coordination　framework　for　the　regional　Energy　Internet　is　designed.　The　upper　scheduling　aims　at　minimizing　the　integrated　cost　and　optimizes　the　day-ahead　scheduling　of　primary　energy　to　realize　energy　economy　and　safe　and　stable　operation　of　the　multi-energy　main　network.　The　two-level　control　strategy　based　on　the　consistency　theory　is　utilized　to　realize　the　precise　control　of　the　active　power　and　the　tie-line　power　of　distributed　generators　in　the　security　domain　of　voltage　and　frequency.　Also,　the　influence　of　communication　delay　is　discussed.　Finally,　the　experiment　verifies　the　effectiveness　of　the　proposed　strategy.　©　2017　Automation　of　Electric　Power　Systems　Press.