As　the　merging　section　of　an　urban　expressway　main　road　and　on-ramp　traffic　flow,　on-ramp　areas　have　serious　merging　conflicts.　In　the　dynamic　game　between　two　streams　of　traffic　flow,　traffic　congestion　can　occur　and　cause　bottlenecks　on　expressways.　With　the　development　of　vehicle　intelligence　and　network　technology,　connected　vehicles　can　form　a　platoon　or　a　group　and　realize　real-time　dynamic　communication.　In　this　paper,　the　vehicle　group　collaboration　behaviors　of　the　main　road　and　ramps　in　an　on-ramp　area　are　investigated　with　a　connected　vehicle　environment,　and　the　vehicle　group　model　based　on　the　theory　of　acceptable　gaps　is　constructed.　Using　VISSIM　simulation　software,　simulation　scenarios　are　built　based　on　time　slice　and　virtual　signal　control　strategies,　and　then,　the　proposed　schemes　are　simulated　and　evaluated.　By　continuous　optimization,　the　ideal　time　difference　between　the　control　ramp　and　the　main　road　can　be　determined　to　form　a　better　vehicle　group　length　and　acceptable　gap,　comprehensively　utilize　the　limited　road　resources,　improve　the　utilization　rate　of　on-ramp　areas　and　reduce　delay.　The　experimental　results　show　that　the　combined　implementation　of　the　mainline　and　ramp　control　strategies　is　more　effective　than　that　of　only　the　mainline　strategy.　When　the　strategy　cycle　is　large　(more　than　3　min),　the　time　slice　simulation　strategy　is　more　effective　than　the　virtual　signal　strategy　for　improving　the　traffic　efficiency　of　on-ramp　areas.　If　the　virtual　signal　strategy　is　adopted,　a　suitable　signal　cycle　should　be　selected.　The　performance　effect　of　vehicle　generation　schemes　under　different　background　traffic　congestion　is　also　discussed.　Finally,　the　effect　of　micro　car-following　parameters　(different　levels　of　driving　behaviors)　on　the　platoon　operation　is　evaluated　in　this　paper.