Due　to　the　uncertainty　of　wind　power,　the　intra-day　market　is　introduced　for　the　system　with　large-scale　wind　power　integration,　to　reduce　the　power　deviation　in　a　real-time　market　and　promote　the　economy　and　stability　of　power　system　operation.　By　using　the　interval　model　to　express　the　uncertainty　of　wind　power　output,　this　paper　proposes　a　joint　optimization　model　of　day-ahead　and　simulative　intra-day　markets　with　consideration　of　uncertainty　of　wind　power　deviation　of　the　intra-day　day　market　in　the　day-ahead　market.　Similarly,　a　joint　optimization　model　of　intra-day　and　simulative　real-time　markets　is　built　with　consideration　of　uncertainty　of　wind　power　variation　of　the　related　real-time　market　in　each　intra-day　market.　Considering　the　regret　psychology　of　power　purchase　decision-makers　in　day-ahead　and　intra-day　markets,　the　two-stage　clearing　schedules　for　minimizing　the　robust　regret　are　obtained　according　to　different　time　scales.　Thus,　the　economy　and　the　robustness　of　the　power　purchase　scheme　in　day-ahead　and　intra-day　markets　are　improved.　Finally,　a　case　is　studied　and　analyzed　in　detail　to　verify　the　feasibility　and　superiority　of　the　proposed　model.　©　2019　Automation　of　Electric　Power　Systems　Press.