To support the high proportion grid interconnection of renewable energy power generation such as wind power and photovoltaic, this paper integrates the wind power plant (WPP), photovoltaic power station (PV), convention gas turbine (CGT) and pumped storage power station (PHSP) as the multi-power Hybrid System (MPHS), and establishes a WPP-PV-CGT-PHSP stochastic scheduling optimization model based on robust stochastic optimization theory. First, this paper introduces the basic structure of MPHS and establishes a multi-power output model. Then, based on the optimization goal of system operation economic benefit maximization and output power fluctuation minimization, the MPHS scheduling operation multi-objective model is established. Simultaneously, to overcome the uncertainty impact on the system, this paper presents an uncertainty characterization method of wind power and photovoltaic based on predictive power, transforms the constraints containing uncertainty variables into the constraints reflecting the decision makers’ risk attitude, and then forms MPHS stochastic scheduling optimization model. Finally, WPP with 9.6 MW installed capacity, PV with 6.5 WM installed capacity, 3 CGT and 1 upper reservoir with 10 MW·h equivalent capacity are taken as the simulation system. The simulation results show as follow: 1) MPHS can make full use of the power generation complementary characteristics of different energy, and achieve optimal overall system operation results. 2) PHSP can optimize and adjust the operational behavior to provide greater capacity for wind power and photovoltaic grid connection. 3) The introduction of robust coefficients can provide scheduling decision-making tool for different risk-preference decision makers, minimize the operational risk during chasing the maximum economic benefits.