In the optimization scheduling of power systems containing renewable energy, to ensure that the scheduling results have higher security and better economy, this study proposes a multi-objective forecasting, scenario generation, and decision scheduling integrated stochastic optimal scheduling method. Firstly, to improve the accuracy and stability of wind-photovoltaic power forecasting, a novel multiobjective wind-photovoltaic forecasting model is proposed based on the Laguerre polynomial, pseudoinverse learning, and hybrid multi-objective Runge-Kutta algorithm (HMORUN). Secondly, to deal with wind-photovoltaic uncertainty, scenarios with representative wind-photovoltaic uncertainty characteristics are generated by scenario generation and reduction techniques using wind-photovoltaic power forecast results. Finally, considering wind-photovoltaic output fluctuations, the degree of source-load matching, wind-photovoltaic utilization, and system economic efficiency factors, with the objective of maximizing the tracking of the load curve and minimizing system economic costs, a stochastic optimized scheduling model for power systems is established. This study uses HMORUN as a solution tool for the multi-objective stochastic optimization scheduling problem (MOSSP). This study uses constraint repair techniques to deal with the complex constraints of the MOSSP model to avoid system load shedding and minimize wind and photovoltaic generation curtailment. To verify the effectiveness of the proposed model, a 10-generator power system, including a wind farm and a photovoltaic plant, is used as a test case for simulation experiments and compared with other multi-objective scheduling models. The experimental results show that the proposed stochastic optimal scheduling model has higher safety and better economy.INDEX TERMS Laguerre polynomial, Pseudo-inverse learning, Forecasting, Hybrid multi-objective Runge-Kutta algorithm, Stochastic optimal scheduling.