The construction of large-scale wind power and photovoltaic bases (referred to as "grand base") focusing on deserts, the Gobi, and desert areas in China is accelerating, and the rate of their scale expansion continues to increase. In this context, how to effectively optimize the allocation of resources, and comprehensively enhance the operational efficiency of the integrated renewable energy system has become a core issue constraining the sustainable development of energy. This study proposes a seasonal dispatch optimization model that integrates machine learning and Single-Objective Optimization (SOO) to maximize system benefits while ensuring efficient resource utilization. The model is validated through a case study of a large-scale renewable energy project in Qinghai Province. The results show that there is a clear seasonal pattern in power generation: wind dominates in spring, summer, and winter, and solar and storage dominate in the autumn. During periods of low renewable energy production, energy storage provides the necessary support. Only in the autumn do storage systems need to be recharged at 16:00. For different load demands, wind energy reaches its maximum power before midday in spring and winter, while solar power generation peaks in the afternoon to evening hours in summer and autumn. The model constructed in this study was able to increase the average profit of the wind and solar energy storage system by 0.31 % in all seasons (in one day, low load scenario). The results of the study provide valuable guidance for optimizing energy system configuration and facilitating the transition to a sustainable energy system. The study helps to advance renewable energy integration and support the resilient development of power systems.