As one of new long-term energy storage technology, the vanadium redox flow battery (VRFB) is suitable for peak-shaving, frequency modulation and new energy support, etc. However, there are issues such as complex power to capacity ratio, larger occupied area, and lower energy efficiency. Based on solving the problem of abandoning of wind, solar, and power limit in new energy power generation, this paper forwards a theoretical method for the optimal design of energy storage capacity configuration by establishing a mathematical model and optimizing the objective function. By optimizing the topology design of energy storage station (ESS), system integration and reliability of power access are improved. By establishing an energy recovery mathematical model, a method for improving system efficiency through energy recovery based thermal management design is proposed. By establishing a multi-source data mix digital twinning model, the idea of VRFB energy storage operation and maintenance technology based on digital twinning is proposed. Finally, the simulation analysis verifies the feasibility of the optimized design. The peak shaving rate can be reached to 16%, and the power limit is reduced from 45% to 14%. Meanwhile, the occupied area can be saved by 15%, and the efficiency can be increased from 70% to about 90%. It provides a systematic theoretical method and feasible solution for the large-scale VRFB ESS optimization design for new energy.