Abstract: Joint optimal operation for flood defense of reservoir group is an important non- option measures which can improve the flood control capacity of reservoir, and plays the role in regulating flood peak, storing floodwater, mitigating or even avoiding flood disasters. Taking the three large reservoirs (Chushandian, Nanwan and Suyahu) above Wangjiaba station in the main stream of Huaihe River as an example, taking the July 2020 flood as the input, changing its previous independent and regular regulation mode, the joint optimal regulation for flood control of reservoir groups was carried out with the aim of minimizing the maximum flood storage capacity occupied by three reservoirs under the premise of ensuring the safety of flood in the downstream. The corresponding model was established and solved by improved genetic algorithm. The results showed that, the improved genetic algorithm saved the time for the model to determine the penalty coefficients and made the model application more convenient. Compared with the regular regulation, optimal operation could ensure the safety of reservoirs and downstream control stations in the river basin, and the total flood storage capacity occupied by the three reservoirs could be reduced by 123 million m³.