Abstract: Dynamically reconfigurable batteries can achieve intrinsically safe operation through flexible topological reconfiguration by an internal power electronics network. However, the real-time and independent reconfiguration of the battery network inside each energy storage unit leads to disparate operating states and inconsistent constraint boundaries among the massive energy storage units in the corresponding large-scale energy storage power stations, which significantly increases the operation and maintenance costs of the power stations. To address this issue, this paper proposes a dynamic power optimization control method that fully accounts for the operating states and multidimensional dynamic boundary constraints of the massive energy storage units in large-scale dynamically reconfigurable battery energy storage stations. The method fully takes into account the constraint boundaries of energy storage units after each dynamic reconfiguration, and establishes a system operation and maintenance optimization model incorporating key factors such as the degradation cost, temperature regulation cost, operation and maintenance cost of dynamically reconfigurable batteries, and the state of charge balance. In addition, a multi-variate fixed weight method is applied to determine the optimization weights of each factor. Finally, case study simulations demonstrate that the proposed method can significantly improve the operation and maintenance economy of the power station while ensuring the independent operation constraints of the massive energy storage units in the dynamically reconfigurable battery energy storage power station, and simultaneously enhance the consistency of their state parameters.