A grid-scale battery energy storage station usually contains multiple battery containers and corresponding electric links. Each link and battery container could become a controllable subsystem when receiving network operators' orders. Battery containers will perform differently after long-term use, especially considering a mixed utilization of new and recycled batteries. Currently the storage inflow/outflow power is distributed equally among containers. The existing power allocation and control strategy in battery energy storage stations mainly focus on batteries' capacity constraint, rather than their performance, temperature, and aging conditions. This paper proposed a novel power allocation approach for multiple battery containers in a battery energy storage station considering batteries' state of charge, temperature, and potential aging caused. This method established the mathematical modeling between battery aging with state of charge, depth of discharge, operating temperature and charging state transition times, thus the potential battery aging can be accurately calculated. The battery aging is used as the optimization objective to ensure the total battery aging occurred in all containers is minimized when executing every power dispatch order. A methodological framework is conducted based on a pure-simulation case study to evaluate the potential benefits and aging risks of different power allocation strategies. The simulation results showed this novel method helps to balance the aging, temperature, and state of charge between battery containers, leading to a more reliable, and secure operation of battery energy storage station.