There is a problem of uncertain optimization in an active distribution network in the cooperation of source, load and storage. Thus an active distribution network operational strategy based on an affinely adjustable robust optimization is proposed. First, the controllable model of distributed PV, battery energy storage model with cycle life degradation and the ZIP load model are considered. Then the optimization model of the active distribution network is constructed. Then, an ellipsoid set is used to describe the uncertainty of distributed PV output, and the linear decision-making mechanism of decision variables and uncertainty variables of distributed PV output in the optimization model is formed by combining with an affine strategy. Finally, the robust optimization model is transformed into a deterministic mixed integer second-order cone programming model by dual transformation. The effectiveness and advantages of the proposed affinely adjustable robust optimization method are compared and verified in the modified IEEE-33 node system.