A cooperative distributed algorithm for handling reactive power saturation in active distribution networks

In Active Distribution Networks (ADNs), voltage regulation can be achieved by local controllers acting on the reactive power injected by Distributed Energy Resources (DERs). An optimization algorithm at the secondary control level evaluates the voltage set points for the local controllers at the primary level. However, the continuous variations of the ADN operating conditions may cause saturation of the reactive powers, which causes the integral wind-up phenomenon of the local PI regulators, leading to degraded performance, as well as the inability to achieve the optimal voltage profile. To manage these issues, the present paper proposes a cooperative strategy among DERs implemented by a distributed algorithm in which all DERs cooperate to solve the saturation. The distributed algorithm locally calculates the variation of the set point of each DER controller to pursue two objectives: avoiding the wind-up phenomenon and achieving a new voltage profile closer to the optimal one. The algorithm uses fixed average weights and does not depend on the control techniques used to design the PI regulators. Only voltage errors are exchanged among DERs, and the communication requirements are limited. The results of accurate numerical simulations give evidence of the validity of the proposed cooperative distributed algorithm.