A Distributed Cloud-based Wide-Area Controller with SDN-Enabled Delay Optimization

The use of cloud computing for wide-area control of power systems using distributed computational resources has recently garnered significant attention. The upshot in such implementation, however, is that Synchrophasor data from one computational agent, commonly referred to as a virtual machine (VM), located in a certain location in the cloud need to be transmitted to VMs located in other locations through a shared communication network. Due to fluctuating bandwidth of these shared links, the wide-area control action can suffer from unacceptable delays, which may not only degrade closed-loop performance but can also threaten grid stability if they exceed a certain limit. This paper proposes the adoption of a simple latency control algorithm implemented in a software-defined network (SDN) connecting the VMs in the wide-area cloud by which delays in the control loop can always be optimized in order to retain their stability limit. The algorithm uses a greedy routing path selection based on active latency measurement for data transfer, and can be run periodically to pick the best available routes while the wide-area control loop is running.