Secure Control Regions for Distributed Stochastic Systems with Application to Distributed Energy Resource Dispatch

Abstract

With the increased adoption of distributed energy resources (DERs) in distribution networks, their coordinated control with a DER management system (DERMS) that provides grid services (e.g., voltage regulation, virtual power plant) is becoming more necessary. One particular type of DERMS using primal-dual control has recently been found to be very effective at providing multiple grid services among an aggregation of DERs; however, the main parameter, the primaldual step size, must be manually tuned for the DERMS to be effective, which can take a considerable amount of engineering time and labor. To this end, we design a simple method that self-tunes the step size(s) and adapts it to changing system conditions. Additionally, it gives the DER management operator the ability to prioritize among possibly competing grid services. We evaluate the automatic tuning method on a simulation model of a real-world feeder in Colorado with data obtained from an electric utility. Through a variety of scenarios, we demonstrate that the DERMS with automatically and adaptively tuned step sizes provides higher-quality grid services than a DERMS with a manually tuned step size.

Probabilistic Constraint Construction for Network-safe Load Coordination

“…= e nsν(δ−(1+δ) ln(1+δ)) . (18) We substitute c/ν, with c ∈ [0, 1 − ν], for δ and obtain Pr (M ns − ν ≥ c) ≤ e ns(c−(ν+c) ln(1+ c ν ))…”

“…Ref. [18] proposes an optimization problem to obtain a hyper-rectangular constraint set on the net power consumption of controllable DERs at each node in order to satisfy chance constraints on the voltage at each node. However, these approaches all require constraints to be applied at each node, rather than applying a constraint on aggregate power deviations by DERs located across a network.…”

“…Furthermore, in contrast to [20], we assume the aggregator has incomplete information about the TCLs to reduce communication requirements and preserve some level of privacy. Lastly, though some past work leveraged chance constraints to develop networksafe DER coordination approaches, e.g., [18], [21]- [26], these papers all assume that the controller has detailed distribution network information (enabling the formulation of a chanceconstrained optimal power flow problem), which is inconsistent with our utility-aggregator coordination framework.…”

Virtual Power Plants and Energy Justice

“…FERC Order 2222 allows VPPs to participate in wholesale markets, whereas DERs are usually not able to do so, which limits the grid services that they can provide to retail only. The ability to be compensated for grid services, which provide distributional justice benefits by improving electricity grid stability, reducing costs, and avoiding traditional infrastructure upgrades, is a key area that separates VPPs and DERMS from individual DERs from a grid operator's perspective (grid services are discussed in more detail in Section 3.1) [18].…”

DER Information Unaware Coordination via Day-ahead Dynamic Power Bounds

“…They do not consider transformer limits and do not fully explore the extension of their method to calculating bounds for multiple points in the grid simultaneously. Finally, the authors in [13] determine bounds for each node by solving a similar maximum volume box problem. In contrast to the work presented in this paper, however, their paper (i) only considers voltage constraints, (ii) the bounds are determined based only on the network model with no consideration of consumer loads (e.g., demand bounds), which can lead to unfairly allocating too much flexibility to nodes with smaller loads at the expense of nodes with larger loads, (iii) does not include any local controllers, (iv) has very limited simulation results, and (v) does not provide any performance evaluation relative to other DER coordination schemes.…”