Data-driven Security-Constrained AC-OPF for Operations and Markets

Abstract

In this paper, we propose a data-driven preventive security-constrained AC optimal power flow (SC-OPF), which ensures small-signal stability and N-1 security. Our approach can be used by both system and market operators for optimizing redispatch or AC based market-clearing auctions. We derive decision trees from large datasets of operating points, which capture all security requirements and allow to define tractable decision rules that are implemented in the SC-OPF using mixedinteger nonlinear programming (MINLP). We propose a secondorder cone relaxation for the non-convex MINLP, which allows us to translate the non-convex and possibly disjoint feasible space of secure system operation to a convex mixed-integer OPF formulation. Our case study shows that the proposed approach increases the feasible space represented in the SC-OPF compared to conventional methods, can identify the global optimum as opposed to tested MINLP solvers and significantly reduces computation time due to a decreased problem size.

Recent Developments in Machine Learning for Energy Systems Reliability Management

“…In [115], the authors propose a preventive control scheme by rescheduling generating units. First, they assess the transient stability of the system [34], [38], [115] Neural networks [116] Tree-based methods [117]- [121] Linear models [122] with a hybrid method based on a SVM model and time-domain simulation. Then, if the system is unstable, they compute from the SVM model the sensitivity of each generator to a transient stability assessment index (derived from the SVM model) to rank the generators and select the ones that are more effective for improving the stability.…”

“…The decision trees rules consist in conditional line transfer limits, that can be embedded in the SCOPF in order for the operator to take decisions already in line with the small-signal stability margin. An extension of this work to solve an AC-SCOPF instead of a DC-SCOPF is proposed in [121], while still incorporating N-1 security and small-signal stability with decision tree-learnt rules. In [116], the authors are solving an OPF considering transient stability constraints.…”

DeepOPF: Deep Neural Network for DC Optimal Power Flow

“…In [22], [23], decision trees are used to derive tractable rules from large data sets of operating points, which can efficiently represent the feasible region and identify possible solutions. However, the proposed heuristic schemes are still iteration based and may still incur a significant amount of running time for large-scale instances.…”

Convex Relaxations and Approximations of Chance-Constrained AC-OPF Problems

“…Citation information: DOI 10.1109/TPWRS.2018.2874072, IEEE Transactions on Power Systems xi consideration of integer variables and large uncertainty ranges in the proposed framework. Furthermore, we are planning to use a combination of data-driven methods from our work in [40], [41], convex relaxations, and the iterative solution framework to develop a scalable approach to an integrated security-and chanceconstrained OPF. In [40], [41] we propose a novel approach which efficiently incorporates N-1 and stability considerations in an optimization framework and is suitable for integration in the proposed CC-SOC-OPF framework.…”