Neural Networks for Encoding Dynamic Security-Constrained Optimal Power Flow

Murzakhanov, Ilgiz; Venzke, Andreas; Misyris, George S.; Chatzivasileiadis, Spyros

doi:10.48550/arxiv.2003.07939

Cited by 6 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With its ability to efficiently learn from complex datasets, provide universal function approximation, and generate predictive solutions on fast timescales, Machine Learning (ML) is emerging as one such tool which can help overcome intractability. Furthermore, emerging research [4], [5] is convincingly showing that embedding surrogate ML models directly inside safety-critical grid routines can transform intractable optimization problems into tractable ones. The recent increase of research on ML within the power systems community is enormous, as comprehensively reviewed by [6].…”

Section: Legacy System Emerging Systemmentioning

confidence: 99%

“…In order to solve an AC Unit Commitment problem, [4] replaces the AC power flow equations with a piecewise linear approximation learned by an optimally compact NN. In [5], a NN learns to classify regions of small-signal stability in order to solve a security constrained OPF problem. Input convex NNs, as popularized in [72], are used in [73] to learn a reactive power control law for optimal voltage regulation.…”

Section: E Embedding Neural Network In Optimization Problemsmentioning

confidence: 99%

See 1 more Smart Citation

Closing the Loop: A Framework for Trustworthy Machine Learning in Power Systems

Stiasny¹,

Chevalier²,

Nellikkath³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Deep decarbonization of the energy sector will require massive penetration of stochastic renewable energy resources and an enormous amount of grid asset coordination; this represents a challenging paradigm for the power system operators who are tasked with maintaining grid stability and security in the face of such changes. With its ability to learn from complex datasets and provide predictive solutions on fast timescales, machine learning (ML) is well-posed to help overcome these challenges as power systems transform in the coming decades. In this work, we outline five key challenges (dataset generation, data pre-processing, model training, model assessment, and model embedding) associated with building trustworthy ML models which learn from physics-based simulation data. We then demonstrate how linking together individual modules, each of which overcomes a respective challenge, at sequential stages in the machine learning pipeline can help enhance the overall performance of the training process. In particular, we implement methods that connect different elements of the learning pipeline through feedback, thus "closing the loop" between model training, performance assessments, and re-training. We demonstrate the effectiveness of this framework, its constituent modules, and its feedback connections by learning the N-1 small-signal stability margin associated with a detailed model of a proposed North Sea Wind Power Hub system.

show abstract

Section: Legacy System Emerging Systemmentioning

confidence: 99%

Section: E Embedding Neural Network In Optimization Problemsmentioning

confidence: 99%

Closing the Loop: A Framework for Trustworthy Machine Learning in Power Systems

Stiasny¹,

Chevalier²,

Nellikkath³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The integration of various machine learning models, e.g., neural networks, into optimization models may be done by reformulating the trained neural network as mixed-integer constraints [10], [20] or by directly differentiating the machine learning-informed optimization model, using first-order methods [11], [12]. The latter strategy, as the one employed in this paper, has the flexibility of integrating any NN architecture without the need of implementing its exact mixed-integer reformulation.…”

Section: Introductionmentioning

confidence: 99%

Capturing Electricity Market Dynamics in the Optimal Trading of Strategic Agents using Neural Network Constrained Optimization

Dolányi¹,

Bruninx²,

Toubeau³

et al. 2022

Preprint

View full text Add to dashboard Cite

In competitive electricity markets the optimal trading problem of an electricity market agent is commonly formulated as a bi-level program, and solved as mathematical program with equilibrium constraints (MPEC). In this paper, an alternative paradigm, labeled as mathematical program with neural network constraint (MPNNC), is developed to incorporate complex market dynamics in the optimal bidding strategy. This method uses input-convex neural networks (ICNNs) to represent the mapping between the upper-level (agent) decisions and the lower-level (market) outcomes, i.e., to replace the lower-level problem by a neural network. In a comparative analysis, the optimal bidding problem of a load agent is formulated via the proposed MPNNC and via the classical bi-level programming method, and compared against each other.

show abstract

“…Although the exact MILP reformulation of a NN is often used for verification purposes [8], [9], it can also be embedded within a larger optimization problem as a function approximation. This technique allows us to transform a problem that may have originally been a challenging mixed-integer nonlinear program (MINLP) into a more tractable MILP.…”

mentioning

confidence: 99%

“…Researchers have embedded NNs as MILPs within optimization problems for a variety of applications: [10]- [13]. Specifically in the field of power systems, [14] and [15] encode frequency constraints using the MILP reformulation in microgrid scheduling and UC problems, respectively, and [8] encodes security constraints into the OPF problem.…”

mentioning

confidence: 99%

Modeling the AC Power Flow Equations with Optimally Compact Neural Networks: Application to Unit Commitment

Kody¹,

Chevalier²,

Chatzivasileiadis³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Nonlinear power flow constraints render a variety of power system optimization problems computationally intractable. Emerging research shows, however, that the nonlinear AC power flow equations can be successfully modeled using Neural Networks (NNs). These NNs can be exactly transformed into Mixed Integer Linear Programs (MILPs) and embedded inside challenging optimization problems, thus replacing nonlinearities that are intractable for many applications with tractable piecewise linear approximations. Such approaches, though, suffer from an explosion of the number of binary variables needed to represent the NN. Accordingly, this paper develops a technique for training an "optimally compact" NN, i.e., one that can represent the power flow equations with a sufficiently high degree of accuracy while still maintaining a tractable number of binary variables. We show that the resulting NN model is more expressive than both the DC and linearized power flow approximations when embedded inside of a challenging optimization problem (i.e., the AC unit commitment problem). Index Terms-AC power flow, AC unit commitment (AC-UC), mixed-integer linear program (MILP), neural networks, piecewise linear model I. INTRODUCTIONThe AC power flow equations are routinely used to model network constraints in optimization problems related to the control, operation, and planning of power systems. These constraints, however, are both nonlinear and non-convex, resulting in optimization problems that can be NP-hard [1]. In practice, many power systems operation problems, like unit commitment (UC), optimal power flow (OPF), and optimal transmission switching (OTS), are solved using linearized approximations of the nonlinear AC power flow equations for the sake of computational tractability. However, linear approximations can result in suboptimal solutions or solutions that are infeasible in the original, nonlinear problem [2].Piecewise linear models offer a method of improving solution accuracy by capturing some of the nonlinearity of the † denotes an equal contribution among authors.

show abstract

Neural Networks for Encoding Dynamic Security-Constrained Optimal Power Flow

Cited by 6 publications

References 0 publications

Closing the Loop: A Framework for Trustworthy Machine Learning in Power Systems

Closing the Loop: A Framework for Trustworthy Machine Learning in Power Systems

Capturing Electricity Market Dynamics in the Optimal Trading of Strategic Agents using Neural Network Constrained Optimization

Modeling the AC Power Flow Equations with Optimally Compact Neural Networks: Application to Unit Commitment

Contact Info

Product

Resources

About