arXiv subjects

Meiyi Li

Publications and source records attributed to Meiyi Li.

14 recordsLinked to original sources

arXiv2026

Beyond Scalar Flexibility: From Eligible AI Workloads to Dependable Load Relief

Grid studies often represent data-center flexibility as a fixed percentage of load, although no public production trace has shown how much eligible load persists across event durations or co-moves across clusters. We reconstruct 4,439 hourly power observations from a 185-day trace of 155,410 GPUs and derive a workload-semantic flexibility envelope. The fleet's time-averaged Monte Carlo median facility demand is 55.8 MW, while immediate eligible curtailment averages 3.55 MW after retaining allocated-GPU idle power: 12.1% of workload power and 6.35% of median facility power. Under full realization of that eligibility, 95%-available relief falls from 2.51 MW for one hour to 2.32 MW for four hours and 1.95 MW for 24 hours; a common realizable fraction q scales every value exactly by q. A mean-calibrated scalar overstates these quantities by 17%, 25%, and 47%, while a scalar tail-calibrated at four hours understates the one-hour product by 6% and overstates the 24-hour product by 17%; the share that reproduces the surface varies by a factor of 1.6 across durations and reliability levels. Aggregating 13 clusters raises four-hour firmness from 0.38 to 0.66, but cross-cluster covariance limits the gain. The production scheduler exposes almost no additional delay-based capacity: newly deferrable arrivals average 0.008 MW and have zero 95%-available capacity. These results replace an assumed flexibility percentage with duration, reliability, portfolio, and realizability terms that can be written into interconnection and demand-response contracts.

eess.SY↗

arXiv2025

Towards Reliable Neural Optimizers: Permutation-Equivariant Neural Approximation in Dynamic Data Driven Applications Systems

Dynamic Data Driven Applications Systems (DDDAS) motivate the development of optimization approaches capable of adapting to streaming, heterogeneous, and asynchronous data from sensor networks. Many established optimization solvers, such as branch-and-bound, gradient descent, and Newton-Raphson methods, rely on iterative algorithms whose step-by-step convergence makes them too slow for real-time, multi-sensor environments. In our recent work, we introduced LOOP-PE (Learning to Optimize the Optimization Process, Permutation Equivariance version), a feed-forward neural approximation model with an integrated feasibility recovery function. LOOP-PE processes inputs from a variable number of sensors in arbitrary order, making it robust to sensor dropout, communication delays, and system scaling. Its permutation-equivariant architecture ensures that reordering the input data reorders the corresponding dispatch decisions consistently, without retraining or pre-alignment. Feasibility is enforced via a generalized gauge map, guaranteeing that outputs satisfy physical and operational constraints. We illustrate the approach in a DDDAS-inspired case study of a Virtual Power Plant (VPP) managing multiple distributed generation agents (DERs) to maximize renewable utilization while respecting system limits. Results show that LOOP-PE produces near-optimal, feasible, and highly adaptable decisions under dynamic, unordered, and distributed sensing conditions, significantly outperforming iterative algorithm based solvers in both speed and flexibility. Here, we extend our earlier work by providing additional analysis and explanation of LOOP-PE design and operation, with particular emphasis on its feasibility guarantee and permutation equivariance feature.

eess.SY↗

arXiv2025

Impact of Data Poisoning Attacks on Feasibility and Optimality of Neural Power System Optimizers

The increased integration of clean yet stochastic energy resources and the growing number of extreme weather events are narrowing the decision-making window of power grid operators. This time constraint is fueling a plethora of research on Machine Learning-, or ML-, based optimization proxies. While finding a fast solution is appealing, the inherent vulnerabilities of the learning-based methods are hindering their adoption. One of these vulnerabilities is data poisoning attacks, which adds perturbations to ML training data, leading to incorrect decisions. The impact of poisoning attacks on learning-based power system optimizers have not been thoroughly studied, which creates a critical vulnerability. In this paper, we examine the impact of data poisoning attacks on ML-based optimization proxies that are used to solve the DC Optimal Power Flow problem. Specifically, we compare the resilience of three different methods-a penalty-based method, a post-repair approach, and a direct mapping approach-against the adverse effects of poisoning attacks. We will use the optimality and feasibility of these proxies as performance metrics. The insights of this work will establish a foundation for enhancing the resilience of neural power system optimizers.

cs.LG↗

arXiv2025

Learning to Optimize Joint Chance-constrained Power Dispatch Problems

The ever-increasing integration of stochastic renewable energy sources into power systems operation is making the supply-demand balance more challenging. While joint chance-constrained methods are equipped to model these complexities and uncertainties, solving these models using the traditional iterative solvers is time-consuming and can hinder real-time implementation. To overcome the shortcomings of today's solvers, we propose a fast, scalable, and explainable machine learning-based optimization proxy. Our solution, called Learning to Optimize the Optimization of Joint Chance-Constrained Problems (LOOP-JCCP), is iteration-free and solves the underlying problem in a single-shot. Our model uses a polyhedral reformulation of the original problem to manage constraint violations and ensure solution feasibility across various scenarios through customizable probability settings. To this end, we build on our recent deterministic solution (LOOP-LC 2.0) by incorporating a set aggregator module to handle uncertain sample sets of varying sizes and complexities. Our results verify the feasibility of our near-optimal solutions for joint chance-constrained power dispatch scenarios. Additionally, our feasibility guarantees increase the transparency and interpretability of our method, which is essential for operators to trust the outcomes. We showcase the effectiveness of our model in solving the stochastic energy management problem of Virtual Power Plants (VPPs). Our numerical findings complement our theoretical justifications and demonstrate great flexibility in parameter tuning, adaptability to diverse datasets, and increased computational speed.

eess.SY↗

arXiv2024

Balancing Passenger Transport and Power Distribution: A Distributed Dispatch Policy for Shared Autonomous Electric Vehicles

Shared autonomous electric vehicles can provide on-demand transportation for passengers while also interacting extensively with the electric distribution system. This interaction is especially beneficial after a disaster when the large battery capacity of the fleet can be used to restore critical electric loads. We develop a dispatch policy that balances the need to continue serving passengers (especially critical workers) and the ability to transfer energy across the network. The model predictive control policy tracks both passenger and energy flows and provides maximum passenger throughput if any policy can. The resulting mixed integer linear programming problem is difficult to solve for large-scale problems, so a distributed solution approach is developed to improve scalability, privacy, and resilience. We demonstrate that the proposed heuristic, based on the alternating direction method of multipliers, is effective in achieving near-optimal solutions quickly. The dispatch policy is examined in simulation to demonstrate the ability of vehicles to balance these competing objectives with benefits to both systems. Finally, we compare several dispatch behaviors, demonstrating the importance of including operational constraints and objectives from both the transportation and electric systems in the model.

eess.SY↗

arXiv2024

Speeding Ticket: Unveiling the Energy and Emission Burden of AI-Accelerated Distributed and Decentralized Power Dispatch Models

As the modern electrical grid shifts towards distributed systems, there is an increasing need for rapid decision-making tools. Artificial Intelligence (AI) and Machine Learning (ML) technologies are now pivotal in enhancing the efficiency of power dispatch operations, effectively overcoming the constraints of traditional optimization solvers with long computation times. However, this increased efficiency comes at a high environmental cost, escalating energy consumption and carbon emissions from computationally intensive AI/ML models. Despite their potential to transform power systems management, the environmental impact of these technologies often remains an overlooked aspect. This paper introduces the first comparison of energy demands across centralized, distributed, and decentralized ML-driven power dispatch models. We provide a detailed analysis of the energy and carbon footprint required for continuous operations on an IEEE 33 bus system, highlighting the critical trade-offs between operational efficiency and environmental sustainability. This study aims to guide future AI implementations in energy systems, ensuring they enhance not only efficiency but also prioritize ecological integrity.

eess.SY↗

arXiv2024

Towards Reliable Neural Optimizers: A Permutation Equivariant Neural Approximation for Information Processing Applications

The complexities of information processing across Dynamic Data Driven Applications Systems drive the development and adoption of Artificial Intelligence-based optimization solutions. Traditional solvers often suffer from slow response times and an inability to adapt swiftly to real-time input variations. To address these deficiencies, we will expand on our previous research in neural-based optimizers by introducing a machine learning-enabled neural approximation model called LOOP-PE (Learning to Optimize the Optimization Process -- Permutation Equivariance version). This model not only enhances decision-making efficiency but also dynamically adapts to variations of data collections from sensor networks. In this work, we focus on mitigating the heterogeneity issues of data collection from sensor networks, including sensor dropout and failures, communication delays, and the complexities involved in integrating new sensors during system scaling. The proposed LOOP-PE model specifically overcomes these issues with a unique structure that is permutation equivariant, allowing it to accommodate inputs from a varying number of sensors and directly linking these inputs to their optimal operational outputs. This design significantly boosts the system's flexibility and adaptability, especially in scenarios characterized by unordered, distributed, and asynchronous data collections. Moreover, our approach increases the robustness of decision-making by integrating physical constraints through the generalized gauge map method, which theoretically ensures the decisions' practical feasibility and operational viability under dynamic conditions. We use a DDDAS case study to demonstrate that LOOP-PE model reliably delivers near-optimal and adaptable solutions, significantly outperforming traditional methods in managing the complexities of multi-sensor environments for real-time deployments.

eess.SY↗

arXiv2023

Toward Rapid, Optimal, and Feasible Power Dispatch through Generalized Neural Mapping

The evolution towards a more distributed and interconnected grid necessitates large-scale decision-making within strict temporal constraints. Machine learning (ML) paradigms have demonstrated significant potential in improving the efficacy of optimization processes. However, the feasibility of solutions derived from ML models continues to pose challenges. It's imperative that ML models produce solutions that are attainable and realistic within the given system constraints of power systems. To address the feasibility issue and expedite the solution search process, we proposed LOOP-LC 2.0(Learning to Optimize the Optimization Process with Linear Constraints version 2.0) as a learning-based approach for solving the power dispatch problem. A notable advantage of the LOOP-LC 2.0 framework is its ability to ensure near-optimality and strict feasibility of solutions without depending on computationally intensive post-processing procedures, thus eliminating the need for iterative processes. At the heart of the LOOP-LC 2.0 model lies the newly proposed generalized gauge map method, capable of mapping any infeasible solution to a feasible point within the linearly-constrained domain. The proposed generalized gauge map method improves the traditional gauge map by exhibiting reduced sensitivity to input variances while increasing search speeds significantly. Utilizing the IEEE-200 test case as a benchmark, we demonstrate the effectiveness of the LOOP-LC 2.0 methodology, confirming its superior performance in terms of training speed, computational time, optimality, and solution feasibility compared to existing methodologies.

eess.SY↗

arXiv2023

Machine Learning Infused Distributed Optimization for Coordinating Virtual Power Plant Assets

Amid the increasing interest in the deployment of Distributed Energy Resources (DERs), the Virtual Power Plant (VPP) has emerged as a pivotal tool for aggregating diverse DERs and facilitating their participation in wholesale energy markets. These VPP deployments have been fueled by the Federal Energy Regulatory Commission's Order 2222, which makes DERs and VPPs competitive across market segments. However, the diversity and decentralized nature of DERs present significant challenges to the scalable coordination of VPP assets. To address efficiency and speed bottlenecks, this paper presents a novel machine learning-assisted distributed optimization to coordinate VPP assets. Our method, named LOOP-MAC(Learning to Optimize the Optimization Process for Multi-agent Coordination), adopts a multi-agent coordination perspective where each VPP agent manages multiple DERs and utilizes neural network approximators to expedite the solution search. The LOOP-MAC method employs a gauge map to guarantee strict compliance with local constraints, effectively reducing the need for additional post-processing steps. Our results highlight the advantages of LOOP-MAC, showcasing accelerated solution times per iteration and significantly reduced convergence times. The LOOP-MAC method outperforms conventional centralized and distributed optimization methods in optimization tasks that require repetitive and sequential execution.

cs.LG↗

arXiv2022

Learning to Solve Optimization Problems with Hard Linear Constraints

Constrained optimization problems appear in a wide variety of challenging real-world problems, where constraints often capture the physics of the underlying system. Classic methods for solving these problems rely on iterative algorithms that explore the feasible domain in the search for the best solution. These iterative methods are often the computational bottleneck in decision-making and adversely impact time-sensitive applications. Recently, neural approximators have shown promise as a replacement for the iterative solvers that can output the optimal solution in a single feed-forward providing rapid solutions to optimization problems. However, enforcing constraints through neural networks remains an open challenge. This paper develops a neural approximator that maps the inputs to an optimization problem with hard linear constraints to a feasible solution that is nearly optimal. Our proposed approach consists of four main steps: 1) reducing the original problem to optimization on a set of independent variables, 2) finding a gauge function that maps the infty-norm unit ball to the feasible set of the reduced problem, 3)learning a neural approximator that maps the optimization's inputs to an optimal point in the infty-norm unit ball, and 4) find the values of the dependent variables from the independent variable and recover the solution to the original problem. We can guarantee hard feasibility through this sequence of steps. Unlike the current learning-assisted solutions, our method is free of parameter-tuning and removes iterations altogether. We demonstrate the performance of our proposed method in quadratic programming in the context of the optimal power dispatch (critical to the resiliency of our electric grid) and a constrained non-convex optimization in the context of image registration problems.

eess.SY↗

arXiv2022

Teaching Networks to Solve Optimization Problems

Leveraging machine learning to facilitate the optimization process is an emerging field that holds the promise to bypass the fundamental computational bottleneck caused by classic iterative solvers in critical applications requiring near-real-time optimization. The majority of existing approaches focus on learning data-driven optimizers that lead to fewer iterations in solving an optimization. In this paper, we take a different approach and propose to replace the iterative solvers altogether with a trainable parametric set function, that outputs the optimal arguments/parameters of an optimization problem in a single feed forward. We denote our method as Learning to Optimize the Optimization Process (LOOP). We show the feasibility of learning such parametric (set) functions to solve various classic optimization problems including linear/nonlinear regression, principal component analysis, transport-based coreset, and quadratic programming in supply management applications. In addition, we propose two alternative approaches for learning such parametric functions, with and without a solver in the LOOP. Finally, through various numerical experiments, we show that the trained solvers could be orders of magnitude faster than the classic iterative solvers while providing near optimal solutions.

cs.LG↗

arXiv2022

Numerical Comparisons of Linear Power Flow Approximations: Optimality, Feasibility, and Computation Time

Linear approximations of the AC power flow equations are of great significance for the computational efficiency of large-scale optimal power flow (OPF) problems. Put differently, the feasibility of the obtained solution is essential for practical use cases of OPF. However, most studies focus on approximation error and come short of comprehensively studying the AC feasibility of different linear approximations of power flow. This paper discusses the merits of widely-used linear approximations of active power in OPF problems. The advantages and disadvantages of the linearized models are discussed with respect to four criteria; accuracy of the linear approximation, optimality, feasibility, and computation time. Each method is tested on five different systems.

eess.SY↗

arXiv2022

A Fully Decentralized Tuning-free Inexact Projection Method for P2P Energy Trading

Agent-based solutions lend themselves well to address privacy concerns and the computational scalability needs of future distributed electric grids and end-use energy exchanges. Decentralized decision-making methods are the key to enabling peer-to-peer energy trading between electricity prosumers. However, the performance of existing decentralized decision-making algorithms highly depends on the algorithmic design and hyperparameter tunings, limiting applicability. This paper aims to address this gap by proposing a decentralized inexact projection method that does not rely on parameter tuning or central coordination to achieve the optimal solution for Peer-to-Peer (P2P) energy trading problems. The proposed algorithm does not require parameter readjustments, and once tuned, it converges for a wide range of P2P setups. Moreover, each prosumer only needs to share limited information (i.e., updated coupled variable) with neighboring prosumers. The IEEE 13 bus test system is used to showcase our proposed method's robustness and privacy advantages.

eess.SY↗

arXiv2021

Virtual Inertia Control of the Virtual Synchronous Generator: A Review

With the increasing impact of low inertia due to the high penetration of distributed generation, virtual synchronous generator (VSG) technology has been proposed to improve the stability of the inverter-interfaced distributed generator by providing "virtual inertia". This paper presents a recent review of virtual inertia control covering significance, features, design principles, and state-of-art inertia strategies from both physical and mathematical perspectives to facilitate the wide application of the VSG. The definition and source of virtual inertia are given to help researchers to establish the concept of "virtual inertia". Then, this paper covers influencing mechanism studies of virtual inertia to reveal its functions. Also, a design framework of the virtual inertia is established by considering both the characteristics of the control system and the limitation of energy storage systems and renewable energy resources. Finally, several novel adaptive inertia control strategies are reviewed, and some aspects of potential future research are recommended.

eess.SY↗