arXiv subjects

Yihong Zhou

Publications and source records attributed to Yihong Zhou.

12 recordsLinked to original sources

arXiv2026

Lifetime Profit-Maximising Co-optimisation of Multi-Service Stacking for Battery Storage

Grid-scale battery energy storage can generate revenue by stacking services across electricity and frequency response markets, yet identifying the lifetime profit-maximising stacking strategy remains challenging. Decisions across services are coupled through shared battery system capacity, constrained by system operator energy management rules, and further shaped by product-specific technical requirements that govern system operation, degradation, and lifetime profitability. This paper presents an ageing-aware receding-horizon framework for co-optimising multi-service stacking that explicitly captures product-specific characteristics and state-of-energy compliance rules. The framework is applied to the Great Britain market, where storage operators can stack electricity trading with multiple dynamic frequency response services procured through the newly introduced 'Enduring Auction Capability' platform under energy management requirements imposed by the National Energy System Operator. Using real market data, we demonstrate that degradation modelling, discount rate, and battery ageing jointly govern both lifetime value and optimal stacking strategy. Accounting for ageing increases lifetime revenue by up to 32% relative to a degradation-agnostic benchmark, while higher-fidelity ageing modelling can provide a further revenue improvement of up to 16% over simpler formulations. Lower discount rates favour strategies that balance calendar and cycling ageing, while higher discount rates favour aggressive operation. Across all strategies, services responding to positive frequency deviations are consistently preferred over those responding to negative deviations. Lifetime profit is maximised by adapting the optimal service mix as the battery ages.

math.OC↗

arXiv2026

Decision-Focused Scenario Generation and Selection for Efficient and Robust Grid Dispatch

The increasing uncertainty from flexible demand and renewable generation has made distributionally robust optimization (DRO) an important tool for robust power system dispatch. DRO relies on forecast scenarios to construct ambiguity sets, but conventional scenario generation pipelines are often trained in an accuracy-oriented manner and may neglect spatial correlations among uncertainties. This mismatch can produce ambiguity sets that are statistically plausible but suboptimal for downstream operation. This work proposes a decision-focused generative framework for correlated scenario generation in DRO-based dispatch. Instead of training generative models solely to fit the historical uncertainty distribution, the proposed framework optimizes generated scenarios according to their induced downstream operational cost. The proposed framework is tailored to mainstream generative models, including variational autoencoders, generative adversarial networks, and diffusion models, while capturing the joint distribution of uncertainties across buses. To improve computational tractability, we further develop a differentiable scenario selector that selects decision-relevant scenarios from a generated pool and can be trained within the same decision-focused pipeline. Case studies demonstrate that the proposed framework effectively reduces 0.80%-2.02% operational cost across different generative models compared to accuracy-oriented methods.

cs.LG↗

arXiv2026

Supervised Reinforcement Learning for the Coordination of Distributed Energy Resources

The increasing integration of distributed energy resources (DERs) is crucial for power system decarbonization, yet unlocking DERs' flexibility is challenged by their inherent uncertainties and modelling complexity. As traditional optimization methods struggle with such uncertainty and complexity of DERs, reinforcement learning (RL) has emerged as a promising alternative for DER management. However, standard RL methods suffer from sample inefficiency and sub-optimality when trained from scratch. Inspired by the training paradigms in large language models, this paper proposes a Supervised Reinforcement Learning (SRL) framework for learning DER coordination policies. This framework first pre-trains a policy on demonstration data in a supervised-learning fashion, which is then further fine-tuned using RL. Furthermore, we propose a two-step fine-tuning process: offline fine-tuning for enhancing policy performance and online fine-tuning for adapting it to the real-world dynamics. Experiments demonstrate that RL implementations based on the proposed framework significantly outperform all benchmarks, achieving high cost efficiency even under low-quality demonstration data.

cs.LG↗

arXiv2026

JAX-Based Batched AC Power Flow for GPU Acceleration and AI Ecosystem Integration

Coordinating growing grid flexibility under uncertainty is becoming increasingly important for efficient and reliable power-system operation. A core computational requirement is the efficient large-scale batched evaluation of AC power flow across candidate operating actions and uncertainty scenarios. Previous work has explored GPU-based batched power-flow evaluation, but has largely relied on hand-written C or CUDA code, creating barriers to customisation, efficient kernel optimisation, and long-term maintenance. JAX is a Python-based framework that enables efficient accelerator execution while keeping implementations in Python. This letter therefore proposes a JAX-based batched AC power-flow solver that uses current JAX functionality to implement Newton--Raphson for transmission networks and Z-Bus power flow for three-phase unbalanced distribution networks, achieving more than 10x speed-ups relative to pandapower and OpenDSS. In addition, JAX integrates seamlessly with the broader JAX-based AI ecosystem, making it straightforward to embed power-flow evaluation within AI methods for future larger-scale and more complex power-system operation.

eess.SY↗

arXiv2026

Strengthened and Faster Linear Approximation to Joint Chance Constraints with Wasserstein Ambiguity

Many real-world decision-making problems have uncertain parameters in constraints. Wasserstein distributionally robust joint chance constraints (WDRJCC) offer a promising solution by explicitly guaranteeing the probability of the simultaneous constraint satisfaction. However, WDRJCC are computationally demanding, and practical applications often require more tractable approaches, especially for large-scale problems such as power system unit commitment problems and multilevel problems with chance constraints in lower levels. To address this, this paper proposes a convex inner-approximation for WDRJCC with right-hand-side uncertainties (RHS-WDRJCC). We propose a Strengthened and Faster Linear Approximation (SFLA) by strengthening an existing convex inner-approximation. This strengthening process reduces the number of constraints and tightens the feasible region for ancillary variables, leading to significant computational speedup. We prove that the proposed SFLA does not introduce extra conservativeness and can be less conservative compared to common approximations such as W-CVaR. We then extend the proposed SFLA to a more interpretable decision-making paradigm: robustness maximization, where the risk level and the Wasserstein radius are determined by maximizing solution robustness subject to a utility degradation limit. We discuss the connection between risk minimization and radius maximization as two formulations of robustness maximization, and show the advantage of radius maximization. In power system unit commitment, the proposed SFLA achieves up to 10x computational speedup compared to the strengthened and exact reformulation. In a bilevel strategic bidding problem where the exact reformulation is not applicable due to non-convexity, the proposed SFLA leads to 90x speedup than W-CVaR. In robustness maximization, the proposed SFLA demonstrated over 100x speedup.

math.OC↗

arXiv2026

GradMAP: Gradient-Based Multi-Agent Proximal Learning for Grid-Edge Flexibility

Coordinating large populations of grid-edge devices requires learning methods that remain fully decentralised in deployment while still respecting three-phase AC distribution-network physics. This paper proposes gradient-based multi-agent proximal learning (GradMAP) to address this challenge. GradMAP trains independent neural-network policies for each agent without any parameter sharing, and each agent uses only its own local observation for online decision-making without communication. During offline training, GradMAP embeds a differentiable three-phase AC power-flow model in a primal-dual learning loop and uses implicit differentiation to propagate exact network-constraint violations to update the policy parameters. To speed up training, GradMAP reuses expensive environment gradients through a proximal surrogate within a trust region defined in the more direct policy-output (action) space, instead of the probability distribution space used in other works, such as PPO. In case studies with 1,000 agents managing batteries, heat pumps, and controllable generators on the IEEE 123-bus feeder, GradMAP learns decentralised policies that minimise three-phase AC load-flow constraint violations within 15 minutes of training on a single workstation-class NVIDIA RTX PRO 5000 Blackwell 48GB GPU. This is a 3--5x training speed-up over gradient-based self-supervised learning benchmarks and substantially better training efficiency than multi-agent reinforcement-learning benchmarks. In out-of-sample tests, GradMAP also delivers among the lowest operating cost and constraint violations.

cs.LG↗

arXiv2025

FICA: Faster Inner Convex Approximation of Chance Constrained Grid Dispatch with Decision-Coupled Uncertainty

This paper proposes a Faster Inner Convex Approximation (FICA) method for solving power system dispatch problems with Wasserstein distributionally robust joint chance constraints (WJCC) and incorporating the modelling of the automatic generation control factors. The problem studied belongs to the computationally challenging class of WJCC with left-hand-side uncertainty (LHS-WJCC). By exploiting the special one-dimensional structure (even if only partially present) of the problem, the proposed FICA incorporates a set of strong valid inequalities to accelerate the solution process. We prove that FICA achieves the same optimality as the well-known conditional value-at-risk (CVaR) inner convex approximation method. Our numerical experiments demonstrate that the proposed FICA can yield 40x computational speedup compared to CVaR, and can even reach up to 500x speedup when the optimisation horizon exceeds 16 time steps. This speedup is achieved when only 50% of constraints in a WJCC have the one-dimensional structure. The approximation quality is numerically verified to be the same as CVaR, and the quality gap is below 1% when compared to the computationally demanding exact reformulation of the LHS-WJCC in most cases. We also discuss the applications of FICA in optimisation problems from other domains that (partially) exhibit the one-dimensional structure.

math.OC↗

arXiv2025

Bilevel Transmission Expansion Planning with Joint Chance-Constrained Dispatch

In transmission expansion planning (TEP), network planners make long-term investment decisions while anticipating market clearing outcomes that are increasingly affected by renewable generation uncertainty. Additionally, market participants' sensitivity to network charges and the requirement for cost recovery by the network planner introduce further complexity. Since the day-ahead market clears before uncertainty realizes, explicitly modelling these uncertainties at the lower-level market clearing becomes important in bilevel TEP problems. In this paper, we introduce a novel bilevel TEP framework with lower-level joint chance-constrained market clearing that manages line flow constraints under wind uncertainty and accounts for the effect of network tariffs on participants' actual marginal costs and utility. To solve this complex problem, we propose a Strengthened Linear Approximation (SLA) technique for handling Wasserstein distributionally robust joint chance constraints with right-hand-side uncertainties (RHS-WDRJCC). The proposed method offers more efficient approximations without additional conservativeness and avoids the numerical issues encountered in existing approaches by introducing valid inequalities. The case study demonstrates that the proposed model achieves the desired out-of-sample constraint satisfaction probability. Moreover, the numerical results highlight the significant computational advantage of SLA, achieving up to a 26x speedup compared to existing methods such as worst-case conditional value-at-risk, while maintaining high solution quality.

math.OC↗

arXiv2025

Large Language Model Interface for Home Energy Management Systems

Home Energy Management Systems (HEMSs) help households tailor their electricity usage based on power system signals such as energy prices. This technology helps to reduce energy bills and offers greater demand-side flexibility that supports the power system stability. However, residents who lack a technical background may find it difficult to use HEMSs effectively, because HEMSs require well-formatted parameterization that reflects the characteristics of the energy resources, houses, and users' needs. Recently, Large-Language Models (LLMs) have demonstrated an outstanding ability in language understanding. Motivated by this, we propose an LLM-based interface that interacts with users to understand and parameterize their ``badly-formatted answers'', and then outputs well-formatted parameters to implement an HEMS. We further use Reason and Act method (ReAct) and few-shot prompting to enhance the LLM performance. Evaluating the interface performance requires multiple user--LLM interactions. To avoid the efforts in finding volunteer users and reduce the evaluation time, we additionally propose a method that uses another LLM to simulate users with varying expertise, ranging from knowledgeable to non-technical. By comprehensive evaluation, the proposed LLM-based HEMS interface achieves an average parameter retrieval accuracy of 88\%, outperforming benchmark models without ReAct and/or few-shot prompting.

cs.AI↗

arXiv2024

AI-focused HPC Data Centers Can Provide More Power Grid Flexibility and at Lower Cost

The recent growth of Artificial Intelligence (AI), particularly large language models, requires energy-demanding high-performance computing (HPC) data centers, which poses a significant burden on power system capacity. Scheduling data center computing jobs to manage power demand can alleviate network stress with minimal infrastructure investment and contribute to fast time-scale power system balancing. This study, for the first time, comprehensively analyzes the capability and cost of grid flexibility provision by GPU-heavy AI-focused HPC data centers, along with a comparison with CPU-heavy general-purpose HPC data centers traditionally used for scientific computing. A data center flexibility cost model is proposed that accounts for the value of computing. Using real-world computing traces from 7 AI-focused HPC data centers and 7 general-purpose HPC data centers, along with computing prices from 3 cloud platforms, we find that AI-focused HPC data centers can offer greater flexibility at 50% lower cost compared to general-purpose HPC data centers for a range of power system services. By comparing the cost to flexibility market prices, we illustrate the financial profitability of flexibility provision for AI-focused HPC data centers. Finally, our flexibility and cost estimates can be scaled using parameters of other data centers through algebraic operations, avoiding the need for re-optimization.

eess.SY↗

arXiv2024

Exploiting Data Centres and Local Energy Communities Synergies for Market Participation

The evolving energy landscape has propelled energy communities to the forefront of modern energy management. However, existing research has yet to explore the potential synergies between data centres and energy communities, necessitating an assessment on their collective capabilities for cost efficiency, waste heat optimisation, and market participation. This paper presents a mixed integer linear programming model to assess the collaborative performance of energy communities, data centres and energy markets. The evaluation focuses on the efficient use of waste heat and the flexibility of job scheduling while minimising system energy costs and maintaining quality of service requirements for data centres. Our results, based on realistic profiles of an energy community and a data centre, showcase significant benefits of these synergies, with a 38% reduction in operating costs and an 87% decrease in heat demand.

eess.SY↗

arXiv2024

Datasets of Great Britain Primary Substations Integrated with Household Heating Information

The growing demand for electrified heating, electrified transportation, and power-intensive data centres challenge distribution networks. If electrification projects are carried out without considering electrical distribution infrastructure, there could be unexpected blackouts and financial losses. Datasets containing real-world distribution network information are required to address this. On the other hand, social data, such as household heating composition, are closely coupled with people's lives. Studying the coupling between the energy system and society is important in promoting social welfare. To fill these gaps, this paper introduces two datasets. The first is the main dataset for the distribution networks in Great Britain (GB), collecting information on firm capacity, peak demands, locations, and parent transmission nodes (the Grid Supply Point, namely GSP) for all primary substations (PSs). PSs are a crucial part of the UK distribution network and are at the lowest voltage level (11 kV) with publicly available data for most UK Distribution Network Operators (DNOs). Substation firm capacity and peak demand facilitate an understanding of the remaining room of the existing network. The parent GSP information helps link the dataset of distribution networks to datasets of transmission networks. The second dataset extends the main network dataset, linking each PS to information about the number of households that use different types of central heating recorded in census data. The derivation of the second dataset is based on locations of PSs collected in the main dataset with appropriate assumptions. The derivation process may also be replicated to integrate other social datasets.

eess.SY↗