Searcharxiv⌕ Search

arXiv subjects

Zhong Fan

Publications and source records attributed to Zhong Fan.

At least 19 recordsLinked to original sources
arXiv2026

Multistep and Multivariate Time Series Explainability with GradientSHAP: A Wind Power Case Study

As the effects of climate change become increasingly apparent, the demand for renewable energy sources such as wind power continues to grow. However, renewables' variable and intermittent nature presents significant challenges in their integration into energy grids. Artificial intelligence (AI) has emerged as a powerful tool to address these issues. Still, its adoption in the energy sector has been hindered by the opacity of many AI models, which often function as black boxes. This paper introduces a carefully adapted GradientSHAP approach to provide interpretability for multivariate, multistep wind power forecasting -- a largely underexplored area. This capability is critical for day-ahead grid management and, to the best of our knowledge, this is the first study to apply GradientSHAP in the context of wind power forecasting. To address the inherent dimensionality, we developed novel visualizations that deliver both local and global explanations, uncovering intricate relationships between environmental factors, site-specific conditions, and temporal dynamics. While GradientSHAP is not an out-of-the-box solution, our findings demonstrate that, with thoughtful handling of SHAP values and domain expertise, it can offer valuable insights into model behaviour, supporting more informed and trustworthy decision-making. Nevertheless, its complexity poses challenges for practical implementation. Future research should focus on streamlining these interpretability methods to make them more actionable for routine deployment in critical energy applications.

cs.AI↗
arXiv2026

Knowledge boundary probing and demand-guided intervention for LLM-based power system code generation

Large language models (LLMs) can turn grid-analysis requests into executable programs for power-system simulation, but utilities and research laboratories often require on-premise deployment. In this setting, first-pass failures frequently arise at an API-knowledge boundary, through hallucinated functions, misused parameters, and mishandled result tables. We present PowerCodeBench, a parameterised benchmark generator released as a frozen 2,000-task suite for pandapower, and a deployment-time workflow that requires no weight updates. Documentation-driven L0-L3 probes produce per-model API profiles for diagnosis, model comparison, documentation allocation, and backend calibration. A query-side demand estimator selects layered API evidence before generation, while execution feedback routes targeted repair. Across ten open-weight LLMs (1.5B-480B) and four mid-tier APIs, the validation-enabled workflow raises scalar-match accuracy by 32-56 percentage points after up to three repair rounds relative to an unassisted first pass, for every model of at least 7B and every API. Open-weight models in the 70B-120B range reach the four-vendor mid-tier accuracy range under matched no-tool conditions. Selective injection approaches the full-layer reference using 41% of its prompt tokens. Among model-item pairs passing numerical checks under both workflows, engineering review confirms the requested analysis in 88% of full-workflow outputs versus 66% under plain repair. Round-0 pilots on OpenDSS and PyPSA motivate staged onboarding from broad retrieval at cold start to calibrated selective injection. Measured throughput, latency, energy, and allocated GPU memory establish a practical on-premise serving envelope.

cs.SE↗
arXiv2026

Action Forcing: Training World Models on Unsupervised Video by Recovering Underlying Egomotion Bases

Synchronised action annotations are needed to train controllable world models and these datasets remain elusive. Existing approaches make use of instrumented platforms with calibrated sensors, costly manual annotation, or latent-action models which lack grounding. We instead turn ordinary unlabelled video into action-supervised training data by recovering (without training) a data-derived egomotion basis. We track pixel displacements across frames and exploit the recurring coherent structure induced by egomotion to obtain grounded control signals directly. Using a method as simple as principal components analysis perform this, we find that the leading components provide signed, scalable, and composable throttle--yaw controls, although the method can recover only motion axes represented in the data. To prevent a high-capacity video DiT from exploiting pixel-level supervision, an online latent critic distils a frozen decoder--tracker--PCA (Principal Components Analysis) teacher without backpropagating through the decoder or tracker. Finally we critique the use of video generation metrics to evaluate WMs and introduce an example of an alternative, reference-free evaluation method. We measure \textit{controllability}, \textit{plausibility}, \textit{conjuring} (creating objects out of thin air) and \textit{geometric integrity}, revealing failures that conventional video metrics miss. We show that most baselines follow familiar action directions but struggle to reverse or remain stationary. Our model handles both while retaining compositional control and generation quality. Despite backwards actions being less than $1\%$ of our training data, we find that the model learns to reverse, scale its response linearly, and compose throttle with steering, all simply by learning through a grounded action space.

cs.CV↗
arXiv2025

Zero Carbon V2X Tariffs for Non-Domestic Customers

With the aim of meeting the worlds net-zero objectives, electricity trading through contractual agreements is becoming increasingly relevant in global and local energy markets. We develop contracts enabling efficient energy trading using Vehicle to Everything technology which can be applied to regulate energy markets and reduce costs and carbon emissions by using electric vehicles with bidirectional batteries to store energy during offpeak hours for export during peak hours. We introduce a contract based on the VCG mechanism which enables fleets of electric vehicles to export electricity to the grid efficiently throughout the day, where each electric vehicle has its energy consumption and exporting schedules and costs.

cs.GT↗
arXiv2025

Digital Player: Evaluating Large Language Models based Human-like Agent in Games

With the rapid advancement of Large Language Models (LLMs), LLM-based autonomous agents have shown the potential to function as digital employees, such as digital analysts, teachers, and programmers. In this paper, we develop an application-level testbed based on the open-source strategy game "Unciv", which has millions of active players, to enable researchers to build a "data flywheel" for studying human-like agents in the "digital players" task. This "Civilization"-like game features expansive decision-making spaces along with rich linguistic interactions such as diplomatic negotiations and acts of deception, posing significant challenges for LLM-based agents in terms of numerical reasoning and long-term planning. Another challenge for "digital players" is to generate human-like responses for social interaction, collaboration, and negotiation with human players. The open-source project can be found at https:/github.com/fuxiAIlab/CivAgent.

cs.LG↗
arXiv2025

Contract Design for V2G Smart Energy Trading

The transition to a net zero energy system necessitates development in a number of directions including developing advanced electricity trading markets. Due to electricity markets being responsible for a large portion of carbon emissions, improving the electricity markets' method for determining energy transactions could have a significant impact on carbon reductions and thus facilitate this transition. V2X technology can be applied to regulate different energy markets, and thus reduce costs and carbon emissions by using the batteries in electric vehicles to store energy during off-peak hours and export it during peak hours. We develop a novel contract based on the VCG-mechanism, for exporting and importing electricity effectively, and show how this mechanism can raise efficiency, facilitate the development of a sustainable and efficient electricity market, and bring us nearer to our Net Zero Goal.

cs.GT↗
arXiv2025

Addressing the sustainable AI trilemma: a case study on LLM agents and RAG

Large language models (LLMs) have demonstrated significant capabilities, but their widespread deployment and more advanced applications raise critical sustainability challenges, particularly in inference energy consumption. We propose the concept of the Sustainable AI Trilemma, highlighting the tensions between AI capability, digital equity, and environmental sustainability. Through a systematic case study of LLM agents and retrieval-augmented generation (RAG), we analyze the energy costs embedded in memory module designs and introduce novel metrics to quantify the trade-offs between energy consumption and system performance. Our experimental results reveal significant energy inefficiencies in current memory-augmented frameworks and demonstrate that resource-constrained environments face disproportionate efficiency penalties. Our findings challenge the prevailing LLM-centric paradigm in agent design and provide practical insights for developing more sustainable AI systems.

cs.CY↗
arXiv2023

Smart Energy Network Digital Twins: Findings from a UK-Based Demonstrator Project

Digital Twins promise to deliver a step-change in distribution system operations and planning, but there are few real-world examples that explore the challenges of combining imperfect model and measurement data, and then use these as the basis for subsequent analysis. In this work we propose a Digital Twin framework for electrical distribution systems and implement that framework on the Smart Energy Network Demonstrator microgrid in the UK. The data and software implementation are made available open-source, and consist of a network model, power meter measurements, and unbalanced power flow-based algorithms. Measurement and network uncertainties are shown to have a substantial impact on the quality of Digital Twin outputs. The potential benefits of a dynamic export limit and voltage control are estimated using the Digital Twin, using simulated measurements to address data quality challenges, with results showing curtailment for an exemplar day could be reduced by 56%. Power meter data and a network model are shown to be necessary for developing algorithms that enable decision-making that is robust to real-world uncertainties, with possibilities and challenges of Digital Twin development clearly demonstrated.

eess.SY↗
arXiv2022

Federated Learning for Short-term Residential Load Forecasting

Load forecasting is an essential task performed within the energy industry to help balance supply with demand and maintain a stable load on the electricity grid. As supply transitions towards less reliable renewable energy generation, smart meters will prove a vital component to facilitate these forecasting tasks. However, smart meter adoption is low among privacy-conscious consumers that fear intrusion upon their fine-grained consumption data. In this work we propose and explore a federated learning (FL) based approach for training forecasting models in a distributed, collaborative manner whilst retaining the privacy of the underlying data. We compare two approaches: FL, and a clustered variant, FL+HC against a non-private, centralised learning approach and a fully private, localised learning approach. Within these approaches, we measure model performance using RMSE and computational efficiency. In addition, we suggest the FL strategies are followed by a personalisation step and show that model performance can be improved by doing so. We show that FL+HC followed by personalisation can achieve a $\sim$5\% improvement in model performance with a $\sim$10x reduction in computation compared to localised learning. Finally we provide advice on private aggregation of predictions for building a private end-to-end load forecasting application.

cs.LG↗
arXiv2022

Battery and Hydrogen Energy Storage Control in a Smart Energy Network with Flexible Energy Demand using Deep Reinforcement Learning

Smart energy networks provide for an effective means to accommodate high penetrations of variable renewable energy sources like solar and wind, which are key for deep decarbonisation of energy production. However, given the variability of the renewables as well as the energy demand, it is imperative to develop effective control and energy storage schemes to manage the variable energy generation and achieve desired system economics and environmental goals. In this paper, we introduce a hybrid energy storage system composed of battery and hydrogen energy storage to handle the uncertainties related to electricity prices, renewable energy production and consumption. We aim to improve renewable energy utilisation and minimise energy costs and carbon emissions while ensuring energy reliability and stability within the network. To achieve this, we propose a multi-agent deep deterministic policy gradient approach, which is a deep reinforcement learning-based control strategy to optimise the scheduling of the hybrid energy storage system and energy demand in real-time. The proposed approach is model-free and does not require explicit knowledge and rigorous mathematical models of the smart energy network environment. Simulation results based on real-world data show that: (i) integration and optimised operation of the hybrid energy storage system and energy demand reduces carbon emissions by 78.69%, improves cost savings by 23.5% and renewable energy utilisation by over 13.2% compared to other baseline models and (ii) the proposed algorithm outperforms the state-of-the-art self-learning algorithms like deep-Q network.

eess.SY↗
arXiv2021

Renewable energy integration and microgrid energy trading using multi-agent deep reinforcement learning

In this paper, multi-agent reinforcement learning is used to control a hybrid energy storage system working collaboratively to reduce the energy costs of a microgrid through maximising the value of renewable energy and trading. The agents must learn to control three different types of energy storage system suited for short, medium, and long-term storage under fluctuating demand, dynamic wholesale energy prices, and unpredictable renewable energy generation. Two case studies are considered: the first looking at how the energy storage systems can better integrate renewable energy generation under dynamic pricing, and the second with how those same agents can be used alongside an aggregator agent to sell energy to self-interested external microgrids looking to reduce their own energy bills. This work found that the centralised learning with decentralised execution of the multi-agent deep deterministic policy gradient and its state-of-the-art variants allowed the multi-agent methods to perform significantly better than the control from a single global agent. It was also found that using separate reward functions in the multi-agent approach performed much better than using a single control agent. Being able to trade with the other microgrids, rather than just selling back to the utility grid, also was found to greatly increase the grid's savings.

cs.MA↗
arXiv2021

Multi-Agent Deep Deterministic Policy Gradient Algorithm for Peer-to-Peer Energy Trading Considering Distribution Network Constraints

In this paper, we investigate an energy cost minimization problem for prosumers participating in peer-to-peer energy trading. Due to (i) uncertainties caused by renewable energy generation and consumption, (ii) difficulties in developing an accurate and efficient energy trading model, and (iii) the need to satisfy distribution network constraints, it is challenging for prosumers to obtain optimal energy trading decisions that minimize their individual energy costs. To address the challenge, we first formulate the above problem as a Markov decision process and propose a multi-agent deep deterministic policy gradient algorithm to learn optimal energy trading decisions. To satisfy the distribution network constraints, we propose distribution network tariffs which we incorporate in the algorithm as incentives to incentivize energy trading decisions that help to satisfy the constraints and penalize the decisions that violate them. The proposed algorithm is model-free and allows the agents to learn the optimal energy trading decisions without having prior information about other agents in the network. Simulation results based on real-world datasets show the effectiveness and robustness of the proposed algorithm.

eess.SY↗
arXiv2021

A Review on Cybersecurity in Smart Local Energy Systems: Requirements, Challenges, and Standards

Smart local energy system (SLES) is considered as a promising pathway facilitating a more effective and localised operation, benefited from the complex information and communication technology (ICT) infrastructures and Internet of things (IoT) technologies. As a part of the critical infrastructure, it is important to not only put effective detection and management to tackle potential cybersecurity issues, but also require considerable numbers of standards to ensure the security of the internet of things system to minimise the risks. This study aims to review the existing standards, investigate how the compatibility with SLES development, and identify the area to focus on in the future. Although existing standards and protocols are highly fragmented, our findings suggest that many of them can meet the requirements of the applications and infrastructures of SLES. Additionally, many standards have been introduced to protect information security and personal privacy due to their increasing importance. The research also suggests that the industry needs to produce more affordable and cyber-secured devices and services. For the government and regulators, relevant guidelines on the minimum function and security requirements for applications should be provided. Additionally, compliance testing and certifications should be in place and carried out by an independent third party to ensure the components of SLES ecosystem with a satisfied security level by design.

cs.CR↗
arXiv2021

The energy revolution: cyber physical advances and opportunities for smart local energy systems

We have designed a two-stage, 10-step process to give organisations a method to analyse small local energy systems (SLES) projects based on their Cyber Physical System components in order to develop future-proof energy systems. SLES are often developed for a specific range of use cases and functions, and these match the specific requirements and needs of the community, location or site under consideration. During the design and commissioning, new and specific cyber physical architectures are developed. These are the control and data systems that are needed to bridge the gap between the physical assets, the captured data and the control signals. Often, the cyber physical architecture and infrastructure is focused on functionality and the delivery of the specific applications. But we find that technologies and approaches have arisen from other fields that, if used within SLES, could support the flexibility, scalability and reusability vital to their success. As these can improve the operational data systems then they can also be used to enhance predictive functions If used and deployed effectively, these new approaches can offer longer term improvements in the use and effectiveness of SLES, while allowing the concepts and designs to be capitalised upon through wider roll-out and the offering of commercial services or products.

eess.SY↗
arXiv2021

Data-driven battery operation for energy arbitrage using rainbow deep reinforcement learning

As the world seeks to become more sustainable, intelligent solutions are needed to increase the penetration of renewable energy. In this paper, the model-free deep reinforcement learning algorithm Rainbow Deep Q-Networks is used to control a battery in a small microgrid to perform energy arbitrage and more efficiently utilise solar and wind energy sources. The grid operates with its own demand and renewable generation based on a dataset collected at Keele University, as well as using dynamic energy pricing from a real wholesale energy market. Four scenarios are tested including using demand and price forecasting produced with local weather data. The algorithm and its subcomponents are evaluated against two continuous control benchmarks with Rainbow able to outperform all other method. This research shows the importance of using the distributional approach for reinforcement learning when working with complex environments and reward functions, as well as how it can be used to visualise and contextualise the agent's behaviour for real-world applications.

cs.LG↗
arXiv2020

Privacy Preserving Demand Forecasting to Encourage Consumer Acceptance of Smart Energy Meters

In this proposal paper we highlight the need for privacy preserving energy demand forecasting to allay a major concern consumers have about smart meter installations. High resolution smart meter data can expose many private aspects of a consumer's household such as occupancy, habits and individual appliance usage. Yet smart metering infrastructure has the potential to vastly reduce carbon emissions from the energy sector through improved operating efficiencies. We propose the application of a distributed machine learning setting known as federated learning for energy demand forecasting at various scales to make load prediction possible whilst retaining the privacy of consumers' raw energy consumption data.

cs.LG↗
arXiv2020

A Review of Privacy-preserving Federated Learning for the Internet-of-Things

The Internet-of-Things (IoT) generates vast quantities of data, much of it attributable to individuals' activity and behaviour. Gathering personal data and performing machine learning tasks on this data in a central location presents a significant privacy risk to individuals as well as challenges with communicating this data to the cloud. However, analytics based on machine learning and in particular deep learning benefit greatly from large amounts of data to develop high-performance predictive models. This work reviews federated learning as an approach for performing machine learning on distributed data with the goal of protecting the privacy of user-generated data as well as reducing communication costs associated with data transfer. We survey a wide variety of papers covering communication-efficiency, client heterogeneity and privacy preserving methods that are crucial for federated learning in the context of the IoT. Throughout this review, we identify the strengths and weaknesses of different methods applied to federated learning and finally, we outline future directions for privacy preserving federated learning research, particularly focusing on IoT applications.

cs.LG↗