arXiv subjects

Ning Lu

Publications and source records attributed to Ning Lu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Backdoor Graph Condensation

Graph condensation has recently emerged as a prevalent technique to improve the training efficiency for graph neural networks (GNNs). It condenses a large graph into a small one such that a GNN trained on this small synthetic graph can achieve comparable performance to a GNN trained on the large graph. However, while existing graph condensation studies mainly focus on the best trade-off between graph size and the GNNs' performance (model utility), they overlook the security issues of graph condensation. To bridge this gap, we first explore backdoor attack against the GNNs trained on the condensed graphs. We introduce an effective backdoor attack against graph condensation, termed BGC. This attack aims to (1) preserve the condensed graph quality despite trigger injection, and (2) ensure trigger efficacy through the condensation process, achieving a high attack success rate. Specifically, BGC consistently updates triggers during condensation and targets representative nodes for poisoning. Extensive experiments demonstrate the effectiveness of our attack. BGC achieves a high attack success rate (close to 1.0) and good model utility in all cases. Furthermore, the results against multiple defense methods demonstrate BGC's resilience under their defenses. Finally, we analyze the key hyperparameters that influence the attack performance. Our code is available at: https://github.com/JiahaoWuGit/BGC.

cs.LG↗

arXiv2025

Design and Implementation of Scalable Communication Interfaces for Reliable and Stable Real-time Co-Simulation of Power Systems

Co-simulation offers an integrated approach for modeling the large-scale integration of inverter-based resources (IBRs) into transmission and distribution grids. This paper presents a scalable communication interface design and implementation to enable reliable and stable real-time co-simulation of power systems with high IBR penetration. The communication interface is categorized into two types: local and remote. In local scenarios, where subsystems are connected within a single local area network (LAN), low-latency communication facilitates the seamless integration of electromagnetic transient (EMT) and phasor-domain models, enabling efficient interactions with power and energy management algorithms. For remote scenarios, data exchange is achieved via internet-based file sharing or VPN-enabled communication. The performance of both methods is evaluated using OPAL-RT as a real-time simulator, demonstrating scalability, effectiveness, and challenges specific to real-time co-simulation applications. To mitigate instability arising from data resolution mismatches in time-sensitive co-simulations, a real-time data extrapolation method is proposed. This approach significantly enhances stability and reliability, ensuring more accurate simulation outcomes. The implementation code is available on GitHub, providing researchers the tools to replicate and expand upon this work.

eess.SY↗

arXiv2024

New $^{63}$Ga(p,$γ$)$^{64}$Ge and $^{64}$Ge(p,$γ$)$^{65}$As reaction rates corresponding to the temperature regime of thermonuclear X-ray bursts

We compute the $^{63}$Ga(p,$γ$)$^{64}$Ge and $^{64}$Ge(p,$γ$)$^{65}$As thermonuclear reaction rates using the latest experimental input supplemented with theoretical nuclear spectroscopic information. The experimental input consists of the latest proton thresholds of $^{64}$Ge and $^{65}$As, and the nuclear spectroscopic information of $^{65}$As, whereas the theoretical nuclear spectroscopic information for $^{64}$Ge and $^{65}$As are deduced from the full pf-shell space configuration-interaction shell-model calculations with the GXPF1A Hamiltonian. Both thermonuclear reaction rates are determined with known uncertainties at the energies that correspond to the Gamow windows of the temperature regime relevant to Type I X-ray bursts, covering the typical temperature range of the thermonuclear runaway of the GS 1826$-$24 periodic bursts and SAX J1808.4$-$3658 photospheric radius expansion bursts.

nucl-th↗

arXiv2024

Invisible Manipulation Deep Reinforcement Learning Enhanced Stealthy Attacks on Battery Energy Management Systems

This paper introduces "invisible manipulation," an innovative cyber-attack mechanism achieved through strategically timed stealthy false data injection attacks (SFDIAs). By stealthily manipulating measurements of a critical asset prior to the target time period, the attacker can subtly guide the engineering system toward a predetermined operational state without detection. Using the battery energy management system (BEMS) as a case study, we employ deep reinforcement learning (DRL) to generate synthetic measurements, such as battery voltage and current, that align closely with actual measurements. These synthetic measurements, falling within the acceptable error margin of residual-based bad data detection algorithm provided by state estimation, can evade detection and mislead Extended Kalman-filter-based State of Charge estimation. Subsequently, considering the deceptive data as valid inputs, the BEMS will operate the BESS towards the attacker desired operational states when the targeted time period come. The use of the DRL-based scheme allows us to covert an online optimization problem into an offline training process, thereby alleviating the computational burden for real-time implementation. Comprehensive testing on a high-fidelity microgrid real-time simulation testbed validates the effectiveness and adaptability of the proposed methods in achieving different attack objectives.

eess.SY↗

arXiv2024

Training Overhead Ratio: A Practical Reliability Metric for Large Language Model Training Systems

Large Language Models (LLMs) are revolutionizing the AI industry with their superior capabilities. Training these models requires large-scale GPU clusters and significant computing time, leading to frequent failures that significantly increase training costs. Despite its significance, this field lacks a metric for evaluating reliability. In this work, we introduce a novel reliability metric called \emph{Training Overhead Ratio} (TOR) to evaluate the reliability of fault-tolerant LLM training systems. TOR is defined as the ratio of optimal training time to the observed training time of a system, serving as a practical tool for users to estimate the actual time required to train an LLM on a given system. Furthermore, our investigation identifies the key factor for enhancing reliability and present TOR equations for various types of failures encountered in practice.

cs.DC↗

arXiv2024

A Two-Stage Optimization Method for Real-Time Parameterization of PV-Farm Digital Twin

Digital twins (DTs) are high-fidelity virtual models of physical systems. This paper details a novel two-stage optimization method for real-time parameterization of photovoltaic digital twins (PVDTs) using field measurements. Initially, the method estimates equivalent irradiance from PV power, voltage, and current data, eliminating the need for direct irradiance sensors. This is crucial for tuning the DT's parameters to actual environmental conditions, thereby improving power prediction accuracy. The second stage focuses on refining these parameters by minimizing discrepancies between measured and predicted outputs. This optimization utilizes the estimated equivalent irradiance as a model input, maintaining synchronization with real-world conditions. Parameter updates are event-trigger, launched when deviations exceed predefined thresholds. This strategy optimizes prediction accuracy and manages communication loads efficiently. Validated with extensive data from a PV farm, this approach outperforms existing methodologies in predictive accuracy and operational efficiency, significantly improving the performance DTs in real-time grid operations.

eess.SY↗

arXiv2024

Highly Efficient and Stable Perovskite Solar Cells via MultiFunctional Curcumin Modified Buried Interface

The buried interface between the electron transport layer and the perovskite layer suffers from severe interface defects and imperfect energy level alignment. To address this issue, this study employs a multifunctional organic molecule, curcumin, to modify the interface between SnO2 and the perovskite layer. The functional groups on curcumin effectively passivate the defects on both sides of the interface, reducing -OH and oxygen vacancy defects on the SnO2 surface and passivating uncoordinated Pb2+ in the perovskite layer. This results in a more compatible energy level alignment and lower defect density at the interface, enhancing carrier transport across it. Consequently, the devices based on curcumin achieve an impressive champion power conversion efficiency (PCE) of 24.46%, compared to 22.03% for control devices. This work demonstrates a simple, green, hydrophobic, and efficient molecular modification method for the buried interface, laying the foundation for the development of high-performance and stable perovskite solar cells.

cond-mat.mtrl-sci↗

arXiv2024

Applying Fine-Tuned LLMs for Reducing Data Needs in Load Profile Analysis

This paper presents a novel method for utilizing fine-tuned Large Language Models (LLMs) to minimize data requirements in load profile analysis, demonstrated through the restoration of missing data in power system load profiles. A two-stage fine-tuning strategy is proposed to adapt a pre-trained LLMs, i.e., GPT-3.5, for missing data restoration tasks. Through empirical evaluation, we demonstrate the effectiveness of the fine-tuned model in accurately restoring missing data, achieving comparable performance to state-of-the-art specifically designed models such as BERT-PIN. Key findings include the importance of prompt engineering and the optimal utilization of fine-tuning samples, highlighting the efficiency of few-shot learning in transferring knowledge from general user cases to specific target users. Furthermore, the proposed approach demonstrates notable cost-effectiveness and time efficiency compared to training models from scratch, making it a practical solution for scenarios with limited data availability and computing resources. This research has significant potential for application to other power system load profile analysis tasks. Consequently, it advances the use of LLMs in power system analytics, offering promising implications for enhancing the resilience and efficiency of power distribution systems.

cs.LG↗

arXiv2024

Large Language Models can be Guided to Evade AI-Generated Text Detection

Large language models (LLMs) have shown remarkable performance in various tasks and have been extensively utilized by the public. However, the increasing concerns regarding the misuse of LLMs, such as plagiarism and spamming, have led to the development of multiple detectors, including fine-tuned classifiers and statistical methods. In this study, we equip LLMs with prompts, rather than relying on an external paraphraser, to evaluate the vulnerability of these detectors. We propose a novel Substitution-based In-Context example Optimization method (SICO) to automatically construct prompts for evading the detectors. SICO is cost-efficient as it requires only 40 human-written examples and a limited number of LLM inferences to generate a prompt. Moreover, once a task-specific prompt has been constructed, it can be universally used against a wide range of detectors. Extensive experiments across three real-world tasks demonstrate that SICO significantly outperforms the paraphraser baselines and enables GPT-3.5 to successfully evade six detectors, decreasing their AUC by 0.5 on average. Furthermore, a comprehensive human evaluation show that the SICO-generated text achieves human-level readability and task completion rates, while preserving high imperceptibility. Finally, we propose an ensemble approach to enhance the robustness of detectors against SICO attack. The code is publicly available at https://github.com/ColinLu50/Evade-GPT-Detector.

cs.CL↗

arXiv2024

Less is More: Understanding Word-level Textual Adversarial Attack via n-gram Frequency Descend

Word-level textual adversarial attacks have demonstrated notable efficacy in misleading Natural Language Processing (NLP) models. Despite their success, the underlying reasons for their effectiveness and the fundamental characteristics of adversarial examples (AEs) remain obscure. This work aims to interpret word-level attacks by examining their $n$-gram frequency patterns. Our comprehensive experiments reveal that in approximately 90\% of cases, word-level attacks lead to the generation of examples where the frequency of $n$-grams decreases, a tendency we term as the $n$-gram Frequency Descend ($n$-FD). This finding suggests a straightforward strategy to enhance model robustness: training models using examples with $n$-FD. To examine the feasibility of this strategy, we employed the $n$-gram frequency information, as an alternative to conventional loss gradients, to generate perturbed examples in adversarial training. The experiment results indicate that the frequency-based approach performs comparably with the gradient-based approach in improving model robustness. Our research offers a novel and more intuitive perspective for understanding word-level textual adversarial attacks and proposes a new direction to improve model robustness.

cs.CL↗

arXiv2024

A Novel Vision Transformer based Load Profile Analysis using Load Images as Inputs

This paper introduces ViT4LPA, an innovative Vision Transformer (ViT) based approach for Load Profile Analysis (LPA). We transform time-series load profiles into load images. This allows us to leverage the ViT architecture, originally designed for image processing, as a pre-trained image encoder to uncover latent patterns within load data. ViT is pre-trained using an extensive load image dataset, comprising 1M load images derived from smart meter data collected over a two-year period from 2,000 residential users. The training methodology is self-supervised, masked image modeling, wherein masked load images are restored to reveal hidden relationships among image patches. The pre-trained ViT encoder is then applied to various downstream tasks, including the identification of electric vehicle (EV) charging loads and behind-the-meter solar photovoltaic (PV) systems and load disaggregation. Simulation results illustrate ViT4LPA's superior performance compared to existing neural network models in downstream tasks. Additionally, we conduct an in-depth analysis of the attention weights within the ViT4LPA model to gain insights into its information flow mechanisms.

eess.SY↗

arXiv2024

Assessment of Transmission-level Fault Impacts on 3-phase and 1-phase Distribution IBR Operation

The widespread deployment of inverter-based resources (IBRs) renders distribution systems susceptible to transmission-level faults. This paper presents a comprehensive analysis of the impact of transmission-level faults on 3-phase and 1-phase distribution IBR operation. To evaluate distributed IBR tripping across various phases and locations on a distribution feeder, we conduct simulations of both symmetrical and unsymmetrical transmission faults at progressively greater electrical distances on a real-time transmission and distribution (T&D) co-simulation platform. The IBR power-to-load ratios (PLRs) at 50%, 100%, and 300% are considered to emulate low, medium, and high IBR conditions. Our results indicate that, while 1-phase and 2-phase faults typically trigger fewer IBR trips when compared to 3-phase faults, a significant power imbalance arises from the tripping of 1-phase IBRs on the affected phases. The imbalance can result in significant power quality problems and unintended equipment tripping. It may be necessary to design fault-ride-through mechanisms specifically tailored to 1-phase IBRs to help mitigate the power imbalances caused by unbalanced faults.

eess.SY↗

arXiv2024

Nuclear ground-state properties probed by the relativistic Hartree-Bogoliubov approach

Using the relativistic Hartree-Bogoliubov framework with separable pairing force coupled with the latest covariant density functionals, i.e., PC-L3R, PC-X, DD-PCX, and DD-MEX, we systematically explore the ground-state properties of all isotopes of Z=8-110. These properties consist of the binding energies, one- and two-neutron separation energies ($S_\mathrm{n}$ and $S_\mathrm{2n}$), root-mean-square radius of matter, of neutron, of proton, and of charge distributions, Fermi surfaces, ground-state spins and parities. We then predict the edges of nuclear landscape and bound nuclei for the isotopic chains from oxygen (Z=8) to darmstadtium (Z=110) based on these latest covariant density functionals. The number of bound nuclei predicted by PC-L3R, PC-X, DD-PCX, and DD-MEX, are 9004, 9162, 6799, and 7112, respectively. The root-mean-square deviations of $S_\mathrm{n}$ ($S_\mathrm{2n}$) yielded from PC-L3R, PCX, DD-PCX, and DD-MEX are 0.962 (1.300) MeV, 0.920 (1.483) MeV, 0.993 (1.753) MeV, and 1.010 (1.544) MeV, respectively. The root-mean-square deviations of charge radius distributions of comparing the available experimental values with the theoretical counterparts resulted from PC-L3R, PC-X, DD-PCX, and DD-MEX are 0.035 fm, 0.037 fm, 0.035 fm, and 0.034 fm, respectively. We notice pronounced differences between the empirical and theoretical root-mean-square radii of neutron at nuclei near the neutron drip line of the Mg, Ca, and Kr isotopic chains, suggesting the possible existence of the halo or giant halo phenomena.

nucl-th↗

arXiv2023

Self-distillation Regularized Connectionist Temporal Classification Loss for Text Recognition: A Simple Yet Effective Approach

Text recognition methods are gaining rapid development. Some advanced techniques, e.g., powerful modules, language models, and un- and semi-supervised learning schemes, consecutively push the performance on public benchmarks forward. However, the problem of how to better optimize a text recognition model from the perspective of loss functions is largely overlooked. CTC-based methods, widely used in practice due to their good balance between performance and inference speed, still grapple with accuracy degradation. This is because CTC loss emphasizes the optimization of the entire sequence target while neglecting to learn individual characters. We propose a self-distillation scheme for CTC-based model to address this issue. It incorporates a framewise regularization term in CTC loss to emphasize individual supervision, and leverages the maximizing-a-posteriori of latent alignment to solve the inconsistency problem that arises in distillation between CTC-based models. We refer to the regularized CTC loss as Distillation Connectionist Temporal Classification (DCTC) loss. DCTC loss is module-free, requiring no extra parameters, longer inference lag, or additional training data or phases. Extensive experiments on public benchmarks demonstrate that DCTC can boost text recognition model accuracy by up to 2.6%, without any of these drawbacks.

cs.CV↗

arXiv2023

Effective and Imperceptible Adversarial Textual Attack via Multi-objectivization

The field of adversarial textual attack has significantly grown over the last few years, where the commonly considered objective is to craft adversarial examples (AEs) that can successfully fool the target model. However, the imperceptibility of attacks, which is also essential for practical attackers, is often left out by previous studies. In consequence, the crafted AEs tend to have obvious structural and semantic differences from the original human-written text, making them easily perceptible. In this work, we advocate leveraging multi-objectivization to address such issue. Specifically, we reformulate the problem of crafting AEs as a multi-objective optimization problem, where the attack imperceptibility is considered as an auxiliary objective. Then, we propose a simple yet effective evolutionary algorithm, dubbed HydraText, to solve this problem. To the best of our knowledge, HydraText is currently the only approach that can be effectively applied to both score-based and decision-based attack settings. Exhaustive experiments involving 44237 instances demonstrate that HydraText consistently achieves competitive attack success rates and better attack imperceptibility than the recently proposed attack approaches. A human evaluation study also shows that the AEs crafted by HydraText are more indistinguishable from human-written text. Finally, these AEs exhibit good transferability and can bring notable robustness improvement to the target model by adversarial training.

cs.CL↗

arXiv2023

An Iterative Bidirectional Gradient Boosting Approach for CVR Baseline Estimation

This paper presents a novel Iterative Bidirectional Gradient Boosting Model (IBi-GBM) for estimating the baseline of Conservation Voltage Reduction (CVR) programs. In contrast to many existing methods, we treat CVR baseline estimation as a missing data retrieval problem. The approach involves dividing the load and its corresponding temperature profiles into three periods: pre-CVR, CVR, and post-CVR. To restore the missing load profile during the CVR period, the method employs a three-step process. First, a forward-pass GBM is executed using data from the pre-CVR period as inputs. Subsequently, a backward-pass GBM is applied using data from the post-CVR period. The two restored load profiles are reconciled, considering pre-calculated weights derived from forecasting accuracy, and only the leftmost and rightmost points are retained. The newly restored points are then included as inputs for the subsequent iteration. This iterative procedure continues until the original load data in the CVR period is fully restored. We develop IBi-GBM using actual smart meter and Supervisory Control and Data Acquisition (SCADA) data. Our results demonstrate that IBi-GBM exhibits robust performance across various data resolutions and in different seasons and outperforms existing methods by achieving a 1-2% reduction in normalized Root Mean Square Error (nRMSE).

eess.SY↗

arXiv2023

Multi-Feeder Restoration using Multi-Microgrid Formation and Management

This papers highlights the benefit of coordinating resources on mulitple active distribution feeders during severe long duration outages through multi-microgrid formation. A graph-theory based multi-microgrid formation algorithm is developed which is agnostic of the underlying energy management scheme of the microgrids and solved in a rolling horizon fashion. The algorithm is then enhanced to handle multiple feeders where formation of long laterals needs to be avoided due to potential voltage control issues in distribution systems. The algorithm is evaluated on a synthetic two feeder system derived from interconnecting two IEEE 123 node system. The results indicate increased service to loads in the system and better utilization of renewable resources.

eess.SY↗

arXiv2023

BERT-PIN: A BERT-based Framework for Recovering Missing Data Segments in Time-series Load Profiles

Inspired by the success of the Transformer model in natural language processing and computer vision, this paper introduces BERT-PIN, a Bidirectional Encoder Representations from Transformers (BERT) powered Profile Inpainting Network. BERT-PIN recovers multiple missing data segments (MDSs) using load and temperature time-series profiles as inputs. To adopt a standard Transformer model structure for profile inpainting, we segment the load and temperature profiles into line segments, treating each segment as a word and the entire profile as a sentence. We incorporate a top candidates selection process in BERT-PIN, enabling it to produce a sequence of probability distributions, based on which users can generate multiple plausible imputed data sets, each reflecting different confidence levels. We develop and evaluate BERT-PIN using real-world dataset for two applications: multiple MDSs recovery and demand response baseline estimation. Simulation results show that BERT-PIN outperforms the existing methods in accuracy while is capable of restoring multiple MDSs within a longer window. BERT-PIN, served as a pre-trained model, can be fine-tuned for conducting many downstream tasks, such as classification and super resolution.

eess.AS↗