arXiv subjects

Yiyan Li

Publications and source records attributed to Yiyan Li.

At least 19 recordsLinked to original sources

arXiv2026

IDSTune: A Multi-Agent Collaborative Framework for Integrated Database System Tuning

Database tuning is critical for achieving high performance in modern database management systems (DBMSs). Existing methods typically optimize a single component---knobs, indexes, or materialized views---without accounting for their interdependencies. This limitation arises because these components require different tuning strategies and are difficult to integrate within a unified framework. As a result, directly extending a method to multiple components or simply combining separate methods often fails to capture cross-component collaboration and shared tuning signals. Moreover, existing methods are insufficient for handling diverse workloads, evolving data, and dynamic query patterns. To address these limitations, we propose IDSTune, an integrated tuning framework that jointly optimizes multiple configuration components through LLM-driven multi-agent collaboration. IDSTune operates in two phases: (i) workload compression, which extracts and selects task-relevant features, and (ii) configuration recommendation, where specialized agents collaboratively generate and refine configurations for knobs, indexes, and materialized views under the supervision of a centralized coordinator. By incorporating feedback and external knowledge retrieval, IDSTune achieves efficient and globally consistent tuning. Extensive experiments show that IDSTune achieves up to 38% performance improvement and 57% faster tuning, with strong adaptability across diverse scenarios.

cs.DB↗

arXiv2026

A Physics-guided Fine-tuned LLM-based Framework for Customized Power Distribution System Feeder Generation

Power distribution system feeder models (e.g., IEEE 33-bus system, IEEE 13-bus system, etc.) are cornerstones for conducting power distribution system studies. As real-world feeder models are hard to acquire due to energy security concerns, generating high-quality synthetic feeders becomes an important alternative to satisfy the fast-growing and diversified needs of power system researchers and engineers. In this paper, we propose an LLM-based synthetic feeder generation framework that can achieve end-to-end generation from natural language specifications to physically consistent feeder models. First, Supervised Fine-Tuning (SFT) is performed on a dataset created following physical laws to empower the LLM with syntactic understanding of complex feeder structures. Second, Group Relative Policy Optimization (GRPO) with a specially-designed multi-stage gated reward function is introduced to better align the generation results with user intent and physical constraints. Third, a dual-agent architecture is deployed to refine and evaluate the generated feeders. Specifically, a refinement agent calibrates the feeder model parameters referring to the industrial feeder design standards, while a judge agent provides quality assessments. Case studies demonstrate that the proposed framework generates customizable feeders with valid formats, physical consistency and high engineering applicability.

eess.SY↗

arXiv2026

Flow-PIN: A Two-Stage Power-Flow-Guided Method for System-Wide Multivariate Profile Inpainting in Distribution Networks

High-quality system measurement data is critical for power distribution system operation. As deep generative models (e.g., GAN, Diffusion, etc.) have been widely studied to solve the missing data restoration problem to enhance the data quality, their results may look "realistic" but not sufficiently "accurate" due to lacking physical guarantees. To address this limitation, a two-stage physics-guided framework, Flow-PIN, is proposed in this paper for system-wide multivariate profile inpainting. The first stage employs a conditional flow matching model, conditioned on topological and correlation graphs, to generate candidate values. A physical penalty is integrated into the loss function to constrain the generative vector field based on grid physical laws. The second stage introduces a topology-aware power-flow-guided refiner that utilizes Laplacian positional encoding to inject topology information into node embeddings. By coupling alternating current power flow equations with a differentiable correlation alignment mechanism, this refiner further corrects numerical deviations. Evaluations on an active distribution network dataset benchmark the proposed framework against ten representative baselines. The results show that Flow-PIN achieves high-fidelity profile inpainting across three dimensions: maximizing numerical accuracy, capturing temporal fluctuation, and preserving spatial topological correlations.

eess.SP↗

arXiv2026

An LLM-Enabled Frequency-Aware Flow Diffusion Model for Natural-Language-Guided Power System Scenario Generation

Diverse and controllable scenario generation (e.g., wind, solar, load, etc.) is critical for robust power system planning and operation. As AI-based scenario generation methods are becoming the mainstream, existing methods (e.g., Conditional Generative Adversarial Nets) mainly rely on a fixed-length numerical conditioning vector to control the generation results, facing challenges in user conveniency and generation flexibility. In this paper, a natural-language-guided scenario generation framework, named LLM-enabled Frequency-aware Flow Diffusion (LFFD), is proposed to enable users to generate desired scenarios using plain human language. First, a pretrained LLM module is introduced to convert generation requests described by unstructured natural languages into ordered semantic space. Second, instead of using standard diffusion models, a flow diffusion model employing a rectified flow matching objective is introduced to achieve efficient and high-quality scenario generation, taking the LLM output as the model input. During the model training process, a frequency-aware multi-objective optimization algorithm is introduced to mitigate the frequency-bias issue. Meanwhile, a dual-agent framework is designed to create text-scenario training sample pairs as well as to standardize semantic evaluation. Experiments based on large-scale photovoltaic and load datasets demonstrate the effectiveness of the proposed method.

eess.SP↗

arXiv2025

A Causal-Guided Multimodal Large Language Model for Generalized Power System Time-Series Data Analytics

Power system time series analytics is critical in understanding the system operation conditions and predicting the future trends. Despite the wide adoption of Artificial Intelligence (AI) tools, many AI-based time series analytical models suffer from task-specificity (i.e. one model for one task) and structural rigidity (i.e. the input-output format is fixed), leading to limited model performances and resource wastes. In this paper, we propose a Causal-Guided Multimodal Large Language Model (CM-LLM) that can solve heterogeneous power system time-series analysis tasks. First, we introduce a physics-statistics combined causal discovery mechanism to capture the causal relationship, which is represented by graph, among power system variables. Second, we propose a multimodal data preprocessing framework that can encode and fuse text, graph and time series to enhance the model performance. Last, we formulate a generic "mask-and-reconstruct" paradigm and design a dynamic input-output padding mechanism to enable CM-LLM adaptive to heterogeneous time-series analysis tasks with varying sample lengths. Simulation results based on open-source LLM Qwen and real-world dataset demonstrate that, after simple fine-tuning, the proposed CM-LLM can achieve satisfying accuracy and efficiency on three heterogeneous time-series analytics tasks: missing data imputation, forecasting and super resolution.

eess.SP↗

arXiv2025

Deep-Learning-based Frequency-Domain Watermarking for Energy System Time Series Data Asset Protection

Data has been regarded as a valuable asset with the fast development of artificial intelligence technologies. In this paper, we introduce deep-learning neural network-based frequency-domain watermarking for protecting energy system time series data assets and secure data authenticity when being shared or traded across communities. First, the concept and desired watermarking characteristics are introduced. Second, a deep-learning neural network-based watermarking model with specially designed loss functions and network structure is proposed to embed watermarks into the original dataset. Third, a frequency-domain data preprocessing method is proposed to eliminate the frequency bias of neural networks when learning time series datasets to enhance the model performances. Last, a comprehensive watermarking performance evaluation framework is designed for measuring its invisibility, restorability, robustness, secrecy, false-positive detection, generalization, and capacity. Case studies based on practical load and photovoltaic time series datasets demonstrate the effectiveness of the proposed method.

eess.SP↗

arXiv2024

A Two-Stage AI-Powered Motif Mining Method for Efficient Power System Topological Analysis

Graph motif, defined as the microstructure that appears repeatedly in a large graph, reveals important topological characteristics of the large graph and has gained increasing attention in power system analysis regarding reliability, vulnerability and resiliency. However, searching motifs within the large-scale power system is extremely computationally challenging and even infeasible, which undermines the value of motif analysis in practice. In this paper, we introduce a two-stage AI-powered motif mining method to enable efficient and wide-range motif analysis in power systems. In the first stage, a representation learning method with specially designed network structure and loss function is proposed to achieve ordered embedding for the power system topology, simplifying the subgraph isomorphic problem into a vector comparison problem. In the second stage, under the guidance of the ordered embedding space, a greedy-search-based motif growing algorithm is introduced to quickly obtain the motifs without traversal searching. A case study based on a power system database including 61 circuit models demonstrates the effectiveness of the proposed method.

eess.SY↗

arXiv2024

A Glass-Box Deep-Learning Method for Electrical Energy System Modeling Based on Kolmogorov-Arnold Network

Deep learning methods have been widely used as an end-to-end modeling strategy of electrical energy systems because of their conveniency and powerful pattern recognition capability. However, due to the "closed-box" nature, deep learning methods have long been blamed for their poor interpretability when modeling a physical system. In this paper, we introduce a novel neural network structure, Kolmogorov-Arnold Network (KAN), to achieve "glass-box" modeling for electrical energy systems to enhance the interpretability. The most distinct feature of KAN lies in the learnable activation function together with the sparse training and symbolification process. Consequently, KAN can express the physical process with concise and explicit mathematical formulas while remaining the nonlinear-fitting capability of deep neural networks. Simulation results based on three electrical energy systems demonstrate the effectiveness of KAN in the aspects of interpretability, accuracy, robustness and generalization ability.

eess.SP↗

arXiv2024

Is Large Language Model Good at Database Knob Tuning? A Comprehensive Experimental Evaluation

Knob tuning plays a crucial role in optimizing databases by adjusting knobs to enhance database performance. However, traditional tuning methods often follow a Try-Collect-Adjust approach, proving inefficient and database-specific. Moreover, these methods are often opaque, making it challenging for DBAs to grasp the underlying decision-making process. The emergence of large language models (LLMs) like GPT-4 and Claude-3 has excelled in complex natural language tasks, yet their potential in database knob tuning remains largely unexplored. This study harnesses LLMs as experienced DBAs for knob-tuning tasks with carefully designed prompts. We identify three key subtasks in the tuning system: knob pruning, model initialization, and knob recommendation, proposing LLM-driven solutions to replace conventional methods for each subtask. We conduct extensive experiments to compare LLM-driven approaches against traditional methods across the subtasks to evaluate LLMs' efficacy in the knob tuning domain. Furthermore, we explore the adaptability of LLM-based solutions in diverse evaluation settings, encompassing new benchmarks, database engines, and hardware environments. Our findings reveal that LLMs not only match or surpass traditional methods but also exhibit notable interpretability by generating responses in a coherent ``chain-of-thought'' manner. We further observe that LLMs exhibit remarkable generalizability through simple adjustments in prompts, eliminating the necessity for additional training or extensive code modifications. Drawing insights from our experimental findings, we identify several opportunities for future research aimed at advancing the utilization of LLMs in the realm of database management.

cs.DB↗

arXiv2024

A Neural-Network-Embedded Equivalent Circuit Model for Lithium-ion Battery State Estimation

Equivalent Circuit Model(ECM)has been widelyused in battery modeling and state estimation because of itssimplicity, stability and interpretability.However, ECM maygenerate large estimation errors in extreme working conditionssuch as freezing environmenttemperature andcomplexcharging/discharging behaviors,in whichscenariostheelectrochemical characteristics of the battery become extremelycomplex and nonlinear.In this paper,we propose a hybridbattery model by embeddingneural networks as 'virtualelectronic components' into the classical ECM to enhance themodel nonlinear-fitting ability and adaptability. First, thestructure of the proposed hybrid model is introduced, where theembedded neural networks are targeted to fit the residuals of theclassical ECM,Second, an iterative offline training strategy isdesigned to train the hybrid model by merging the battery statespace equation into the neural network loss function. Last, thebattery online state of charge (SOC)estimation is achieved basedon the proposed hybrid model to demonstrate its applicationvalue,Simulation results based on a real-world battery datasetshow that the proposed hybrid model can achieve 29%-64%error reduction for $OC estimation under different operatingconditions at varying environment temperatures.

eess.SP↗

arXiv2024

Unsupervised and Interpretable Synthesizing for Electrical Time Series Based on Information Maximizing Generative Adversarial Nets

Generating synthetic data has become a popular alternative solution to deal with the difficulties in accessing and sharing field measurement data in power systems. However, to make the generation results controllable, existing methods (e.g. Conditional Generative Adversarial Nets, cGAN) require labeled dataset to train the model, which is demanding in practice because many field measurement data lacks descriptive labels. In this paper, we introduce the Information Maximizing Generative Adversarial Nets (infoGAN) to achieve interpretable feature extraction and controllable synthetic data generation based on the unlabeled electrical time series dataset. Features with clear physical meanings can be automatically extracted by maximizing the mutual information between the input latent code and the classifier output of infoGAN. Then the extracted features are used to control the generation results similar to a vanilla cGAN framework. Case study is based on the time series datasets of power load and renewable energy output. Results demonstrate that infoGAN can extract both discrete and continuous features with clear physical meanings, as well as generating realistic synthetic time series that satisfy given features.

eess.SP↗

arXiv2024

E2ETune: End-to-End Knob Tuning via Fine-tuned Generative Language Model

Database knob tuning is a significant challenge for database administrators, as it involves tuning a large number of configuration knobs with continuous or discrete values to achieve optimal database performance. Traditional methods, such as manual tuning or learning-based approaches, typically require numerous workload replays and are both time-consuming and resource-intensive. To address this challenge, we introduce E2ETune, an end-to-end knob tuner powered by a fine-tuned generative language model. The key idea is to leverage the exceptional sequence-to-sequence modeling capabilities of generative language models to capture the complex mapping between workloads (inputs) and their corresponding promising configurations (outputs). To achieve this goal, we propose a novel data generation framework to efficiently produce a large amount of training data, where each data sample consists of a workload and its promising configuration. Then, these data are used to fine-tune a generative language model, yielding an end-to-end knob tuner. This tuner offers out-of-the-box configuration recommendations for new workloads. We conduct extensive experiments to evaluate E2ETune's efficiency and effectiveness using 10 representative and 3 real-world benchmarks. Compared to state-of-the-art methods, E2ETune can identify competitive configurations in significantly less time.

cs.AI↗

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

MultiLoad-GAN: A GAN-Based Synthetic Load Group Generation Method Considering Spatial-Temporal Correlations

This paper presents a deep-learning framework, Multi-load Generative Adversarial Network (MultiLoad-GAN), for generating a group of synthetic load profiles (SLPs) simultaneously. The main contribution of MultiLoad-GAN is the capture of spatial-temporal correlations among a group of loads that are served by the same distribution transformer. This enables the generation of a large amount of correlated SLPs required for microgrid and distribution system studies. The novelty and uniqueness of the MultiLoad-GAN framework are three-fold. First, to the best of our knowledge, this is the first method for generating a group of load profiles bearing realistic spatial-temporal correlations simultaneously. Second, two complementary realisticness metrics for evaluating generated load profiles are developed: computing statistics based on domain knowledge and comparing high-level features via a deep-learning classifier. Third, to tackle data scarcity, a novel iterative data augmentation mechanism is developed to generate training samples for enhancing the training of both the classifier and the MultiLoad-GAN model. Simulation results show that MultiLoad-GAN can generate more realistic load profiles than existing approaches, especially in group level characteristics. With little finetuning, MultiLoad-GAN can be readily extended to generate a group of load or PV profiles for a feeder or a service area.

eess.SP↗

arXiv2023

A Novel Feeder-level Microgrid Unit Commitment Algorithm Considering Cold-load Pickup, Phase Balancing, and Reconfiguration

This paper presents a novel 2-stage microgrid unit commitment (Microgrid-UC) algorithm considering cold-load pickup (CLPU) effects, three-phase load balancing requirements, and feasible reconfiguration options. Microgrid-UC schedules the operation of switches, generators, battery energy storage systems, and demand response resources to supply 3-phase unbalanced loads in an islanded microgrid for multiple days. A performance-based CLPU model is developed to estimate additional energy needs of CLPU so that CLPU can be formulated into the traditional 2-stage UC scheduling process. A per-phase demand response budget term is added to the 1st stage UC objective function to meet 3-phase load unbalance limits. To reduce computational complexity in the 1st stage UC, we replace the spanning tree method with a feasible reconfiguration topology list method. The proposed algorithm is developed on a modified IEEE 123-bus system and tested on the real-time simulation testbed using actual load and PV data. Simulation results show that Microgrid-UC successfully accounts for CLPU, phase imbalance, and feeder reconfiguration requirements.

eess.SY↗

arXiv2022

Load Profile Inpainting for Missing Load Data Restoration and Baseline Estimation

This paper introduces a Generative Adversarial Nets (GAN) based, Load Profile Inpainting Network (Load-PIN) for restoring missing load data segments and estimating the baseline for a demand response event. The inputs are time series load data before and after the inpainting period together with explanatory variables (e.g., weather data). We propose a Generator structure consisting of a coarse network and a fine-tuning network. The coarse network provides an initial estimation of the data segment in the inpainting period. The fine-tuning network consists of self-attention blocks and gated convolution layers for adjusting the initial estimations. Loss functions are specially designed for the fine-tuning and the discriminator networks to enhance both the point-to-point accuracy and realisticness of the results. We test the Load-PIN on three real-world data sets for two applications: patching missing data and deriving baselines of conservation voltage reduction (CVR) events. We benchmark the performance of Load-PIN with five existing deep-learning methods. Our simulation results show that, compared with the state-of-the-art methods, Load-PIN can handle varying-length missing data events and achieve 15-30% accuracy improvement.

eess.SP↗

arXiv2022

Feeder Microgrid Management on an Active Distribution System during a Severe Outage

Forming a microgrid on a distribution system with large scale outage after a severe weather event is emerging as a viable solution to improve resiliency at the distribution level. This option becomes more attractive when the distribution system has high levels of distributed PV. The management of such feeder-level microgrid has however many challenges, such as limited resources that can be deployed on the feeder quickly, and the limited real-time monitoring and control on the distribution system. Effective use of the distributed PV is also challenging as they are not monitored and controlled. To handle these challenges, the paper proposes a 2-stage hierarchical energy management scheme to securely operate these feeder level micorgrids. The first stage of the scheme solves a sequential rolling optimization problem to optimally schedule the main resources (such as a mobile diesel generator and battery storage unit). The second stage adopts a dispatching scheme for the main resources to adjust the stage-1 set-points closer to real- time. The proposed scheme has unique features to assure that the scheme is robust under highly varying operating conditions with limited system observability: (i) an innovative PV forecast error adjustment and a dynamic reserve adjustment scheme to handle the extreme uncertainty on PV power output, and (ii) an intelligent fuel management scheme to assure that the resources are utilized optimally over the multiple days of the restoration period. The proposed algorithm is tested on sample system with real-time data. The results show that the proposed scheme performs well in maximizing service to loads by effective use of all the resources and by properly taking into account the challenging operating conditions.

eess.SY↗

arXiv2022

A TCN-based Hybrid Forecasting Framework for Hours-ahead Utility-scale PV Forecasting

This paper presents a Temporal Convolutional Network (TCN) based hybrid PV forecasting framework for enhancing hours-ahead utility-scale PV forecasting. The hybrid framework consists of two forecasting models: a physics-based trend forecasting (TF) model and a data-driven cloud-event forecasting (CF) model. Three TCNs are integrated in the framework for: i) blending the inputs from different Numerical Weather Prediction sources for the TF model to achieve superior performance on forecasting hourly PV profiles, ii) capturing spatial-temporal correlations between detector sites and the target site in the CF model to achieve more accurate forecast of intra-hour PV power drops, and iii) reconciling TF and CF results to obtain coherent hours-ahead PV forecast with both hourly trends and intra-hour fluctuations well preserved. To automatically identify the most contributive neighboring sites for forming a detector network, a scenario-based correlation analysis method is developed, which significantly improves the capability of the CF model on capturing large power fluctuations caused by cloud movements. The framework is developed, tested, and validated using actual PV data collected from 95 PV farms in North Carolina. Simulation results show that the performance of 6 hours ahead PV power forecasting is improved by approximately 20% compared with state-of-the-art methods.

eess.SP↗