arXiv subjects

Mingjian Tuo

Publications and source records attributed to Mingjian Tuo.

15 recordsLinked to original sources

arXiv2026

Economic Evaluation of V2G-Enabled Fast Charging Stations Under Endogenous EV Adoption Dynamics

Building fast charging stations (FCSs) is crucial for transportation electrification, but there exists an indirect network effect: while the increasing number of electric vehicles (EVs) decides the FCS capacity expansion, the spatial locations of these facilities strongly influence drivers' willingness to adopt EVs. Ignoring this interaction can lead to bad capital investments and exacerbate power grid vulnerabilities during tidal traffic peaks. Therefore, we explicitly model the EV adoption dynamics as decision-dependent uncertainties (DDUs) in a new multi-period collaborative planning framework. This framework evaluates the economic viability of V2G-enabled FCSs across both transportation and distribution networks. To simplify the complex calculation, we introduce an aggregated fleet virtual battery model to catch macroscopic vehicle-to-grid (V2G) flexibility. This successfully circumvents the dimension curse inherent in tracking microscopic state-of-charge. To further guarantee calculation speed, the nonlinear infrastructure exposure is transformed into a mixed-integer program by using Special Ordered Set type 2 (SOS2) variables and Second-Order Cone Programming (SOCP) relaxations for grid limits. Finally, numerical studies on a coupled Sioux Falls and IEEE 33-bus testbed prove that our framework achieves superior expected social welfare. Also, macroscopic V2G aggregation is highlighted for its capability to mitigate distribution grid congestion penalties.

eess.SY↗

arXiv2026

Inertia-Constrained Generation Scheduling: Sample Selection, Learning-Embedded Optimization Modeling, and Computational Enhancement

Day-ahead generation scheduling is typically conducted by solv-ing security-constrained unit commitment (SCUC) problem. However, with fast-growing of inverter-based resources, grid inertia has been dramatically reduced, compromising the dy-namic stability system. Traditional SCUC (T-SCUC), without any inertia requirements, may no longer be effective for renewa-bles-dominated grids. To address this, we propose the active linearized sparse neural network-embedded SCUC (ALSNN-SCUC) model, utilizing machine learning (ML) to incorporate system dynamic performance. A multi-output deep neural net-work (DNN) model is trained offline on strategically-selected data samples to accurately predict frequency stability metrics: locational RoCoF and frequency nadir. Structured sparsity and active ReLU linearization are implemented to prune redundant DNN neurons, significantly reducing its size while ensuring pre-diction accuracy even at high sparsity levels. By embedding this ML-based frequency stability predictor into SCUC as con-straints, the proposed ALSNN-SCUC model minimizes its com-putational complexity while ensuring frequency stability follow-ing G-1 contingency. Case studies show that the proposed ALSNN-SCUC can enforce pre-specified frequency requirements without being overly conservative, outperforming five bench-mark models including T-SCUC, two physics-based SCUC, and two ML-based SCUC. The proposed sparsification and active linearization strategies can reduce the DNN-SCUC computing time by over 95% for both IEEE 24-bus and 118-bus systems, demonstrating the effectiveness and scalability of the proposed ALSNN-SCUC model.

eess.SY↗

arXiv2026

Machine Learning-assisted Dynamics-Constrained Day-Ahead Energy Scheduling

TThe rapid expansion of inverter-based resources, such as wind and solar power plants, will significantly diminish the presence of conventional synchronous generators in fu-ture power grids with rich renewable energy sources. This transition introduces in-creased complexity and reduces dynamic stability in system operation and control, with low inertia being a widely recognized challenge. However, the literature has not thoroughly explored grid dynamic performance associated with energy scheduling so-lutions that traditionally only consider grid steady-state constraints. This paper will bridge the gap by enforcing grid dynamic constraints when conducting optimal energy scheduling; particularly, this paper explores locational post-contingency rate of change of frequency (RoCoF) requirements to accommodate substantial inertia reductions. This paper introduces a machine learning-assisted RoCoF-constrained unit commit-ment (ML-RCUC) model designed to ensure RoCoF stability after the most severe generator outage while maintaining operational efficiency. A graph-informed NN (GINN)-based RoCoF predictor is first trained on a high-fidelity simulation dataset to track the highest locational RoCoF, which is then reformulated as mixed-integer linear programming constraints that are integrated into the unit commitment model. Case studies, by solving the optimization problem ML-RCUC and validating its solutions with time-domain simulations, demonstrate that the proposed method can ensure loca-tional RoCoF stability with minimum conservativeness.

eess.SY↗

arXiv2024

Comparative Analysis of Machine Learning Models for Short-Term Distribution System Load Forecasting

Accurate electrical load forecasting is crucial for optimizing power system operations, planning, and management. As power systems become increasingly complex, traditional forecasting methods may fail to capture the intricate patterns and dependencies within load data. Machine learning (ML) techniques have emerged as powerful alternatives, offering superior prediction accuracy and the ability to model non-linear and complex temporal relationships. This study presents a comprehensive comparison of prominent ML models: feedforward neural networks, recurrent neural networks, long short-term memory networks, gated recurrent units, and the attention temporal graph convolutional network; for short-term load forecasting of the Energy Corridor distribution system in Houston, Texas. Using a 24-hour look-back window, we train the models on datasets spanning one and five years, to predict the load demand for the next hour and assess performance. Our findings aim to identify the most effective ML approach for accurate load forecasting, contributing to improved grid reliability and system optimization.

eess.SY↗

arXiv2023

Analysis of Weather and Time Features in Machine Learning-aided ERCOT Load Forecasting

Accurate load forecasting is critical for efficient and reliable operations of the electric power system. A large part of electricity consumption is affected by weather conditions, making weather information an important determinant of electricity usage. Personal appliances and industry equipment also contribute significantly to electricity demand with temporal patterns, making time a useful factor to consider in load forecasting. This work develops several machine learning (ML) models that take various time and weather information as part of the input features to predict the short-term system-wide total load. Ablation studies were also performed to investigate and compare the impacts of different weather factors on the prediction accuracy. Actual load and historical weather data for the same region were processed and then used to train the ML models. It is interesting to observe that using all available features, each of which may be correlated to the load, is unlikely to achieve the best forecasting performance; features with redundancy may even decrease the inference capabilities of ML models. This indicates the importance of feature selection for ML models. Overall, case studies demonstrated the effectiveness of ML models trained with different weather and time input features for ERCOT load forecasting.

cs.LG↗

arXiv2023

Graph Neural Network-based Power Flow Model

Power flow analysis plays a crucial role in examining the electricity flow within a power system network. By performing power flow calculations, the system's steady-state variables, including voltage magnitude, phase angle at each bus, active/reactive power flow across branches, can be determined. While the widely used DC power flow model offers speed and robustness, it may yield inaccurate line flow results for certain transmission lines. This issue becomes more critical when dealing with renewable energy sources such as wind farms, which are often located far from the main grid. Obtaining precise line flow results for these critical lines is vital for next operations. To address these challenges, data-driven approaches leverage historical grid profiles. In this paper, a graph neural network (GNN) model is trained using historical power system data to predict power flow outcomes. The GNN model enables rapid estimation of line flows. A comprehensive performance analysis is conducted, comparing the proposed GNN-based power flow model with the traditional DC power flow model, as well as deep neural network (DNN) and convolutional neural network (CNN). The results on test systems demonstrate that the proposed GNN-based power flow model provides more accurate solutions with high efficiency comparing to benchmark models.

eess.SY↗

arXiv2023

Selectively Linearized Neural Network based RoCoF-Constrained Unit Commitment in Low-Inertia Power Systems

Conventional synchronous generators are gradually being replaced by inverter-based resources, such transition introduces more complicated operation conditions. And the reduction in system inertia imposes challenges for system operators on maintaining system rate-of-change-of-frequency (RoCoF) security. This paper presents a selectively linearized neural network (SNLNN) based RoCoF-constrained unit commitment (SLNN-RCUC) model. A RoCoF predictor is first trained to predict the system wide highest locational RoCoF based on a high-fidelity simulation dataset. Instead of incorporating the complete neural network into unit commitment, a ReLU linearization method is implemented on active selected neurons to improve the algorithm computational efficiency. The effectiveness of proposed SLNN-RCUC model is demonstrated on the IEEE 24-bus system by conducting time domain simulation on PSS/E

eess.SY↗

arXiv2022

Machine Learning Assisted Inertia Estimation using Ambient Measurements

With the increasing penetration of converter-based renewable resources, different types of dynamics have been introduced to the power system. Due to the complexity and high order of the modern power system, mathematical model-based inertia estimation method becomes more difficult. This paper proposes two novel machine learning assisted inertia estimation methods based on long-recurrent convolutional neural (LRCN) network and graph convolutional neural (GCN) network respectively. Informative features are extracted from ambient measurements collected through phasor measurement units (PMU). Spatial structure with high dimensional features and graphical information are then incorporated to improve the accuracy of the inertia estimation. Case studies are conducted on the IEEE 24-bus system. The proposed LRCN and GCN based inertia estimation models achieve an accuracy of 97.34% and 98.15% respectively. Furthermore, the proposed zero generation injection bus based optimal PMU placement (ZGIB-OPP) has been proved to be able to maximize the system observability, which subsequently improves the performance of all proposed inertia estimation models.

eess.SY↗

arXiv2022

Deep Learning based Security-Constrained Unit Commitment Considering Locational Frequency Stability in Low-Inertia Power Systems

With the goal of electricity system decarbonization, conventional synchronous generators are gradually replaced by converter-interfaced renewable generations. Such transition is causing concerns over system frequency and rate-of-change-of-frequency (RoCoF) security due to significant reduction in system inertia. Existing efforts are mostly derived from uniform system frequency response model which may fail to capture all characteristics of the systems. To ensure the locational frequency security, this paper presents a deep neural network (DNN) based RoCoF-constrained unit commitment (DNN-RCUC) model. RoCoF predictor is trained to predict the highest locational RoCoF based on a high-fidelity simulation dataset. Training samples are generated from models over various scenarios, which can avoid simulation divergence and system instability. The trained network is then reformulated into a set of mixed-integer linear constraints representing the locational RoCoF-limiting constraints in unit commitment. The proposed DNN-RCUC model is studied on the IEEE 24-bus system. Time domain simulation results on PSS/E demonstrate the effectiveness of the proposed algorithm.

eess.SY↗

arXiv2022

Security-Constrained Unit Commitment Con-sidering Locational Frequency Stability in Low-Inertia Power Grids

With increasing installation of wind and solar generation, conventional synchronous generators in power systems are gradually displaced resulting in a significant reduction in system inertia. Maintaining system frequency within acceptable ranges becomes more critical for the stability of a power system. In this paper, we first study the impact of inter-area oscillations on the system rate-of-change-of-frequency (RoCoF) security; then, the limitations on locational RoCoFs accounting for G-1 contingency stability are derived. By enforcing these frequency related constraints, a location based RoCoF constrained security constrained unit commitment (LRC-SCUC) model is proposed. Furthermore, an effective piecewise linearization (PWL) technique is employed to formulate a RoCoF linearization problem and linearize the nonlinear function representing the location based RoCoF constraints in SCUC. Simulation results reveal that the inclusion of inertia-related constraints can substantially improve the system stability at the cost of higher operation cost. The results also show that deploying virtual inertia techniques not only reduces the total cost, but also improves the system market efficiency.

eess.SY↗

arXiv2021

Wholesale Electricity Price Forecasting using Integrated Long-term Recurrent Convolutional Network Model

Electricity price is a key factor affecting the decision-making for all market participants. Accurate forecasting of electricity prices is very important and is also very challenging since electricity price is highly volatile due to various factors. This paper proposes an integrated long-term recurrent convolutional network (ILRCN) model to predict electricity prices considering the majority contributing attributes to the market price as input. The proposed ILRCN model combines the functionalities of convolutional neural network and long short-term memory (LSTM) algorithm along with the proposed novel conditional error correction term. The combined ILRCN model can identify the linear and non-linear behavior within the input data. We have used ERCOT wholesale market price data along with load profile, temperature, and other factors for the Houston region to illustrate the proposed model. The performance of the proposed ILRCN electricity price forecasting model is verified using performance/evaluation metrics like mean absolute error and accuracy. Case studies reveal that the proposed ILRCN model is accurate and efficient in electricity price forecasting as compared to the support vector machine (SVM) model, fully-connected neural network model, LSTM model and the LRCN model without the conditional error correction stage.

cs.LG↗

arXiv2021

Long-Term Recurrent Convolutional Network-based Inertia Estimation using Ambient Measurements

Conventional synchronous machines are gradually replaced by converter-based renewable resources. As a result, synchronous inertia, an important time-varying quantity, has substantially more impact on modern power systems stability. The increasing integration of renewable energy resources imports different dynamics into traditional power systems; therefore, the estimation of system inertia using mathematical model becomes more difficult. In this paper, we propose a novel learning-assisted inertia estimation model based on long-term recurrent convolutional network (LRCN) that uses system wide frequency and phase voltage measurements. The proposed approach uses a non-intrusive probing signal to perturb the system and collects ambient measurements with phasor measurement units (PMU) to train the proposed LRCN model. Case studies are conducted on the IEEE 24-bus system. Under a signal-to-noise ratio (SNR) of 60dB condition, the proposed LRCN based inertia estimation model achieves an accuracy of 97.56% with a mean squared error (MSE) of 0.0552. Furthermore, with a low SNR of 45dB, the proposed learning-assisted inertia estimation model is still able to achieve a high accuracy of 93.07%.

eess.SY↗

arXiv2021

Optimal Allocation of Virtual Inertia Devices for Enhancing Frequency Stability in Low-Inertia PowerSystems

As renewable resources gradually replace conventional generation based synchronous machines, the dynamics of the modern grid changes significantly and the system synchronous inertia decreases substantially. This transformation poses severe challenges for power system stability; for instance, it may lead to larger initial rate of change of frequency and increase frequency excursions. However, new opportunities also arise as novelconverter control techniques, so-called grid-forming strategies, show higher efficiency and faster response than conventional synchronous generators. They mainly involve virtual inertia (VI) emulation to mimic the behavior of synchronous machines. In this study, a state-space model for the power system network is developed with VI as a frequency regulation method. A reduced model based-norm algorithm (RMHA) considering the Fiedler mode impact is proposed in this paper to optimize the allocation of VI devices and improve power system frequency stability. Finally, case studies conducted on the IEEE 24-bus system demonstrate the efficacy of the proposed RMHA approach.

eess.SY↗

arXiv2020

Dynamic Estimation of Power System Inertia Distribution Using Synchrophasor Measurements

Integration of intermittent renewable energy sources in modern power systems is increasing very fast. Replacement of synchronous generators with zero-to-low variable renewables substantially decreases the system inertia. In a large system, inertia distribution may vary significantly, areas of low inertia are more susceptible to frequency deviation, posing risks of load shedding and generation trip. Therefore, it is necessary for operators to evaluate and quantify the system inertia and its distribution in real time. This paper proposes a novel synchronized phasor measurement units (PMUs)-based dynamic system inertia estimation method. The proposed inertia estimation method is derived using electrical distance and clustering algorithm, which considers the impact of location of measurements relative to in-feed load and impact of oscillations. The center of inertia (COI) area and area of low inertia are also determined during the estimation. Numerical simulations are conducted on the IEEE 24-bus system with various load profiles using Transient Security Analysis Tools (TSAT), a core module of the DSATools, which demonstrate the efficacy of the proposed approach.

eess.SP↗

arXiv2020

Benefits and Cyber-Vulnerability of Demand Response System in Real-Time Grid Operations

With improvement in smart grids through two-way communication, demand response (DR) has gained significant attention due to the inherent flexibility provided by shifting non-critical loads from peak periods to off-peak periods, which can greatly improve grid reliability and reduce cost of energy. Operators utilize DR to enhance operational flexibility and alleviate network congestion. However, the intelligent two-way communication is susceptible to cyber-attacks. This paper studies the benefits of DR in security-constrained economic dispatch (SCED) and then the vulnerability of the system to line overloads when cyber-attack targets DR signals. This paper proposes a false demand response signal and load measurement injection (FSMI) cyber-attack model that sends erroneous DR signals while hacking measurements to make the attack undetectable. Simulation results on the IEEE 24-bus system (i) demonstrate the cost-saving benefits of demand response, and (ii) show significant line overloads when the demand response signals are altered under an FSMI attack.

eess.SY↗