J.L. Cremer
Please Note
66 records found
1
Network topology optimization (NTO) via busbar splitting can mitigate transmission grid congestion and reduce redispatch costs. However, solving this mixed-integer nonlinear problem for large-scale systems in near-real-time is currently intractable with existing solvers. Machine learning (ML) approaches have emerged as a promising alternative, but they have limited generalization to unseen topologies, varying operating conditions, and different systems, which limits their practical applicability. This paper formulates NTO for congestion management considering linearized AC power flow, and proposes a graph neural network (GNN)-accelerated approach. We develop a heterogeneous edge-aware message passing GNN to predict effective nodes for busbar splitting actions as candidate NTO solutions. The proposed GNN captures local flow patterns, improves generalization to unseen topology changes, and enhances transferability across systems. Case studies show up to 104 × speed-up, delivering AC-feasible solutions within one minute and a 2.3% optimality gap on the GOC 2000-bus system. These results demonstrate a significant step toward near-real-time NTO for large-scale systems with topology and cross-system generalization.
The electrification and ongoing energy transition lead to systematic changes in electricity loading and variability in power systems. Distribution systems were designed for regular operating patterns, assuming constant low loading. Now, operators need to assess whether their assets can withstand more, as well as time-varying loading. Operating the system at or near its ampacity potentially accelerates thermal ageing, so the question arises: ‘how much can one operate at the limits while keeping maintenance and failures low?’ This paper introduces a novel approach that derives a time-varying Weibull approximation of failure rates using thermal models and provides a shortcut method to quantify maintenance implications under time-varying loading for heterogeneous MV cable populations. The case studies investigate a dataset from Denmark and the Oberrhein Medium Voltage (MV) system in Germany, studying ageing assets and the interplay with loading, and replacement paradigms of two different cable insulation types. The studies demonstrate that a small fraction of 25% of old, low-quality cables leads to 82% of failures, and 1.4% of the time of highest loading can cause 46% of cable ageing. The case studies also demonstrate that maintenance needs may be between 10-300 times higher under future loading conditions associated with the energy transition, specifically in networks that have older PILC cables. This paper provides a new tool for operators to plan maintenance under more realistic, future operating conditions.
Suppose one is interested in identifying the weakest link of the electrical system at 3 simultaneous faults caused by an extreme weather event. Current techniques cannot identify this; however, knowing such information can help reinforce the system at the weakest link to increase system security. Current techniques typically apply a forward process to the security assessment: assign a contingency list, study its impact, analyse the list to obtain a shortlist, and improve the system. However, this process does not scale well to the number of contingencies, specifically, when one is interested in the combination of k faults, as the process needs to run once per combination. This paper proposes a new backwards process using quantum effects from quantum annealing (QA). Our proposal formulates a quadratic optimisation to find the worst-case N-k contingency using disjunctive power transfer distribution factors. Then, we use a quantum annealer and propose a search algorithm to solve the problem, using the distribution of solutions to obtain the shortlist of contingencies. We propose a meta-heuristic to make the approach feasible on quantum computers with limited qubits. The case studies focus on the IEEE 118-bus system, showing a 200 x speed-up for 50 faults compared to exhaustive search. The case study extrapolates to the 2383wp system, showing the approach scales well to larger power systems; however, the current quantum hardware limits the number of single faults to consider to around 50. Case studies demonstrate weak lines can be identified for reinforcement by analysing the QA solution distribution, potentially improving system security for multiple N-k faults.
Configure-and-Bound
A fast heuristic for network topology reconfiguration
Substation reconfiguration via busbar splitting can mitigate transmission grid congestion and reduce operational costs. However, existing approaches neglect the security of substation topology, particularly for substations without busbar splitting (i.e., closed couplers), which can lead to severe consequences. Additionally, the computational complexity of optimizing substation topology remains a challenge. This paper introduces a MILP formulation for security-constrained substation reconfiguration (SC-SR), considering N-1 line, coupler and busbar contingencies to ensure secure substation topology. To efficiently solve this problem, we propose a heuristic approach with multiple master problems (HMMP). A central master problem optimizes dispatch, while independent substation master problems determine individual substation topologies in parallel. Linear AC power flow equations ensure PF accuracy, while feasibility and optimality sub-problems evaluate contingency cases. The proposed HMMP significantly reduces computational complexity and enables scalability to large-scale power systems. Case studies on the IEEE 14-bus, 118-bus, and PEGASE 1354-bus system show the effectiveness of the approach in mitigating the impact of coupler and busbar tripping, balancing system security and cost, and computational efficiency.
Coordination between transmission system operators (TSOs) and distribution system operators (DSOs) can support TSOs in using the distribution system (DS) flexibility while ensuring feasible operation. Flexibility areas (FAs) can support TSO-DSO coordination, aggregating the total feasible flexibility within the DS. However, existing real-time estimation approaches do not consider the limited measurements within DS. This paper proposes a Bayesian neural network (BNN) to estimate the operating conditions that bound the operational flexibility, including epistemic and aleatoric uncertainties. These uncertainties stem from the limited real-time measurements in DSs and the measurement noise. TSOs can select a threshold that confirms a probability of safety, considering uncertainty margins. The paper also provides FA estimation in DS topologies with (Formula presented.) points of common coupling (PCC) with the transmission system. Case studies in the CIGRE and Oberrhein networks compare the proposed BNNs to baseline statistic-based approaches for forecast and measurement uncertainty in FAs. The case studies show the proposed FA estimation under various safety margins and systems with 2-PCC. Case studies also assess various measurement noise levels and evaluate model performance for different DS topologies.
AC Optimal Power Flow (ACOPF) is computationally intensive for large-scale grids, often requiring prohibitive solution times with conventional solvers. Machine learning offers significant speedups, but existing models struggle with scalability and topology flexibility. To address these challenges, we propose a Hybrid Heterogeneous Message Passing Neural Network (HH-MPNN) that integrates a heterogeneous graph neural network (GNN) with a scalable transformer and physics-informed positional encodings. Our approach differs from previous GNN-based ACOPF approaches, which rely on local message passing alone. Our architecture explicitly models distinct power system components to capture local features while using global attention for long-range dependencies. Evaluated on diverse benchmarks including PGLearn and GridFM-DataKit datasets, HH-MPNN achieves <1% optimality gap on default topologies across grid sizes from 14 to 2000 buses on average. For N-1 contingencies, our approach demonstrates zero-shot N-1 generalization with <3% optimality gap on several test cases despite training only on default topologies. We further develop an approach that ensures robust N-1 generalization to high-impact contingencies through targeted augmentation of the training data, showing that exhaustive simulation is unnecessary for topologically flexible models. Finally, size generalization experiments demonstrate that pre-training on small grids significantly improves performance on large-scale systems. Achieving computational speedups of up to 5000 times compared to interior point solvers, these results advance practical, generalizable machine learning for real-time power system operations.
Batteries with silicon-graphite-based anodes, which offer higher energy density and improved charging performance, introduce pronounced voltage hysteresis, making state-of-charge (SoC) estimation particularly challenging. Existing approaches to modeling hysteresis rely on exhaustive high-fidelity tests or focus on conventional graphite-based lithium-ion batteries, without considering uncertainty quantification or computational constraints. This work introduces a data-driven approach for probabilistic hysteresis factor prediction, with a particular emphasis on applications involving silicon-graphite anode-based batteries. A data harmonization framework is proposed to standardize heterogeneous driving cycles across varying operating conditions. Statistical learning and deep learning models are applied to assess performance in predicting the hysteresis factor with uncertainties while considering computational efficiency. Extensive experiments are conducted to evaluate the generalizability of the optimal model configuration in unseen vehicle models through retraining, zero-shot prediction, fine-tuning, and joint training. By addressing key challenges in SoC estimation, this research facilitates the adoption of advanced battery technologies.
OrderFusion
Encoding orderbook for end-to-end probabilistic intraday electricity price forecasting
Ground fault detection in inverter-based microgrid (IBM) systems is challenging, particularly in a real-time setting, as the fault current deviates slightly from the nominal value. This difficulty is reinforced when there are partially decoupled disturbances and modeling uncertainties. The conventional solution of installing more relays to obtain additional measurements is costly and also increases the complexity of the system. In this brief, we propose a data-assisted diagnosis scheme based on an optimization-based fault detection filter with the output current as the only measurement. Modeling the microgrid dynamics and the diagnosis filter, we formulate the filter design as a quadratic programming (QP) problem that accounts for decoupling partial disturbances, robustness to nondecoupled disturbances and modeling uncertainties by training with data, and ensuring fault sensitivity simultaneously. To ease the computational effort, we also provide an approximate but analytical solution to this QP. Additionally, we use classical statistical results to provide a thresholding mechanism that enjoys probabilistic false-alarm guarantees. Finally, we implement the IBM system with Simulink and real-time digital simulator (RTDS) to verify the effectiveness of the proposed method through simulations.
TensorConvolutionPlus
A python package for distribution system flexibility area estimation
Power system operators need new, efficient operational tools to use the flexibility of distributed resources and deal with the challenges of highly uncertain and variable power systems. Transmission system operators can consider the available flexibility in distribution systems (DSs) without breaching the DS constraints through flexibility areas. However, there is an absence of open-source packages for flexibility area estimation. This paper introduces TensorConvolutionPlus, a user-friendly Python-based package for flexibility area estimation. The main features of TensorConvolutionPlus include estimating flexibility areas using the TensorConvolution+ algorithm, the power flow-based algorithm, an exhaustive PF-based algorithm, and an optimal power flow-based algorithm. Additional features include adapting flexibility area estimations from different operating conditions and including flexibility service providers offering discrete setpoints of flexibility. The TensorConvolutionPlus package facilitates a broader adaptation of flexibility estimation algorithms by system operators and power system researchers.
Learning to control a battery through reinforcement
Balancing lifetime and profit
Battery energy storage systems offer control over energy use and enable energy arbitrage (EA) helping to lower energy costs. However, battery owners currently fail to optimally exploit these systems for EA as the battery lifetime decreases, and many EA approaches incorrectly assume constant battery capacity. Battery performance declines over time resulting in reduced capacity that limits the economic benefits. Therefore, considering battery degradation is key to balancing economic profit and lifetime. In response, this work applies reinforcement learning to control a battery providing residential EA services and proposes a semi-supervised learning model to consider degradation. Case studies investigate three scenarios: 1) the approach is trained on a battery with an unrealistic constant maximum capacity to serve as a baseline, 2) the actions from the first scenario are applied to a real-world environment with a battery experiencing capacity decay to acknowledge the effect of neglecting degradation and 3) the approach considers a battery with a real decreasing capacity. Results show not considering degradation when operating a battery (scenario 2), leads to profits 13% lower than the ones obtained in the ideal case (scenario 1). If degradation is considered (scenario 3), the profits are only 4% lower than the profits obtained in the ideal case (scenario 1) and the battery's lifetime is extended by 20% compared to the lifetime achieved when not considering degradation (scenario 2).
Transmission network topology control offers cheap flexibility to system operators for mitigating grid congestion. However, finding the optimal sequence of topology actions remains a challenge due to the large number of possible actions. Although reinforcement learning (RL) approaches have attracted interest for long-term planning in large combinatorial action spaces, they encounter challenges such as training stability, sample efficiency, and unforeseen consequences of RL actions. Addressing these challenges, this paper proposes a hybrid curriculum-trained RL and Monte Carlo tree search (MCTS) approach to determine sequential topological actions for mitigating grid congestion. The curriculum-based approach stabilizes training by first pre-training a policy network through supervised imitation learning, followed by RL training. The policy network guides the MCTS to simulate promising future trajectories, mitigating unforeseen consequences and identifying long-term strategies to improve grid security. Moreover, the MCTS-verified actions are used for RL training, enhancing sample efficiency and training time. A distance factor is added to the MCTS, which improves convergence by prioritizing actions closer to congestion. Numerical results on the IEEE 118-bus system show that the proposed hybrid approach improves the timesteps survived by 30% compared to a standard RL approach, and by 5% compared to a brute-force baseline. Additionally, the inclusion of the distance factor increases the timesteps survived by 15%. These results highlight the potential of the proposed method for real-world applications of using sequential topological actions to effectively relieve grid congestion.
Coordination between power system operators can improve the power system stability and effectively deploy resources in distribution systems (DS). The research work of this paper provides a coordination method to mitigate the impact of dynamic events on transmission systems (TS). The proposed method uses a machine learning (ML)-based model to estimate the collective dynamic response of DS under varying TS dynamic properties, DS operating conditions, and share of inverter base resources (IBRs). In addition, the ML-based model enables TS operators (TSOs) to provide feedback to DS operators (DSOs) for controlling the IBRs’ active power output to prevent post-fault instabilities. The proposed TSO-DSO coordination method includes a risk-based active power setpoint optimizer for instability prevention. The proposed method uses existing measurement and IBR control platforms available in DS and estimates the post-fault DS dynamic response considering IBR active power control actions. Case studies on synthetic models of TS and DS covering the Zeeland province in The Netherlands illustrate the application of the proposed coordination and the instability risk mitigation when optimizing IBR setpoints.
Increasing renewable energy supply and distributed generating sources in the power grid lead to lower inertia levels. Lower inertia combined with higher uncertainty in operation can cause drastic frequency fluctuations when a disturbance occurs. System operators must know whether the transmission system is secure against a disturbance. Real-time models attempt to predict the frequency in near-real time however require pretraining on a large variety of possible disturbances. Training models in real-time would not require pre-training as they are directly trained on the occurring disturbance. However, training in real-time is not feasible until now as fast-occurring system dynamics require shorter prediction (and training) times for security and operation which standard machine learning models are not capable of. For the first time, this work proposes a fast training strategy that learns Neural Ordinary Differential Equations (NODE) in near real-time directly on the occurring disturbance, simultaneously addressing the inaccuracy issue of model-based dynamic studies. NODE learns the dynamics or derivatives of an ODE system that standard ODE solvers can solve. NODE provides a continuous function for predicting future dynamics in a decentralized way, hence faster frequency stability assessment for longer time spans. We propose a collocation-based sampling using the collocation gradients. Case studies on the IEEE 39-bus system show the approach is feasible for near real-time operation, accurately predicts future system states, and enables operators to apply predesigned corrective control actions, potentially making the system secure for future disturbances.