Search for: All records

Reinforcement learning (RL) has shown great potential for designing voltage control policies, but their performance often degrades under changing system conditions such as topology reconfigurations and load variations. We introduce a topology-aware online policy optimization framework that leverages data-driven estimation of voltage-reactive power sensitivities to achieve efficient policy adaptation for medium-voltage radial distribution networks. Exploiting the sparsity of topology-switching events, where only a few lines change at a time, our method efficiently detects topology changes and identifies the affected lines and parameters, enabling fast and accurate sensitivity updates without recomputing the full sensitivity matrix. The estimated sensitivity is subsequently used for online policy optimization of a pre-trained neural-network-based RL controller. Simulations on both the IEEE 13-bus and SCE 56-bus systems demonstrate over 90% line identification accuracy, using only 15 data points. The proposed method also significantly improves voltage regulation performance compared with non-adaptive policies and adaptive policies that rely on regression-based online optimization methods for sensitivity estimation. 

Real-world control systems require policies that are not only high-performing but also interpretable and robust. A promising direction toward this goal is model-based control, which learns system dynamics and cost functions from historical data and then uses these models to inform decision-making. Building on this paradigm, we introduce DiffOP, a novel framework for learning optimization-based control policies defined implicitly through optimization control problems. Without relying on value function approximation, DiffOP jointly learns the cost and dynamics models and directly optimizes the actual control costs using policy gradients. To enable this, we derive analytical policy gradients by applying implicit differentiation to the underlying optimization problem and integrating it with the standard policy gradient framework. Under standard regularity conditions, we establish that DiffOP converges to an epsilon-stationary point within O(1/epsilon) iterations. We demonstrate the effectiveness of DiffOP through experiments on nonlinear control tasks and power system voltage control with constraints. 

Energy-efficient ventilation control plays a vital role in reducing building energy consumption while ensuring occupant health and comfort. While Computational Fluid Dynamics (CFD) simulations provide detailed and physically accurate representations of indoor airflow, their high computational cost limits their use in real-time building control. In this work, we present a neural operator learning framework that combines the physical accuracy of CFD with the computational efficiency of machine learning to enable building ventilation control with the high-fidelity fluid dynamics models. Our method jointly optimizes the airflow supply rates and vent angles to reduce energy use and adhere to air quality constraints. We train an ensemble of neural operator transformer models to learn the mapping from building control actions to airflow fields using high-resolution CFD data. This learned neural operator is then embedded in an optimization-based control framework for building ventilation control. Experimental results show that our approach achieves significant energy savings compared to maximum airflow rate control, rule-based control, as well as data-driven control methods using spatially averaged CO2 prediction and deep learning–based reduced-order models, while consistently maintaining safe indoor air quality. These results highlight the practicality and scalability of our method in maintaining energy efficiency and indoor air quality in real-world buildings. 

Despite advances in learning-based methods, finding valid Lyapunov functions for nonlinear dynamical systems remains challenging. Current neural network approaches face two main issues: challenges in scalable verification and limited interpretability. To address these, we propose an end-to-end framework using transformers to construct analytical Lyapunov functions (local), which simplifies formal verification, enhances interpretability, and provides valuable insights for control engineers. Our framework consists of a transformer-based trainer that generates candidate Lyapunov functions and a falsifier that verifies candidate expressions and refines the model via risk-seeking policy gradient. Unlike Alfarano et al. (2024), which utilizes pre-training and seeks global Lyapunov functions for low-dimensional systems, our model is trained from scratch via reinforcement learning (RL) and succeeds in finding local Lyapunov functions for high-dimensional and non-polynomial systems. Given the symbolic nature of the Lyapunov function candidates, we employ efficient optimization methods for falsification during training and formal verification tools for the final verification. We demonstrate the efficiency of our approach on a range of nonlinear dynamical systems with up to ten dimensions and show that it can discover Lyapunov functions not previously identified in the control literature. Full implementation is available on https://github.com/JieFeng-cse/Analytical-Lyapunov-Function-Discovery. 

We consider the problem of designing learning-based reactive power controllers that perform voltage regulation in distribution grids while ensuring closed-loop system stability. In contrast to existing methods, where the provably stable controllers are restricted to be decentralized, we propose a unified design framework that enables the controllers to take advantage of an arbitrary communication infrastructure on top of the physical power network. This allows the controllers to incorporate information beyond their local bus, covering existing methods as a special case and leading to less conservative constraints on the controller design. We then provide a design procedure to construct input convex neural network (ICNN) based controllers that satisfy the identified stability constraints by design under arbitrary communication scenarios, and train these controllers using supervised learning. Simulation results on the University of California, San Diego (UCSD) microgrid testbed illustrate the effectiveness of the framework and highlight the role of communication in improving control performance. 

Ensuring the frequency stability of electric grids with increasing renewable resources is a key problem in power system operations. In recent years, a number of advanced controllers have been designed to optimize frequency control. These controllers, however, almost always assume that the net load in the system remains constant over a sufficiently long time. Given the intermittent and uncertain nature of renewable resources, it is becoming important to explicitly consider net load that is time-varying. This paper proposes an adaptive approach to frequency control in power systems with significant time-varying net load. We leverage the advances in short-term load forecasting, where the net load in the system can be accurately predicted using weather and other features. We integrate these predictions into the design of adaptive controllers, which can be seamlessly combined with most existing controllers including conventional droop control and emerging neural network-based controllers. We prove that the overall control architecture achieves frequency restoration decentralizedly. Case studies verify that the proposed method improves both transient and frequency-restoration performances compared to existing approaches.