skip to main content

Title: SAVER: Safe Learning-Based Controller for Real-Time Voltage Regulation
Fast and safe voltage regulation algorithms can serve as fundamental schemes for achieving a high level of renewable penetration in modern distribution power grids. Faced with uncertain or even unknown distribution grid models and fast changing power injections, model-free deep reinforcement learning (DRL) algorithms have been proposed to find the reactive power injections for inverters while optimizing the voltage profiles. However, such data-driven controllers can not guarantee the satisfaction of the hard operational constraints, such as maintaining voltage profiles within a certain range of the nominal value. To this end, we propose SAVER: SAfe Voltage Regulator, which is composed of an RL learner and a specifically designed, computationally efficient safety projection layer. SAVER provides a plug-and-play interface for a set of DRL algorithms that guarantees the system voltages are within safe bounds. Numerical simulations on real-world data validate the performance of the proposed algorithm.  more » « less
Award ID(s):
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
2022 IEEE Power & Energy Society General Meeting (PESGM)
Page Range / eLocation ID:
1 to 5
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Fast mechanical disconnect switches are an integral part of hybrid circuit breakers, which are proposed as protection devices to clear faults in medium voltage distribution systems. Compared to their conventional counterparts, hybrid circuit breakers can have the ability to limit the fault current, which could allow more interconnections between substations with advantages with respect to grid reliability and resiliency. Furthermore, they enable the integration of distributed generation such as small solar power installations without expensive changes to the grid infrastructure. The proposed design of an ultrafast mechanical disconnect switch operates in vacuum, carries continuous current similar to conventional vacuum interrupters, opens at current zero, features minimum moving mass, and has an open contact separation of less than a millimeter. The limited separation distance requires an optimized contact geometry to keep the electric field within safe limits, minimize the moving mass, and reduce contact resistance. This paper proposes different contact geometries and uses finite element analysis to compare the contact geometries with respect to maximum electric field, mass, and contact resistance. Reductions of mass by 50% and reduction of contact resistance of 10% have been achieved. 
    more » « less
  2. Global sensitivity analysis (GSA) of distribution system with respect to stochastic PV variations plays an important role in designing optimal voltage control schemes. This paper proposes a Kriging, i.e., Gaussian process modeling enabled data-driven GSA method. The key idea is to develop a surrogate model that captures the hidden global relationship between voltage and real and reactive power injections from the historical data. With the surrogate model, the Sobol index can be conveniently calculated to assess the global sensitivity of voltage to various power injection variations. Comparison results with other model-based GSA methods on the IEEE 37-bus feeder, such as the polynomial chaos expansion and the Monte Carlo approaches demonstrate that the proposed method can achieve accurate GSA outcomes while maintaining high computational efficiency. 
    more » « less
  3. For fast timescales or long prediction horizons, the AC optimal power flow (OPF) problem becomes a computational challenge for large-scale, realistic AC networks. To overcome this challenge, this paper presents a novel network reduction methodology that leverages an efficient mixed-integer linear programming (MILP) formulation of a Kron-based reduction that is optimal in the sense that it balances the degree of the reduction with resulting modeling errors in the reduced network. The method takes as inputs the full AC network and a pre-computed library of AC load flow data and uses the graph Laplacian to constraint nodal reductions to only be feasible for neighbors of non-reduced nodes. This results in a highly effective MILP formulation which is embedded within an iterative scheme to successively improve the Kron-based network reduction until convergence. The resulting optimal network reduction is, thus, grounded in the physics of the full network. The accuracy of the network reduction methodology is then explored for a 100+ node medium-voltage radial distribution feeder example across a wide range of operating conditions. It is finally shown that a network reduction of 25-85% can be achieved within seconds and with worst-case voltage magnitude deviation errors within any super node cluster of less than 0.01pu. These results illustrate that the proposed optimization-based approach to Kron reduction of networks is viable for larger networks and suitable for use within various power system applications. 
    more » « less
  4. This paper considers achieving flat voltage profiles in a distribution network based on reactive power optimization (RPO) through voltage regulation devices (VRD). These devices include capacitor banks, load-tap-changing and regulating transformers, whose statuses can only assume pre-determined integer value levels, making this a non-convex problem. Two RPO-based algorithms are proposed, which can be applied to any initial states, node priority, topology and load model types. The first algorithm focuses on finding a practical solution by ensuring the VRD constraints are observed at each step. The second one focuses on finding the globally optimal solution by applying a convex relaxation technique and solving the resulting problem with the barrier interior point method. Here, the gradients are computed numerically, thus requiring no analytical functions of voltages in terms of VRDs. Numerical results and their analysis are examined on two test networks: 1) single feeder; and 2) network with laterals. 
    more » « less
  5. Hybrid electric vehicles can achieve better fuel economy than conventional vehicles by utilizing multiple power sources. While these power sources have been controlled by rule-based or optimization-based control algorithms, recent studies have shown that machine learning-based control algorithms such as online Deep Reinforcement Learning (DRL) can effectively control the power sources as well. However, the optimization and training processes for the online DRL-based powertrain control strategy can be very time and resource intensive. In this paper, a new offline–online hybrid DRL strategy is presented where offline vehicle data are exploited to build an initial model and an online learning algorithm explores a new control policy to further improve the fuel economy. In this manner, it is expected that the agent can learn an environment consisting of the vehicle dynamics in a given driving condition more quickly compared to the online algorithms, which learn the optimal control policy by interacting with the vehicle model from zero initial knowledge. By incorporating a priori offline knowledge, the simulation results show that the proposed approach not only accelerates the learning process and makes the learning process more stable, but also leads to a better fuel economy compared to online only learning algorithms. 
    more » « less