This paper proposes a new deep learning (DL) based model-free robust method for bulk system on-line load restoration with high penetration of wind power. Inspired by the iterative calculation of the two-stage robust load restoration model, the deep neural network (DNN) and deep convolutional neural network (CNN) are respectively designed to find the worst-case system condition of a load pickup decision and evaluate the corresponding security. In order to find the optimal result within a limited number of checks, a load pickup checklist generation (LPCG) algorithm is developed to ensure the optimality. Then, the fast robust load restoration strategy acquisition is achieved based on the designed one-line strategy generation (OSG) algorithm. The proposed method finds the optimal result in a model-free way, holds the robustness to handle uncertainties, and provides real-time computation. It can completely replace conventional robust optimization and supports on-line robust load restoration which better satisfies the changeable restoration process. The effectiveness of the proposed method is validated using the IEEE 30-bus system and the IEEE 118-bus system, showing high computational efficiency and considerable accuracy.
This paper proposes a stochastic optimal preparation and resource allocation method for upcoming extreme weather events in distribution systems, which can assist utilities to achieve faster and more efficient post-event restoration. With the objective of maximizing served load and minimizing operation cost, this paper develops a two-stage stochastic mixed-integer linear programming (SMILP) model. The first-stage determines the optimal positions and numbers of mobile resources, fuel resources, and labor resources. The second-stage considers network operational constraints and repair crew schedu…
In case of high impact low probability events, in order to restore the critical loads of the distribution network as much as possible, it is necessary to employ all available resources such as microgrids and distributed generations. This paper presents a two-stage method for critical load restoration after an extreme event utilizing the coordination of all available distributed generations, microgrids, and demand response programs. In the first stage, the post-disaster reconfiguration of the network is determined, and in the second stage, the demand response outputs of the responsive loads and…
A major outage in the electricity distribution system may affect the operation of water and natural gas supply systems, leading to an interruption of multiple services to critical customers. Therefore, enhancing resilience of critical infrastructures requires joint efforts of multiple sectors. In this paper, a distribution system service restoration method considering the electricity-water-gas interdependency is proposed. The objective is to provide electricity, water, and natural gas supplies to critical customers in the desired ratio according to their needs after an extreme event. The opera…
We analyze the dynamics of the power grid with a high penetration of renewable energy sources using the ORNL-PSERC-Alaska (OPA) model. In particular we consider the power grid of the Balearic Islands with a high share of solar photovoltaic power as a case study. Day-to-day fluctuations of the solar generation and the use of storage are included in the model. Resilience is analyzed through the blackout distribution and performance is measured as the average fraction of the demand covered by solar power generation. We find that with the present consumption patterns and moderate storage, solar ge…
In the literature, smart grids are modeled as cyber-physical power systems without considering the computational social aspects. However, end-users are playing a key role in their operation and response to disturbances via demand response and distributed energy resources. Therefore, due to the critical role of active and passive end-users and the intermittency of renewable energy, smart grids must be planned and operated by considering the computational social aspects in addition to the technical aspects. The level of cooperation, flexibility, and other social features of the various stakehold…
Extreme weather events have a significant impact on the aging and outdated power distribution infrastructures. These high-impact low-probability (HILP) events often result in extended outages and loss of critical services, thus, severely affecting customers' safety. This calls for the need to ensure resilience in distribution networks by quickly restoring the critical services during a disaster. This paper presents an advanced feeder restoration method to restore critical loads using distributed energy resources (DERs). A resilient restoration approach is proposed that jointly maximizes the am…
With the growth of complexity and extent, large scale interconnected network systems, e.g. transportation networks or infrastructure networks, become more vulnerable towards external disruptions. Hence, managing potential disruptive events during the design, operating, and recovery phase of an engineered system therefore improving the system's resilience is an important yet challenging task. In order to ensure system resilience after the occurrence of failure events, this study proposes a mixed-integer linear programming (MILP) based restoration framework using heterogeneous dispatchable agent…
A Resilience Evaluation Framework for Electric Distribution Systems: Historical Weather Conditioning, Sensitivity Analysis, and a Flooding-Aware Extension
To improve the resilience of electric distribution systems, this paper proposes a stochastic multi-period mixed-integer linear programming model that determines where to underground distribution lines and how to coordinate mobile generators in order to serve critical loads during extreme events. The proposed model represents the service restoration process using the linearized DistFlow approximation of the AC power flow equations as well as binary variables for the undergrounding statuses of the lines, the configurations of switches, and the locations of mobile generators during each time peri…
Quick and reliable power system restoration is critically important after natural disasters or other sudden threats, such as cyber-attacks. Leveraging renewable resources in system restoration shortens recovery times, resulting in prevented life-loss and avoided economic-loss, and improves the resilience of the entire grid. However, it is not a common practice today; the inherent variability of these resources represents a challenge for a streamlined restoration process. This paper presents a prioritized method - starting with renewable generator units then lowering priority to conventional un…
Interdisciplinary Graduate School, Nanyang Technological University Singapore（南洋理工大学跨学科研究生院（新加坡））；School of Electrical and Electronic Engineering, Nanyang Technological University Singapore（南洋理工大学电气与电子工程学院（新加坡））
Mobile energy storage systems (MESSs) provide mobility and flexibility to enhance distribution system resilience. The paper proposes a Markov decision process (MDP) formulation for an integrated service restoration strategy that coordinates the scheduling of MESSs and resource dispatching of microgrids. The uncertainties in load consumption are taken into account. The deep reinforcement learning (DRL) algorithm is utilized to solve the MDP for optimal scheduling. Specifically, the twin delayed deep deterministic policy gradient (TD3) is applied to train the deep Q-network and policy network, t…
The ever-increasing penetration of Distributed Generators (DGs) in distribution networks suggests to enable their potentials in better fulfilling the restoration objective. The objective of the restoration problem is to resupply the maximum energy of loads considering their priorities using minimum switching operations. Basically, it is desired to provide a unique configuration that is valid regarding the load and generation profiles along the entire restorative period. However, this unique configuration may not satisfy at the same time: I) the DG start-up requirements at the beginning of the …
Community-Centered Resilience Enhancement of Urban Power and Gas Networks via Microgrid Partitioning, Mobile Energy Storage, and Data-Driven Risk Assessment
In order to increase the resilience of distribution systems against high-impact low-probability (HILP) events, it is important to prioritize the damaged assets so that the lost loads, especially critical and important loads, can be restored faster. In addition, correctly predicting the number of repair teams during critical times contributes to restoring the network to the initial resilience level. For this reason, this paper discusses the prioritization of electricity supply lines for evaluating the number of required repair teams. To this end, the economic value of distribution system lines …
Power restoration is an urgent task after a black-out, and recovery efficiency is critical when quantifying system resilience. Multiple elements should be considered to restore the power system quickly and safely. This paper proposes a recovery model to solve a direct-current optimal power flow (DCOPF) based on mixed-integer linear programming (MILP). Since most of the generators cannot start independently, the interaction between black-start (BS) and non-black-start (NBS) generators must be modeled appropriately. The energization status of the NBS is coordinated with the recovery status of tr…
