Search

Reinforcement Learning (RL) is a promising technique for scheduling and planning storage systems in microgrids, which are small-scale power networks that can operate independently or in coordination with the main grid. RL can enhance the utilization of local renewable energy sources and reduce the operational costs of microgrids. In this comprehensive study and review the state-of-the-art applications of RL for Microgrid Energy Management (MEM), focusing on battery storage systems is discussed. This work also identify the main challenges, limitations, and future directions in this domain. Furthermore, this article present a novel benchmark algorithm that compares the performance of RL with mixed-integer linear programming (MILP), a widely used optimization technique. This complete work provides a valuable insight into the current status and future prospects of RL for MEM. [ABSTRACT FROM AUTHOR]

As the number of electric vehicles (EVs) significantly increases, the excessive charging demand of parked EVs in the charging station may incur an instability problem to the electricity network during peak hours. For the charging station to take a microgrid (MG) structure, an economical and energy-efficient power management scheme is required for the power provision of EVs while considering the local load demand of the MG. For these purposes, this study presents the power management scheme of interdependent MG and EV fleets aided by a novel EV charging/discharging scheduling algorithm. In this algorithm, the maximum amount of discharging power from parked EVs is determined based on the difference between local load demand and photovoltaic (PV) power production to alleviate imbalances occurred between them. For the power management of the MG with charging/discharging scheduling of parked EVs in the PV-based charging station, multi-objective optimization is performed to minimize the operating cost and grid dependency. In addition, the proposed scheme maximizes the utilization of EV charging/discharging while satisfying the charging requirements of parked EVs. Moreover, a more economical and energy-efficient PV-based charging station is established using the future trends of local load demand and PV power production predicted by a gated recurrent unit (GRU) network. With the proposed EV charging/discharging scheduling algorithm, the operating cost of PV-based charging station is decreased by 167.71% and 28.85% compared with the EV charging scheduling algorithm and the conventional EV charging/discharging scheduling algorithm, respectively. It is obvious that the economical and energy-efficient operation of PV-based charging station can be accomplished by applying the power management scheme with the proposed EV charging/discharging scheduling strategy. [ABSTRACT FROM AUTHOR]

To meet the diverse operational requirements of microgrids with multi-uncertainties, a tri-level two-stage mixed-integer robust optimization model is proposed to solve the optimal scheduling problem with source-load power and unit N-1 uncertainties. In the first stage, the optimal goal is to obtain robust scheduling plans that can be applied effectively in satisfying the operation and adjustment constraints, as well as to find the minimal operating cost in the basic scenario. Considering the system may concern a specific operating factor (e.g., load satisfaction rate) under a N-1 contingency, the second stage of the model minimizes the total energy of load shedding and renewable energy abandonment in the worst scenario. As the proposed robust model is a multi-level coupled mixed-integer optimization problem, which is difficult to address by either the duality method or the Karush–Kuhn–Tucker (KKT) method, a novel nested reformulation-and-decomposition (R&D) algorithm is proposed to handle this complex problem. Numerical simulations and tests have verified the effectiveness and superiority of the proposed robust model and the solution algorithm. Furthermore, the robust optimal schemes satisfy the engineering application more reasonably. [ABSTRACT FROM AUTHOR]

New Information and Communications Technologies (ICT), such as the Internet of Things (IoT), are enabling the evolution of energy grids towards a sophisticated power network called Smart Grid (SG). In the context of SGs, a microgrid is a self-sustained network that can operate in both grid-connected or stand-along modes. The long-term scheduling of the operation of distributed generators (DG) and renewable energy resources (RES) in microgrids is a problem that requires tough planning and the use of advanced tools to be efficiently addressed. This paper discusses different ICT technologies that can enable microgrid communication for control and management of distributed energy resources (DER). Based on such ICT, we propose a novel agent-based model to address the long-term scheduling of DER in microgrids as a distributed constraint optimization problem (DCOP). However, finding the optimal solution for a DCOP is known to be an NP-Hard problem, making difficult to guarantee optimal solutions even for short optimization periods. Hence, for the long-term scheduling of DER, we propose to split the problem into small time windows that can be effectively solved sequentially by off-the-shelf DCOP algorithms. A particular, but general enough case study is used to compare different DCOP algorithms under the proposed model. Results show that DCOP algorithms can find optimal and near-optimal solutions depending on the window size and the scenario considered. [ABSTRACT FROM AUTHOR]

Renewable energy-based multi-energy supply microgrids are expected to play an important role in smart cities. How to schedule such microgrids in grid-connected mode and dispatch power among sources inside the microgrids is a problem. Moreover, as electric vehicles are becoming more and more common, the charging of large numbers of vehicles is also a challenge for the utility grid. In this paper, we build a temporal-spatial electric vehicle charging demand model, which includes three parts: trip plans, duration of stay, and search for the shortest path based on the Dijkstra algorithm. Then, we build a grid-connected gas/electricity/heat microgrid and present a coordinated scheduling method for this microgrid. A day-ahead scheduling method is used to decide the role of the microgrid (i.e., operate as a load or as a generator from the point of view of the utility), a real-time rolling-horizon dispatching algorithm is used to respond to the forecasting errors and at the same time implement the real-time actual power exchange between the microgrid and the main grid. The problem is formulated as a mixed integer linear programming problem. The temporal-spatial electric vehicle charging demands model is simulated based on a 81-node transportation network, while the energy supply network is a combined IEEE-30, gas-20 and heat-14 network. The simulation results show the effectiveness of this coordinated scheduling method. [ABSTRACT FROM AUTHOR]

With increasing concerns about worldwide environmental conditions and rapid development of renewable energy technologies, microgrids have been regarded as a promising solution to reduce the burden of infrastructure-based power systems. However, due to the intrinsically intermittent features of existing renewable energy, along with random residential behavior patterns, unpredictable plugged-in or unplugged actions of electric vehicles (EVs) and the time-varying price of electricity, it is challenging for microgrid operators to efficiently perform load scheduling and energy management. In this paper, we propose an online algorithm to conduct cost-aware scheduling of EV loads and energy supplies for microgrids. We formulate this problem into a stochastic optimization problem with the objective of minimizing the time-average cost of a microgrid, including the purchase cost of electricity from the main grid, the cost of charging and discharging batteries, renewable harvesting costs, and life-cycle greenhouse-gas emission costs. To solve this problem, the key idea is to exploit the dynamics of the price of electricity to conduct battery charging and discharging operations, renewable energy harvesting, and schedule EV loads properly. Our method is based on the Lyapunov optimization technique, which has low computational complexity and only requires limited prediction of price information. The theoretical analysis of our algorithm confirms that the proposed strategy can achieve optimality with explicit bound. By conducting extensive real-data driven simulations, we demonstrate that our proposed algorithm can achieve much lower cost and be more eco-friendly than other alternative solutions. [ABSTRACT FROM PUBLISHER]

• The proposed study focuses on exploiting the potential of IoT technology in mission-critical energy management systems; • IoT-orchestration is enabled for nanogrid energy management architecture that focuses on minimizing the use of non-renewables and maximizing the use of renewables; • The orchestration module automatically generates the energy tasks and assigns them to the corresponding physical devices, which leads to enhanced efficiency and robustness in comparison with conventional energy management architectures; • The study also proposes a time-aware task scheduling algorithm wherein the task's surplus time is optimized to determine the best execution order; • Compared to conventional scheduling techniques, the proposed algorithm improved energy tasks' latency, response time, and round trip time (RTT). Internet of Things (IoT) has brought an immense revolution in diverse fields related to mission-critical systems including healthcare and navigation systems. We argue that the potential of IoT has not been fully exploited in the energy sector. There is a dire need to shift the traditional paradigm of mission-critical electric power architectures to IoT-enabled fully orchestrated architectures to enhance the overall performance. In this study, we present a novel IoT task orchestration architecture for efficient energy management of a nanogrid system that focuses on minimizing the use of non-renewable energy resources and maximizing the use of renewable energy resources. Since network orchestration deals with automating the interaction between multiple components involved to execute a particular service, therefore, scheduling the relevant processes within strict deadlines becomes the core pillar of the system's performance. The mission-critical systems with urgent task execution often suffer from missing task deadlines issues. To overcome this issue, we present a task scheduling algorithm that incorporates the optimized surplus time, and efficiently executes the energy management-related tasks contemplating to their types. The study utilizes sensors to obtain data from physical IoT devices, including photovoltaic (PV), Energy Storage System (ESS), and diesel generator (DG). The performance of the proposed model is evaluated using data set of nanogrid houses. The outcomes revealed that IoT-task orchestration has played a pivotal role in efficient energy management for nanogrid missioncritical systems. Furthermore, the comparison showed that the task starvation rate is reduced to 16% and 12% when compared with state-of-the-art scheduling algorithms. [ABSTRACT FROM AUTHOR]

Using the example of a two-layer energy management system for remote microgrids, a novel scheduling algorithm is proposed that considers the battery lifetime and is expected to reduce the operational cost of the microgrid. This method combines the objectives of minimizing fuel consumption and battery degradation costs into a single weighted goal. [ABSTRACT FROM AUTHOR]

Microgrids are expected to become part of the next electric power system evolution, not only in rural and remote areas but also in urban communities. Since microgrids are expected to coexist with traditional power grids (such as district heating does with traditional heating systems), their planning process must be addressed to economic feasibility, as a long-term stability guarantee. Planning a microgrid is a complex process due to existing alternatives, goals, constraints and uncertainties. Usually planning goals conflict each other and, as a consequence, different optimization problems appear along the planning process. In this context, technical literature about optimization techniques applied to microgrid planning have been reviewed and the guidelines for innovative planning methodologies focused on economic feasibility can be defined. Finally, some trending techniques and new microgrid planning approaches are pointed out. [ABSTRACT FROM AUTHOR]

In developing countries such as in Africa, rural areas are not always connected to the national or regional electric grid. As there is a huge potential for solar energy production, photovoltaics are established to supply electricity for operating critical infrastructures with battery systems and diesel generators installed for redundancy. These systems have the potential for further utilization. As a first step, overproduction of solar energy is offered to households due to simple control rules observing battery levels. This paper proposes an optimization method for energy allocation by process graphs or P-graphs. The method provides preliminary estimations on optimal periodical energy balances between energy producers, storage, and consumers of different priorities. These preliminary estimations serve as reconciliation parameters for advanced real-time control of energy distribution. Each step of transforming the practical problem of multiperiod energy distribution to the language of process network synthesis and then optionally to mathematical programming is detailed. To represent the problem adequately, the original formulation of process network synthesis is extended to take into account potential targets and intermediate entities that cannot be accumulated. The work aims to provide better utilization of renewable energy sources and storage. The microgrid profit and return of investment can also be improved by the proposed methodology. Image 1 [ABSTRACT FROM AUTHOR]