Grid Energy Storage Systems Research Articles

Medium-Duty Delivery Truck Integrated Bidirectional Wireless Power Transfer System With Grid and Stationary Energy Storage System Connectivity

Medium-Duty Delivery Truck Integrated Bidirectional Wireless Power Transfer System With Grid and Stationary Energy Storage System Connectivity

High Throughput Electrochemical Screening of Phosphate-Rich Nonflammable Electrolytes in Lithium-Ion Batteries

Frequent fires and explosions in lithium-ion batteries (LIBs) used in grid energy storage systems (ESS) highlight the necessity of revisiting nonflammable phosphate electrolytes as alternatives to the currently used flammable carbonates. However, previous studies have shown the difficulty of integrating phosphate solvents into LIB electrolytes due to compatibility issues with graphite. In this work, we developed a high-throughput (HTP) electrochemical characterization method, akin to pH test paper, to rapidly screen potential phosphate electrolytes and graphite materials. Through HTP screening, we identified 101 promising combinations out of 1,740. This number was reduced to 26 after testing in Li/Graphite half cells. The optimized phosphate-rich electrolyte (60 v% phosphate) with cosolvents demonstrated 300 stable cycles at 0.1 C in Graphite/LiFePO4 (LFP) full cells with thick electrodes (∼3.0 mAh cm−2), surpassing prior research findings. This unique HTP method provides a powerful tool to expedite the development of safe LIBs for ESS applications.

An efficient hybrid technique for energy management system with renewable energy system and energy storage system in smart grid

This manuscript proposes a hybrid technique for an energy management system (EMS) with a renewable energy system and energy storage system using the smart micro grid. This work's novelty liesin the integration of Flying Foxes Optimization (FFO) algorithm and Deep Attention Dilated Residual Convolutional Neural Network (DADRCNN). Unlike existing methods, the FFO-DADRCNN method effectively leverages both optimization and predictive modelling capabilities to enhance grid stability, efficiency, and load prediction accuracy. The main goals of the proposed approach are to increase grid stability and efficiency, decrease micro grid system costs, and increase load prediction accuracy. The micro grid system incorporates PV, an energy storage system, wind turbine and load demand. To regulate the micro grid system and estimate load utilizing the DADRCNN most effectively, the proposed FFO technique is used. The proposed method's performance is evaluated using the MATLAB platform and compared with existing approaches, including Giza Pyramids Construction, Heap-based optimizer, and Nomadic People Optimizer. The proposed technique shows better outcomes when compared to other methods. Compared to other methods currently available, our proposed approach boasts an impressive 98 % efficiency rate and comes at a low operating cost of $1840.

Comprehensive testing technology for new energy vehicle power batteries based on improved particle swarm optimization

As the new energy industry continues to progress, the health management of power batteries has become the key to ensuring the performance and safety of automobiles. Therefore, accurately predicting battery capacity decline is particularly important. A battery capacity degradation prediction model combining unscented particle filtering, particle swarm optimization, and SVR is constructed. It optimizes regression parameters through the introduced optimization strategy. Unscented particle filtering is used to improve particle swarm optimization and battery detection model. The study tested four various models of lithium-ion batteries. The model predicted a mean square error of 0.0011 for battery 5, 0.0007 for battery 6, 0.0022 for battery 7, and 0.0013 for battery 18. In the prediction of different battery types, the mean square error of the NIMH battery was reduced by 0.0008 compared with the particle swarm optimization-support vector regression algorithm, and by 0.0005 compared with the unscented particle filtering-regression vector regression algorithm. The mean square error of lithium-iron phosphate battery was reduced by 0.0008 and 0.0004 respectively compared with comparison models. The mean square error value of lithium titanate battery was reduced by 0.0007 and 0.0003 respectively in the research model compared with comparison models. It improves the prediction accuracy in lithium-ion batteries. Its application in battery health management can provide important technical support for improving battery performance and extending service cycles. The proposed method can be used for battery monitoring and management of power grid energy storage system. By accurately predicting the capacity decline of battery, the operation strategy of energy storage system can be optimized to ensure the efficient operation and long life of the system. The battery management system can be used for drones and aviation equipment to predict battery health and capacity decline in real time, ensuring the safety and reliability of flight missions.

Stable dendrite-free Zn electrodeposition triggered by hydrophobic-zincophilic and elastic interface for highly reversible Zn battery anode

Aqueous zinc (Zn) metal batteries (ZMBs) are regarded as promising candidates for grid energy storage systems due to their low cost, high safety, and environmental friendliness. However, the notorious Zn dendrites induced by the uneven Zn deposition and severe side reactions hinder the widespread application of aqueous ZMBs. Herein, we ingeniously combine inorganic zincophilicity and organic elasticity to report a silver nanoparticle-decorated polyvinylidene fluoride (PVDF) as a hydrophobic-zincophilic and elastic artificial interface layer (HZE-Zn) to stabilize Zn electrodeposition. The synergistic effect of the elastic hydrophobic PVDF and the in-situ alloyed zincophilic interface constrains the corrosion and achieves homogeneous Zn-ion flux at the electrode interface. Consequently, the modified Zn anode achieves stable Zn deposition with a long lifespan of over 2200 h at a current density of 1.0 mA cm−2 and 1400 h at 2.0 mAh cm−2. When coupled with a MnO2 cathode, the prepared HZE-Zn anode exhibits better full-cell cycling stability than the pristine Zn anode. This interesting concept of building a hydrophobic-zincophilic and elastic artificial interface layer for Zn anodes provides new vitality to develop the practical applications of aqueous ZMBs.

A multifunctional desalination-osmotic energy storage (DOES) system for managing energy and water supply

Net-zero carbon emission target for mitigating climate change accelerates the exploitation of renewable energy, such as solar and wind, as power origin in utilities sector. However, the intermittency of renewable energy escalates the supply-demand mismatch in not only electricity sector but also water sector, as freshwater supply increasingly relies on unconventional, energy-intensive desalination that is increasingly powered by renewables. To enhance the energy-water resilience, we propose a desalination-osmotic energy storage (DOES) system, which alternates the operation of reverse osmosis (RO) for desalination and pressure retarded osmosis (PRO) for electricity generation, achieving multiple functions including freshwater production and storage, grid energy storage, and eventually bulk-scale management of freshwater and energy supply. Via innovative system design and operation integrating semi-closed (SC) and closed-circuit (CC) configurations in RO and PRO modes in the proposed DOES, energy loss arising from over-pressurization in RO and under-pressurization in PRO could be substantially reduced. As a result, DOES can achieve practical maximum energy efficiencies of >75 % for both RO desalination and PRO electricity generation, and a round-trip efficiency of >68 % when used as a grid-scale energy storage system (ESS). Apart from the quantitative analysis of energy performance, a qualitative comparison of various other performance metrics between DOES and other grid ESSs is also conducted. Given its various performance advantages as well as multi-functionality, the DOES system could be an important complement to, though not replacement of, existing grid-scale ESSs.

Electron-outflowing heterostructure hosts for high-voltage aqueous zinc-iodine batteries

Enhancing the energy density is an imperative challenge in the advancement of aqueous zinc-iodine (Zn-I2) batteries, which hold great promise for grid energy storage systems. Achieving the reversible conversion reaction of high-valence iodine species, particularly I+/I0 redox chemistry, offers a substantial potential for improved energy density. In this study, we introduce a novel method of employing electron-outflowing heterostructured hosts to stabilize I+ ions in aqueous halide-free ZnSO4 electrolytes. Both experimental analyses and density functional theory (DFT) calculations reveal that the stabilization of thermodynamically unstable I+ cations originates from the charge transfer interaction between I+ ions and copper phthalocyanine (CuPc) within the hosts. The electron-outflowing effect resulting from the π-π conjugation between CuPc and the carbon framework in heterostructured hosts further intensifies the charge transfer interaction, leading to the reversible I+/I5− conversion reaction in Zn-I2 batteries. Consequently, this emerging redox reaction, validated by in-situ Raman and ex-situ ultraviolet-visible spectroscopy, introduces another voltage plateau at ∼1.6 V and enhances the energy density to 233.8 Wh kg–1 at 3.0 A g–1. This work showcases outstanding electrochemical performance for the Zn-I2 battery, providing valuable insights into advancing iodine redox chemistry featuring multiple electrons.

Accelerated Selective Li+ Transports Assisted by Microcrack-Free Anionic Network Polymer Membranes for Long Cyclable Lithium Metal Batteries.

Rechargeable Li metal batteries have the potential to meet the demands of high-energy density batteries for electric vehicles and grid-energy storage system applications. Achieving this goal, however, requires resolving not only safety concerns and a shortened battery cycle life arising from a combination of undesirable lithium dendrite and solid-electrolyte interphase formations. Here, a series of microcrack-free anionic network polymer membranes formed by a facile one-step click reaction are reported, displaying a high cation conductivity of 3.1 × 10-5 S cm-1 at high temperature, a wide electrochemical stability window up to 5V, a remarkable resistance to dendrite growth, and outstanding non-flammability. These enhanced properties are attributed to the presence of tethered borate anions in microcrack-free membranes, which benefits the acceleration of selective Li+ cations transport as well as suppression of dendrite growth. Ultimately, the microcrack-free anionic network polymer membranes render Li metal batteries a safe and long-cyclable energy storage device at high temperatures with a capacity retention of 92.7% and an average coulombic efficiency of 99.867% at 450 cycles.

Fractional Order Modeling of Lithium-ion Batteries For A Real Smart Grid System

Establishing robust, accurate, and low-complexity battery models is crucial for Smart Grid applications to effectively utilize and manage lithium-ion batteries and estimate battery states. This paper analyzes lithium-ion battery electrochemical impedance spectrograms and integrates impedance elements with fractional-order characteristics into Thevenin circuit models based on fractional calculus theory. These models can be embedded in microprocessors for accurate real-time results. A time domain identification procedure minimizes Integral Absolute Error, yielding simple, high-accuracy fractional order modeling of Energy Storage Systems in Real Smart Grids, with a very acceptable average Nrmse index of fit.

Enhancing Grid Stability and Sustainability: Integrating Lithium-Ion Batteries with Solar Energy for Large-Scale Applications

Solar energy serves as a pollution-free, environmentally friendly energy source and a way to achieve sustainable energy. Nonetheless, the variability of solar energy caused by daily and seasonal changes in sunlight availability may cause an imbalance between energy supply and demand. This passage discussed the limitations of solar energy and problems with large-scale grids combined with solar energy and lithium-ion batteries, and their corresponding solutions, mentioning cost issues and the key point that may need extra attention to achieve an ideal status for a grid energy storage system. To address this issue, this paper presents a detailed analysis and discussion on Lithium-ion batteries and the synergy between lithium-ion batteries and solar power. In order to apply this combination in large-scale grids, high energy requirements can be achieved by changing the anode material from graphite to silicon, but this will inevitably lead to an increase in cost. This paper also discusses high-efficiency power networks and compares them to lithium-metal batteries (e.g., LOBs and LSBs). Solving this problem could work towards reducing CO2 emissions and provide energy to achieve a balance between energy supply and demand, and could also be used to provide emergency power.

Introducing graphene‐enhanced technologies to the regulated market to mitigate carbon emissions

AbstractAmbitious net zero carbon emissions goals have been set by various climate change agencies and government officials, including President Biden. One of the most prominent emissions goals aims to reach net zero carbon emissions globally by 2050. Unfortunately, the United States is not on track to achieving this goal and the United Nations has announced that several nations have fallen short of their carbon emissions agreements. Increases in global temperatures, deadly natural disasters, and species extinction have heightened the pressure facing scientists and policymakers to convince nations to stay on course to reach net zero emissions within the next few decades. Last year, the largest contributor to carbon dioxide emissions in the United States was “U.S. motor gasoline and diesel (distillate) fuel consumption” (U.S. Energy Information Administration, Frequently Asked Questions), making it clear that changes must be made in the energy sector to mitigate this. Recent innovations in the battery industry have shown exciting promise for the future of clean energy, though regulatory barriers must be overcome to fully establish these new technologies. One such technology is the graphene‐enhanced lithium‐ion battery which has five times the energy density of regular lithium‐ion batteries due to the high thermal conductivity of graphene. However, graphene is difficult to produce on a large scale, and many investors and stakeholders have hesitated to participate in the graphene supply chain due to the lack of regulation and health and safety information on graphene. The EPA (Environmental Protection Agency), along with other standardization organizations, has worked to understand nanomaterials like graphene over the past decade, but they seem to be behind in producing substantial data on the material's properties. Furthermore, current graphene manufacturers have varying ways of classifying and labeling graphene products, creating additional confusion in the marketplace. To accelerate the process of researching and regulating graphene, the EPA's adherence to predetermined deadlines for the creation of regulations for up‐and‐coming materials must be mandated, a standardized graphene labeling system should be adopted, universities and federal agencies should be incentivized to collaborate on research related to graphene production methods, and a mandatory “battery passport” system should be implemented to ensure the widespread availability of battery information. These recommendations are aimed at increasing trust and collaboration within the graphene industry and standardizing graphene materials to reduce ambiguity within the supply chain. By implementing these policy changes, the use of graphene‐enhanced lithium‐ion batteries in electric vehicles and grid energy storage systems can become more widespread, and allow for higher efficiency. As improvements are made to clean energy alternatives, the use of fossil fuels can be reduced, thus, decreasing overall carbon emissions.

Analysis of energy storage capacity optimization of distribution network based on quadratic planning algorithm under large scale DG access

Abstract The intermittent, random and uncontrollable characteristics of renewable energy generation bring many challenges to the actual operation of distribution grid (DG) energy storage. This paper firstly establishes a whole life cycle model applicable to the distribution grid energy storage system, which fully considers the various types of costs of VSC converter stations. Secondly, for comprehensive reactive power optimization, a sequential quadratic programming algorithm is used, and inequality constraints are added to the objective function in the form of penalty functions to optimize the state of on-load transformers and capacitors. Finally, the effectiveness of its distribution grid energy storage configuration optimization is verified by applying the distribution grid energy storage optimization based on the quadratic programming algorithm to electric vehicle charging piles. The results show that after two stages of optimization, the configuration capacity of distribution grid energy storage is increased by 53.14Kw/h compared with that before optimization, which is a substantial increase. The average annual cost of battery 0’ access decreases by 0.275, and battery 2’ access decreases by 0.581. This shows that the energy storage capacity optimization strategy based on the sequential quadratic programming algorithm proposed in this paper can improve the efficiency of energy storage utilization. It provides the basis for the user to select the energy storage capacity.

Investigation of Low Order Parameters Affecting Tidal Stream Energy Resource Assessments

Tidal stream energy may have the potential to contribute to the baseline energy needs of the UK. There is a significant difference between the existing assessments of the UK’s resource, highlighting the need for a standardised model. The paper investigates some challenges of quantifying tidal stream energy. This study examines the sensitivity of resource assessments to low-order parameters, in order to inform higher order modelling and contribute to defining a standardised model. A combination of channel characteristics, turbine parameters and simulation settings formed the basis of 96+ cases to determine key parameters affecting resource assessments. Two 0-dimensional, idealised channels were modelled using 2-8 constituents from four UK sites spanning the semi-diurnal form factor range. Blockage-corrected blade element momentum theory was used to represent variable-speed, variable-pitch turbines. Three blockage designed rotors and one rotor designed for an unspecified blockage were simulated. Power capping and combined power and thrust capping were implemented.
 
 The study indicates dynamic balance is a key parameter affecting tidal dynamics when turbines are present in a channel. Variability of annual energy production due to the nodal cycle was most significant in the drag dominated channel with increasing blockage. Constituents with large amplitude and phase differences led to greater variability of the resource across the nodal cycle. Results indicate that the maximum deviation from the average annual energy production over the 18.6-year nodal cycle is +/- 8.8%. At a low form factor site, the difference between using 2 or 4 constituents was insignificant. However, for a higher form factor site, that is still semi-diurnal, the peak velocity increases by 0.32 m/s with 8 constituents compared to 2. Random Forest modelling indicates that the most important characteristic point on a variable-speed, variable-pitch turbine performance curve is the rated speed. Comparison of capping strategies highlights that combined power and thrust capping leads to an increase in energy per swept area. Combined capping decreased the variability of the resource over the nodal cycle, which is helpful to maintain a steady supply to the grid and sizing energy storage systems.

Applications of energy storage systems in power grids with and without renewable energy integration — A comprehensive review

Energy storage system (ESS) is recognized as a fundamental technology for the power system to store electrical energy in several states and convert back the stored energy into electricity when required. Some excellent characteristics such as availability, versatility, flexible performance, fleet response time, modularity etc., make ESS more attractive for power system applications. The ESS has signification contributions and applications to operate the power system optimally in power grids with and without integrating renewable energy (RE) systems. This paper presents a comprehensive review of ESS technologies and their applications in power grids. Five different types of ESS, namely mechanical, chemical, electrical, electro-chemical and thermal, are elaborately explored with their key characteristics and applications. A summary of comparative analysis to find the appropriate ESS for power system applications and an analysis of the practical implementation of different ESS worldwide have been presented briefly, reflecting the suitability of ESS for power system applications. The comparative analysis further demonstrates the nature of various ESS, including their advantages and limitations in specific applications. The major challenges from these analyses are identified to be the capacity of ESS, RE integration, high maintenance and implementation cost of ESS, ESS degradation, waste management and associated environmental pollution. Moreover, the review finds out that there are still numerous challenges in terms of technology and economy that need to be addressed to make the practical applications of ESS viable for power industries. The reviews also identify recent key researches that can effectively address those issues and explore the future directions researchers can undertake to improve various characteristics of ESS. Finally, the transition pathway to promote ESS for power systems applications with and without integration of RE systems is discussed to demonstrate the possibility and potentiality of large-scale ESS deployment for a sustainable climate-friendly future.

Microwave-Assisted Sol-Gel Synthesis of High-Voltage Ruthenium-Doped LiFePO4 Cathodes for Solid-State Lithium-Ion Batteries

The demand for high voltage and high-capacity energy systems has grown due to increased use of portable electronics, electric vehicles, and grid energy storage systems. LiFePO4 (LFP) shows great promise as a cathode material due to its low cost, structural stability, safety characteristics, and low environmental impact. However, LFP cathodes possess a low Li-ion diffusion coefficient and poor electrical conductivity compared to oxide-based cathodes that are used currently. To make LFP competitive with oxide-based cathodes, LFP was modified to improve its characteristics. Doping of materials is a highly effective method for improving electrochemical characteristics in LFP cathodes. Ruthenium-doping of iron sites within LFP has shown positive effects in the promotion of Li-ion diffusion, reduction in band gap energies, and improved electrical conductivity. Previous studies have shown that ruthenium-doping at 0.01 moles has an increased specific capacity of 162 mAh g-1, better cycling characteristics, and reduced resistance compared to pristine LFP.1 Ruthenium doping levels above 0.02 moles have shown a decrease in specific capacity, cycling characteristics, and an increased resistance due to the formation of ruthenium oxide that disrupts the crystal lattice of LFP. In this study, LiFe1-xRuxPO4 (x= 0.01, 0.0075, and 0.005) and pristine LFP were prepared via microwave-assisted sol-gel synthesis to explore the effects of lower levels of ruthenium-doping on LFP electrochemical characteristics. Microwave synthesis was employed for reduced sintering time, beneficial effects on microstructure, and decreased cost. Samples were characterized via scanning electron microscopy, in-situ/ex-situ X-ray diffraction, energy dispersive X-ray spectroscopy, electrochemical impedance spectroscopy, and cyclic voltammetry.The authors acknowledge financial support from the NSF IUCRC program for the “Center for solid-state electric power storage” (#2052631) and support from the South Dakota Board of Regents for the “Governor’s Research Center (GRC) for electrochemical energy storage”.1.) Gao, Y.; Xiong, K.; Zhang, H.; Zhu, B. Effect of RU Doping on the Properties of Lifepo4/c Cathode Materials for Lithium-Ion Batteries. ACS Omega 2021, 6 (22), 14122–14129.

Research on key technologies of large-scale wind-solar hybrid grid energy storage capacity big data configuration optimization

Due to the uncertainty and randomness of large-scale wind and light, the output power of the power grid has great fluctuations. If it is directly connected to the grid, it will affect the main grid. In addition, when the grid switches between on-grid/off-grid operation modes, there will be power shortages, shocks and oscillations. The scientific and reasonable configuration of energy storage system capacity big data can reduce the load power shortage rate, improve the utilization rate of renewable energy, and ensure the reliable operation of the power grid. For this reason, the key technology of large-scale wind-solar hybrid grid energy storage capacity big data configuration optimization is studied. A large-scale wind-solar hybrid grid energy storage structure is proposed, and the working characteristics of photovoltaic power generation and wind power generation are analyzed, and the probability model of photovoltaic power generation, wind power generation and load, as well as the charging and discharging model of battery and super capacitor are established accordingly. On this basis, the optimization objective function is set, the constraints are determined, and the large-scale wind-solar hybrid grid energy storage capacity big data configuration optimization model is constructed. And the PSO algorithm is used to solve the model to realize the big data configuration optimization of large-scale wind-solar hybrid grid energy storage capacity. The research results show that the proposed method of large-scale wind-solar hybrid grid energy storage system has good power supply reliability and economy, and can effectively improve the utilization rate of renewable energy.

Simulation of Secondary Frequency Modulation Process of Wind Power with Auxiliary of Flywheel Energy Storage

With the rapid increase in the proportion of wind power, the frequency stability problem of power system is becoming increasingly serious. Based on MATLAB/Simulink simulation, the role and effect of secondary frequency modulation assisted by Flywheel Energy Storage System (FESS) in regional power grid with certain wind power penetration rates are studied. First, the linear frequency control of the power system is used to establish the primary frequency modulation control model of FESS assisting wind power, and the frequency characteristics of FESS participating in primary frequency modulation are analyzed according to the transfer function. Then, in the case of step disturbance and continuous disturbance of load power, the frequency characteristics of a regional power grid are simulated and demonstrated through time domain simulation, and conclusions are drawn through comparison; a certain proportion of FESS can quickly respond to the frequency deviation signal. During secondary frequency modulation simulation, the maximum frequency deviation of the system is reduced by 57.1% and the frequency fluctuation range is reduced by 53.8%, effectively improving the frequency quality of the power grid.

Viologen Hydrothermal Synthesis and Structure–Property Relationships for Redox Flow Battery Optimization

AbstractAqueous organic redox flow batteries (AORFBs) are an emerging technology for fire safe grid energy storage systems with sustainable material feedstocks. Yet, designing organic redox molecules with the desired solubility, viscosity, permeability, formal potential, kinetics, and stability while remaining synthetically scalable is challenging. Herein, the adaptability is demonstrated of a single‐step, high‐yield hydrothermal reaction for nine viologen chloride salts. New empirical insights are gleaned into fundamental structure–property relationships for multiobjective optimization. A new asymmetric Dex‐DiOH‐Vi derivative showcases an enhanced solubility of 2.7 m with minimal tradeoff in membrane permeability. With a record viologen cycling volumetric capacity (67 Ah L−1 anolyte theoretical), Dex‐DiOH‐Vi exhibits 14‐d of stable cycling performance in anolyte‐limiting AORFB with no crossover or chemical degradation. This work highlights the importance of designing efficient synthetic approaches of organic redox species for molecular engineering high‐performance flow battery electrolytes.

Exploiting High-Voltage Stability of Dual-Ion Aqueous Electrolyte Reinforced by Incorporation of Fiberglass into Zwitterionic Hydrogel Electrolyte.

Rechargeable zinc aqueous batteries are key alternatives for replacing toxic, flammable, and expensive lithium-ion batteries in grid energy storage systems. However, these systems possess critical weaknesses, including the short electrochemical stability window of water and intrinsic fast zinc dendrite growth. Hydrogel electrolytes provide a possible solution, especially cross-linked zwitterionic polymers that possess strong water retention ability and high ionic conductivity. Herein, an in situ prepared fiberglass-incorporated dual-ion zwitterionic hydrogel electrolyte with an ionic conductivity of 24.32 mS cm-1 , electrochemical stability window up to 2.56V, and high thermal stability is presented. By incorporating this hydrogel electrolyte of zinc and lithium triflate salts, a zinc//LiMn0.6 Fe0.4 PO4 pouch cell delivers a reversible capacity of 130 mAh g-1 in the range of 1.0-2.2V at 0.1C, and the test at 2C provides an initial capacity of 82.4 mAh g-1 with 71.8% capacity retention after 1000 cycles with a coulombic efficiency of 97%. Additionally, the pouch cell is fire resistant and remains safe after cutting and piercing.

Hierarchical Intelligent Operation of Energy Storage Systems in Power Distribution Grids

High penetration of distributed energy storage systems (ESS) offers an unparalleled opportunity to reinforce the distribution grid at the local level against upstream disruptions; however, their mass operation under uncertainty of load and renewable generation is computationally expensive. While deep reinforcement learning (DRL) has been suggested to train operator agents capable of handling uncertainty and high dimensionality of the problem, it falls short when safety and feasibility assurances are required in critical operations. This paper proposes a model for hierarchical coupling of DRL and mathematical optimization for operation of ESS in distribution grids, in order to take advantage of DRL fast response while keeping network constraints in check. In the proposed method, strategic scheduling of distributed ESS units are performed locally by fast DRL-trained agents, while critical grid-wide operations such as fault management and voltage control are performed by an optimization-based central controller. The local controller is trained by Twin Delayed Deep Deterministic Policy Gradient (TD3), whose response time is three orders of magnitude faster than stochastic optimization, while the optimality of solutions are similar in both cases.