Virtual Energy Storage System Research Articles

Economic operation of an agent-based virtual storage aggregated residential electric-heating loads in multiple electricity markets

The increasing penetration of intermittent renewable energy sources leads to a higher demand for flexible balancing service in the power system. Consequently, demand-side management techniques, such as utilizing flexible resources from residential electric-heating loads, provide a promising solution. However, due to various uncertainties such as stochastic weather and market prices, it is challenging to develop an economically efficient strategy. Hence, this paper proposes an economic operation optimization approach for residential electric-heating loads to develop optimal bidding and scheduling strategy in the multi-stage market. First, the dynamic characteristics of aggregated electric-heating loads are modeled as the virtual energy storage systems (VESS) to quantify the flexibility potential arising from the combination of building thermal inertia, user tolerance and thermal energy storage (TES). Subsequently, the joint energy and balancing service market is introduced to exploit the operating flexibility of VESS. A multi-agent deep reinforcement learning (MADRL) optimization framework is proposed to address the multi-objective problem at different decision frequency, which involves achieving lower economic costs while maintaining indoor comfort. Finally, we combine the DRL with imitation learning (IL) by introducing the behavior cloning mechanism to mitigate training process instability. It reduces the computational costs and eliminates the need for redundant additional structure. Comparative simulation experiments demonstrate the efficiency of data-driven optimization approach, which provides an adaptive, economic cost-effective and general solution for multi-type aggregators to activate their flexibility in dynamic markets.

Cryptocurrency mining as a novel virtual energy storage system in islanded and grid-connected microgrids

Renewable electrical energy (such as: solar and wind energies) generation in microgrids (MGs), is gaining attention to reduce greenhouse gas emissions. Microgrid operators (MOs) aim to create self-sufficient, environmentally sustainable grids, increasing the capacity of renewable energy sources (RESs) by up to 100%. Despite of the benefits of this trend, challenges arise from non-controlled characteristics of these power generations and their seasonal variations, causing fluctuations and renewable energy curtailment. Although the technical solutions; such as: the demand response (DR) programs, and the conventional electrical energy storage systems (EESSs) can help, however those may face limitations in countries with high seasonal energy generation and consumption variations. This paper introduces cryptocurrency mining loads (CMLs) as innovative virtual energy storage systems (VESSs), named cryptocurrency energy storage systems (CESSs). It proposes a structure to store excess renewable energy in cryptocurrency units (CCUs) like Bitcoin (BTC). CESSs can be charged during off-peak intervals and, conversely, they discharge during high-demand periods to reduce the overall operational cost of MGs. Furthermore, it presents a new energy management system (EMS) formulation for the optimal operation of MGs in the presence of CESSs, providing an opportunity to generate additional electricity from RESs and to mitigate renewable energy curtailment. This paper explores the optimal operation conditions of both islanded and grid-connected MG with the proposed CESS. Utilizing a dataset from an island in Finland as a practical MG, its effectiveness is demonstrated through several case studies. The results of one case study in this paper demonstrate that the proposed CESS can decrease the operating cost of the MG by about 46.5%. Additionally, it is showed that by application of CESS the renewable energy curtailment is significantly reduced, and approached zero.

Flexibility enhancement of multi-district DISCOs considering a trade-off between congestion and extractable reserve capacity from virtual energy storage systems

One of the fundamental concerns of distribution companies (DISCOs) is how to satisfy security requirements under uncertain conditions and real-time contingencies. To address this concern, it is vital to employ operational flexibility levers aimed at mitigating congestion and retaining optimal reserve capacity, particularly in critical feeders. Increased interaction between DISCOs and virtual energy storage systems (VESSs) offers an unparalleled opportunity to unlock the energy flexibility of distribution grids at the local level without increasing operational complexity. According to the state-of-the-art, models that tap the economic performance of VESSs while respecting the flexibility and security setpoints adopted by DISCOs in a distributed manner have remained scarce. This study goes beyond by developing a flexibility-oriented decision-making strategy as a single-leader multi-followers stochastic model; in which a multi-district DISCO involves the flexibility setpoints and congestion management strategy in the market participation plan of each VESS locally to enhance the energy flexibility of each district. In the upper level, DISCO aims to deal with stressful situations in real-time sessions by (i) determining the minimum reserve capacity requirements in each district to be provided by VESSs with respect to the existing uncertainty sources and the importance of each district, and (ii) alleviating the congestion in critical feeders to release the feeder capacity for deployment of the available reserve power promptly. In the lower level, each VESS makes optimal decisions on power and reserve bidding in day-ahead, reserve, and real-time markets that are affected by the flexibility services required by the DISCO. Inspired by the activity schedule of different users, it also examines how demand response programs can affect DISCO’s ability to unlock consumer-side flexibility in realizing congestion management strategies. The developed bi-level scheduling model is first reformulated as single-level mathematical program with equilibrium constraints (MPEC) by employing Karush–Kuhn–Tucker (KKT) optimality conditions and then linearized into a mixed-integer linear program (MILP). The proposed strategy is validated through extensive simulations conducted on the modified IEEE 33-bus test system consisting of three districts with different users. Furthermore, DIgSILENT PowerFactory is used to verify the feasibility of the proposed model under contingency cases and to provide insights for DISCOs to control multiple VESSs and capacity-correlated uncertainties.

A real time peer-to-peer energy trading for prosumers utilizing time-varying building virtual energy storage

With the increasing penetration of distributed energy resources (DER) in the electric power system, Peer-to-Peer (P2P) energy trading has become a promising paradigm for future electric power systems. Building thermal load, which is an important demand side resource, should be considered carefully in the design of a P2P trading method. In this paper, we investigate the application of building thermal energy storage capability in P2P energy trading. We aggregate and model building thermal loads as a virtual energy storage and derive a time-varying virtual energy storage system (T-VESS) model to quantify the flexibility of a building. Key parameters of T-VESS, including charging/discharging rate, energy capacity, and state of charge (SOC), are analyzed so that T-VESS can be embedded in the building prosumer model to participate effectively in electricity-oriented P2P energy trading. We propose a real time P2P energy trading method of prosumers based on model predictive control (MPC). This method consists of an energy supply and demand quantification stage and a transaction price optimization stage, which can effectively capture the time-varying nature of T-VESS. To preserve the trading preference for prosumers, we propose a distributed P2P energy trading actions implementation method based on the continuous double auction (CDA) considering multiple trading preference grades. Numerical results demonstrate that the proposed P2P energy trading method considering T-VESS can reduce the operational cost of prosumers by 3.7%, while also promoting the local integration of renewable energy by 3.1%. Furthermore, compared to existing P2P trading methods considering single preference grade, the proposed method greatly enhances the utilities of both individual prosumers and the overall community.

Virtual energy storage system for peak shaving and power balancing the generation of a MW photovoltaic plant

This article proposes a novel control of a Virtual Energy Storage System (VESS) for the correct management of non-programmable renewable sources by coordinating the loads demand and the battery storage systems operations at the residential level. The proposed novel control aims at covering two main gaps in current state-of-the-art VESSs. The first gap is considering a distributed battery storage system instead of a centralized one, the second gap is providing the electricity grid operator with two services instead of one. To this aim, the authors explore a VESS consisting of residential buildings where each apartment is equipped with an air conditioner but also with a battery storage system. The explored VESS provides the grid operator with both peak shaving and power balancing services for the generation of a megawatt photovoltaic plant located near the VESS. The goodness of the proposed coordinated control is demonstrated via numerical experiments and using real data, measured every 15 min in September 2019. The case study consists of a 1.4 MW photovoltaic plant located near a small town, 21 residential buildings with 168 apartments, each equipped with an air conditioner (continuous power is 1.5 kW) and battery energy storage systems (3 kW /2.5 kWh). The numerical results show that the battery energy storage systems are charged correctly during peak hours (the charging power is between 0.45 and 0.90 kW, and the state of charge varies from 20 % to 78 %) and that the residual photovoltaic plant generation resembles a horizontal line. Later, in the early afternoon, the reference temperature of the air conditioners and the charge/discharge of the battery storage systems are suitably adjusted by solving a mixed linear integer programming problem, to balance the reduction in photovoltaic plant generation, which lasts an hour and a half and peaks at 188 kW. Finally, the numerical results also show that the energy that remained in the batteries is entirely consumed by users in the late afternoon or evening and that the amplitude and the duration of the so-called “load rebound” are so slight that no compensation action (e.g., the bath returning or linear recovery strategy) is required for the considered case study.

Sponge Grid With Numerous Virtual Energy Storage Systems: Concept, Model, and Decentralized Control

High proportion of energy storage systems (ESSs) and flexible loads signify the main features of a modern power system. ESS with its bi-directional flow characteristic can flexibly change power network operations, thus providing a new solution for voltage regulation and control. However, since ESS resources are dispersed throughout the power system, it is necessary to design an effective aggregation scheme to achieve a concise voltage regulation. In this paper, a novel sponge grid is proposed, which is capable of both local and global tasks to offer greater flexibility and initiative in power system operations. On the one hand, it constructs the source and forms virtual energy storage (VES) systems to satisfy local demands. On the other hand, the sponge grid applies the superposition of a large number of VES systems spontaneous effects to present a collective voltage regulation behavior. Then, a message passing-based decentralized algorithm is proposed and designed for the local voltage regulation in the sponge grid. Various constraints including those representing inequality boundary, state of charge, and complex line flows, are expressed analytically and computed fast by the proposed portable projection approach. The proposed sponge grid model is tested on a modified IEEE 33-bus system to verify its effectiveness and superior performance.

A data-driven rolling optimization control approach for building energy systems that integrate virtual energy storage systems

The virtual energy storage system (VESS) is an innovative and cost-effective technique for coupling building envelope thermal storage and release abilities with the electric and heat power conversion characteristics of an air conditioner; this system provides building energy systems (BESs) with adjustable potentials similar to those of conventional battery energy storage systems (BESSs). However, the VESS is a dynamic system, and uncertainties in the outdoor temperature and solar irradiance are difficult to accurately predict, which impacts the quantification accuracy of VESSs; these characteristics challenge the BES control scheme economy and the thermal comfort of occupants. To solve this crucial issue, a data-driven rolling optimization (RO) control approach for a BES that integrates a VESS is proposed. First, a BES state space model integrating the VESS is created to reflect the VESS adjustable potential and dynamic characteristics. Based on the above model, while aiming at a small BES data sample size, a support vector machine (SVM) is combined with RO to correct the day-ahead quantification errors of the VESS adjustable potential and enhance the economical operation and thermal comfort of the BES that integrates the VESS in uncertain environments. Comparative simulations validate the effectiveness of this VESS modelling and data-driven RO control approach.

Capacity of Virtual Energy Storage System for Frequency Regulation Services via a Data-Driven Distributionally Robust Optimization Method

Due to large thermal inertia of buildings and flexibility of interruptible loads, smart buildings pose a remarkable potential for developing virtual energy storage systems (VESSs). However, current literature lacks advanced models to quantify and thus properly optimize available capacity of VESS for power system ancillary services, especially frequency regulation services (FRS). This paper, firstly, presents a novel probabilistic model for explicitly quantifying the VESS capacity in charging and discharging modes, which can further be optimized by scheduling building loads. While the optimized capacity is fully available for FRS, building customer economic benefit and comfort could considerably be impaired by various uncertainties including weather and user related parameters, as well as FRS requests. Due to limited information of the uncertainties, common model-based stochastic and robust optimization approaches are inefficient, leading to either poor out-of-sample performance or over-conservative solutions. Thus, secondly, this paper develops a data-driven distributionally robust optimization (DRO) method to robustly optimize the capacity of VESS against the worst distribution of the uncertainties with a finite training dataset. The proposed data-driven DRO-based model can well match modern data-driven operation environment. Numerical simulations validate the high efficiency and out-of-sample performance of the proposed VESS capacity optimization method under the uncertainties.

An ensemble learning model for estimating the virtual energy storage capacity of aggregated air-conditioners

Renewable energy resources (RES) pose several challenges due to their natural intermittency when integrated into a distribution network. A smart energy storage system (SESS) alleviates these challenges, which is achieved by integrating thermostatically controlled loads (TCLs) such as air conditioners (ACs). Hence, this study proposes a virtual energy storage system (VESS) by modeling the ACs analogous to an electro-chemical battery. Besides, non-linearity in energy consumption patterns of ACs due to consumer behavior makes predictive energy analysis critical to deplete the discrepancy between supply and demand. Therefore, this study proposes an ensemble learning (EL) technique to predict the virtual energy storage (VES) capacity of ACs. The prediction stage relies on various machine learning (ML) methods, like support vector regression (SVR) and artificial neural network (ANN). Based on these ML techniques, this study proposes an SVRs-ANN-based EL model to predict the VES capacity during the discharging cycle. Each weak learner in the EL model is trained using a significant features subset with the aid of the k-fold cross-validation method to build a more generalized model for unseen data. The performance of the EL model is estimated using minute-based smart meter data set and empirically manifested that the EL model outperforms conventional approaches. The R2 and root mean squared error (RMSE) of the EL model reach 1 and 1.03×10−7, respectively, and results demonstrate lower RMSE values and higher performance than conventional methods. The results obtained have shown the robustness of the proposed EL model to operate ACs in energy-saving mode instead of load shifting. Although ACs are not controlled in the experiments, the experimental validation proved that the EL model accurately predicts the daily VES capacity of ACs without violating consumer satisfaction.

Optimal scheduling of multi-energy type virtual energy storage system in reconfigurable distribution networks for congestion management

The virtual energy storage system (VESS) is one of the emerging novel concepts among current energy storage systems (ESSs) due to the high effectiveness and reliability. In fact, VESS could store surplus energy and inject the energy during the shortages, at high power with larger capacities, compared to the conventional ESSs in smart grids. This study investigates the optimal operation of a multi-carrier VESS, including batteries, thermal energy storage (TES) systems, power to hydrogen (P2H) and hydrogen to power (H2P) technologies in hydrogen storage systems (HSS), and electric vehicles (EVs) in dynamic ESS. Further, demand response program (DRP) for electrical and thermal loads has been considered as a tool of VESS due to the similar behavior of physical ESS. In the market, three participants have considered such as electrical, thermal and hydrogen markets. In addition, the price uncertainties were calculated by means of scenarios as in stochastic programming, while the optimization process and the operational constraints were considered to calculate the operational costs in different ESSs. However, congestion in the power systems is often occurred due to the extreme load increments. Hence, this study proposes a bi-level formulation system, where independent system operators (ISO) manage the congestion in the upper level, while VESS operators deal with the financial goals in the lower level. Moreover, four case studies have considered to observe the effectiveness of each storage system and the simulation was modeled in the IEEE 33-bus system with CPLEX in GAMS.

Distribution locational marginal price based hierarchical scheduling framework for grid flexibility from virtual energy storage systems

• VESS's are modeled using residential air-conditioners for DSM. • Local ancillary services TSO-DSO coordination scheme for cooperative management in grids. • Hierarchical framework for scheduling aggregated VESS. • Network constrained optimal VESS participation in ancillary service markets. • DSM based practical market model for SOs, maximizing overall market benefits. Maintaining generation and demand equilibrium under large scale renewable energy integration is challenging. Conventional flexibility providers such as grid-scale Energy Storage Systems (ESS) have potential but are costly solutions. With evolving smart grid scenario, aggregated demand response from Thermostatically Controlled Loads (TCLs) is utilized as Virtual Energy Storage Systems (VESS) and considered as innovative and cost efficient solution to provide fast response reserves. VESS units are often scheduled and controlled in a centralized manner considering limited distribution system constraints. However, accurate operational flexibility potential of diversified VESS units for various grid services needs to consider distribution network constraints. It can be fully utilized when Distribution System Operators (DSO) coordinate with Transmission System Operators (TSO) to transfer grid services from VESS of its area. Local ancillary service TSO-DSO coordination scheme ensures in this perspective significant flexibility of TSO and allows both TSO and DSO to access VESS flexibility. For such controlling of diversified VESS units where DSO is responsible for aggregate flexibility and local area balancing, a hierarchical scheduling framework is more practical. In this context, this paper proposes a two-stage hierarchical approach for optimal scheduling of TCLs as VESS to optimize energy and provide grid flexibility. At the lower stage, TCLs are scheduled based on Distribution Locational Marginal Price (DLMP) provided by DSO, while ensuring consumers’ comfort. At the upper level, DSO provides additional flexibility to TSO through optimal scheduling of aggregated VESS considering local ancillary services TSO-DSO coordination scheme. The simulation results indicate that VESS scheduling improves system flexibility by maintaining a generation-demand balance.

Improved Resilient Model Predictive Control for Enhanced Microgrid Virtual Inertia Emulation by Virtual Energy Storage System Under DoS Attacks

Improved Resilient Model Predictive Control for Enhanced Microgrid Virtual Inertia Emulation by Virtual Energy Storage System Under DoS Attacks

Environmental assessment of prosumer digitalization: The case of virtual pooling of PV battery storage systems

The emergence of decentralized renewable energies together with digitalization enable new possibilities for the provision of system services that are needed for the energy transition. In Germany and worldwide, prosumers with a PV battery storage system are on the rise. With the aid of digital information and communication technologies (ICT), energy service providers can pool individual battery storage units to form a virtual battery energy storage system (VBES). The VBES is supposed to provide flexibility services and, thereby, support the transition towards an energy system with high shares of intermittent renewables. Although the concept is environmentally motivated, a holistic environmental assessment is missing so far. Therefore, we carry out a life-cycle analysis that considers environmental benefits, as well as burdens of VBES. For this, we compare a PV battery storage unit in a German single-family household with and without participation in a VBES for the year 2018. We assess three levels of environmental effects: direct effects due to digitalization (data transfer, ICT resources for VBES operation), indirect effects from external optimization on the household-level (changes in battery life and power supply), and enablement effects on the system level (provision of Frequency Containment Reserve, redispatch, and preventing renewables curtailment). We find that digitalization has a comparably low direct impact but it can enable relatively high environmental benefits for climate protection, and in other impact categories. Conditions for our findings are that battery life and the household's degree of self-sufficiency with PV electricity are not significantly affected by the external control of the VBES operator. We conclude that current and future battery owners should be encouraged to allow external access for system services. Furthermore, the necessary ICT roll out should be implemented in the short- to mid-term.

Optimal control of source–load–storage energy in DC microgrid based on the virtual energy storage system

By integrating controllable source-load in the form of virtual energy storage into the energy storage control system within the DC microgrid, the virtual energy storage system (VESS) with flexible resources can provide a viable solution for the system to effectively accomplish the co-optimization of source–load–energy storage. Firstly, based on the conversion relationship between the rotational kinetic energy of the wind turbine and asynchronous motor and capacitive energy storage, the wind turbine and desalination load are established as virtual energy storage equivalent models; as electric vehicles have both mobile load and energy storage characteristics, the tariff mechanism is used to develop an orderly charging and discharging model to establish a virtual energy storage model for electric vehicles. Secondly, the economic optimization model of the VESS is established jointly with source–load–storage to maximize the intra-day microgrid operation gain, and an energy cooperative optimization control strategy is formed with virtual capacitance value and virtual charge state as operating parameters in each period of time. Finally, it is verified through simulation that the source–load–storage​ energy co-optimization control can not only relieve the regulation pressure of energy storage in the DC microgrid but also significantly improve the operation economy of the microgrid.

Market-Based Resource Allocation of Distributed Cloud Computing Services: Virtual Energy Storage Systems

The cloud-based application is a major developing feature of smart grids. Apart from centralized Internet data centers (IDCs), distributed cloud resources (CRs) can also provide cloud computing services with low latency and high reliability. For the distributed cloud computing, CRs aggregators (CRAs) will integrate the distributed idle computing resources, which are dispersed energy consumers in the system, to form virtual IDCs. This article presents a market-based computing resource allocation method for distributed cloud computing services. The computing resource allocation refers to the approach to allocating the computing workloads to different CRs. First, the batch workload scheduling (BWS)-based virtual energy storage system (VESS) model and thermal inertia (TI)-based VESS model are proposed to help CRAs better aggregate the distributed CRs and characterize the energy consumption flexibility of the virtual IDCs. Then, the energy trading behavior of the CRAs in the transactive energy market is modeled in the resource allocation process. Case studies are conducted on a 55-bus electricity system. It can be found that energy consumption costs can be reduced by applying the proposed methodology, and revenues from providing cloud computing services can be increased.

An optimal control method of microgrid system with household load considering battery service life

Since the disorderly charging of a large number of electric vehicles is not conducive to the stability of the power grid, aiming at the storage function of power batteries for electric vehicles, an optimal control method for microgrid systems with household loads based on electric vehicle virtual energy storage is proposed in this paper. Taking the minimum of electricity cost, capacity attenuation of the battery and carbon dioxide emission as the objective functions, and fully considering the constraints of the charge and discharge power and the available capacity of electric vehicles, an optimal control model for microgrid systems with household load is established based on the multi-objective optimization. The optimal Pareto solution set of the model is obtained by the Non-dominated Sorting Genetic Algorithms III (NSGA-III) algorithm, and the optimal solution is solved by the technique for order preference by similarity to the ideal solution (TOPSIS) algorithm. The effectiveness and correctness of the proposed method are verified by the simulation examples from the actual operation data of the microgrid system containing a photovoltaic station and household load in China. The simulation results show that electric vehicles participating in virtual energy storage not only improve the stability of the power grid operation, but also bring some economic benefits to electric vehicle users. The electricity cost is reduced by 829.0USD, the total charging and discharging of the electric vehicle virtual energy storage system is 2703 kWh, the average service life of the battery is 7.3 years, and the carbon dioxide emission is reduced by 166.37 kg.

Fair Virtual Energy Storage System Operation for Smart Energy Communities

A virtual energy storage system (VESS) logically shares a physical energy storage system among multiple units. In resource sharing, the distribution of benefits is a critical problem. As a resolution, this study proposes a fair VESS operation method for smart energy communities that involve groups of energy consumption units. First, the cost and resource fairness indices are defined as the benefit and VESS usage proportional to the investment cost, respectively. The fair VESS operation problem is formulated considering the fairness indices that could be solved optimally by applying gradient methods without additional computational burden. The simulation results using the dataset in Korea demonstrate that the proposed operation allows the fair distribution of the benefit and resource usage among units with a marginal benefit reduction of approximately 5% in relation to the VESS operation to maximize the benefit. Moreover, it is shown that the resource fairness that controls the VESS usage limits the total benefit, and the cost fairness distributes the benefit among units according to the cost contribution. Furthermore, the proposed VESS operation can manage the VESS lifetime and improve the system performance of the utility grid.

Development of Grid Tied Inverter for Indian Grid and Operating Environmental Condition

Solar Photovoltaic (PV) technology is one of the renewable energy source technologies to realize the shift to decarbonize energy supply because of availability of safe, limitless, free and reliable long-term sources of power. PV modules/cells are instruments to convert this solar energy to DC energy. Grid interconnection with solar power system is comparatively easier to install as the same do not require storage system. This paper presents injection of 3 kW power generated from solar PV array to 230V grid distribution network. Technique uses phase angle difference between grid voltage and voltage source inverter. The grid acts like a virtual energy storage system with an unlimited storage capacity. Hence, the proposed system is designed without a battery backup. This system is simulated in MATLAB/Simulink to confirm the desire output. In the second step, hardware circuit is designed and implemented based on the simulation results.

Dynamic Virtual Energy Storage System Operation Strategy for Smart Energy Communities

The concept of a virtual energy storage system (VESS) is based on the sharing of a large energy storage system by multiple units; however, the capacity allocation for each unit limits the operation performance of the VESS. This study proposes an operation strategy of a dynamic VESS for smart energy communities. The proposed VESS operation strategy considers the usage-limited constraint rather than the capacity allocation constraint and it guarantees the usage of VESS resources of each participant for an operation period. Therefore, the degrees of freedom for VESS operation can be increased at each operation time. The dynamic VESS operation problem is formulated as a mixed-integer linear problem that could be solved optimally by applying gradient methods and dual decomposition. The dataset of a VESS in Korea is used for simulation. The simulation results demonstrate that, when the proposed operation strategy is used, the cost efficiency achieved is more than twice that achieved when the existing VESS operation strategy is used. Furthermore, the proposed strategy accurately reflects the characteristics of the participants; thus, more units can participate in the VESS operation service. The proposed VESS operation can improve the system performance of the utility grid and increase the net benefit of the participants.

Prediction of virtual energy storage capacity of the air-conditioner using a stochastic gradient descent based artificial neural network

Integration of renewable energy sources (RES) poses numerous challenges in smart grids, as RES are largely dependent on natural intermittency. Hence, it is necessary to rely on smart energy storage systems (SESS) to mitigate the fluctuation of RES on the supply side. SESS can be achieved by using demand response management (DRM), i.e., by aggregating thermostatically controlled loads using state-of-art smart grid technologies. In this paper, the air conditioners (ACs) are aggregated into a virtual energy storage system (VESS) by employing an electric model of the ACs. A simple mathematical model was described to evaluate the charging and discharging pattern of the load in terms of power and energy. Based on the VESS, the temperature control strategy was designed to reduce the power consumption of the ACs when making sure the room temperature is below the permissible limit. An artificial neural network (ANN) model was designed to predict the energy capacity of ACs, using which the VESS is achieved. Besides, the stochastic gradient descent (SGD) optimization algorithm was implemented in the back-propagation of the ANN model to achieve the optimum prediction values. Four case studies were presented to demonstrate the energy-saving capacity of ACs by regulating the set point temperature of a selected site. The results illustrate the change in the power consumption pattern of the ACs with and without employing them for virtual energy storage (VES). Hence, the proposed ANN model enables the aggregators to achieve the desired VES without affecting the comfort of the occupants.