*1.1. Motivation*

In the future, power systems will operate with a high proportion of renewable energy, which needs more flexible operational resources to compensate for power imbalances that are currently scarce. Significant intermittence in renewable resources for power systems causes great changes in real-time price, and so requires energy storage to balance power. Distributed, shared energy storage technology has high application value and will be an important and widely used resource in future power systems. At present, large-scale, pumped-storage power stations are the main energy storage resource in power systems, with costs lower than those of battery energy storage systems. High cost causes a scarcity applied battery energy storage technology in power grids. To reduce the cost of energy storage services, cloud energy storage (CES) technology, presented in [1,2], is one strategy for centralizing all distributed energy storage devices from consumers into a cloud service center, as virtual energy storage capacity, instead of real devices.

Depending on the interactive value of the consumer and power grid, a CES can be an opportunity to promote a shared business model, setting the precedent for consumers

**Citation:** Li, J.; Xing, Y.; Zhang, D. Planning Method and Principles of the Cloud Energy Storage Applied in the Power Grid Based on Charging and Discharging Load Model for Distributed Energy Storage Devices. *Processes* **2022**, *10*, 194. https:// doi.org/10.3390/pr10020194

Academic Editors: Alon Kuperman and Alessandro Lampasi

Received: 15 December 2021 Accepted: 31 December 2021 Published: 20 January 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

as energy resource suppliers. Such a shared business model can provide users more opportunity to be involved in the power market and to make energy storage more flexible. The advantages thereof are as follows: (1) having no limits to capacity from time, location, or demand; (2) having lower costs than traditional energy storage technology, only paying for leasing; (3) large-scale, flexible, shared, and distributed energy storage devices from loadside consumers and concentrated energy storage equipment from professional providers, working together in the CES environment, to huge social benefit and improved spare energy usage; (4) the consumer who has distributed energy storage participating in their business can reduce their total electricity expenses paid to the grid.

The major consumers are residents and small businesses, who are sensitive to electricity prices. Additionally, some substations within a power grid need distributed energy storage devices for emergency service. At present, to multiple energy systems (MES) a CES can be a flexible resource for end-user demand responsiveness.

Technical support for such a system includes power-load forecasting, planning optimization, communication technology, data and economic analysis, and so on. Buyers send information, such as their needed storage capacities and payments, to sellers, transferred based on the power market, communications companies, and banks. Buyers who need energy storage capacity can purchase the right of use for a given period. After obtaining the right to its use, these consumers can charge and discharge the cloud battery, according to their demand circumstances. Consumers can buy virtual energy storage capacity in a cloud network instead of building physical energy storage to reduce cost.

A CES is used as a distribution network, wherein users benefit from each other. When its battery needs charging, CES suppliers pay for the power. When its batteries discharge, CES suppliers collect fees from the consumers and the distributed power grid. Cloud energy storage suppliers need to make optimization decisions, considering cost and profit under the constraints of consumers' demand for charging and discharging the cloud battery. Then, energy storage device suppliers control real-world equipment through the building, operating, and maintenance to provide good service.

In summary, the research goal of this paper is as follows.


#### *1.2. Paper Innovation*

This paper is to solve two problems: (1) how to plan and design an orderly, controlled CES system in a realistic power system; (2) how to manage and operate the CES system. To solve the first problem, this paper presents a detailed planning drawing. To solve the second problem, this paper analyzes the load curve of five types of distributed energy storage system using Monte Carlo simulation (MCS). A planning method and principles of the CES applied in the power grid are presented. In this paper, the main innovations are as follows.

(1) The traditional research on summarizing the main factors and probabilistic models of load distribution of different types of distributed energy storage in one paper are seldom. We establish the timely and spatial distribution of charge and discharge load model considering multi-variable series of factors such as electricity price and user demand, after summarizing the impact factors from the aspects of energy storage type, battery capacity, state of charge (SOC), charging mode and user behavior.


### *1.3. Structure of the Article*

The structure of this article is as follows: Section 2 summarizes our literature review. Section 3 presents the planning method and establishes the charging and discharging load model for distributed energy storage behaviors. Section 4 analyzes the impact of the orderly or disorderly charging and discharging of different energy storage behaviors on power grid capacity, load characteristics, and safety margin, in order to summarize the application fields of CES in supporting large power grids. Section 5 concludes the paper.

#### **2. Literature Review**

Over the last ten years, multi-type energy storage technologies for micro grids [3,4] have been the focus of a large number of studies. With the development of large-scale renewable energy resources that easily integrate distributed energy resources into power systems with high power quality and operational control, the building of new transfer lines needs significant investment to meet end-user demand response; to this end, installing suitable scales of energy storage devices is being adopted as a better planning solution. In such a solution, the operation department adopts an active control strategy, such as energy storage systems, to compensate the renewable energy sources' volatility and smooth peak–valley load differences. For producing maximum comprehensive economic benefit, virtual power plants (VPP) [5,6] aggregate scheduled and non-scheduled units, including renewable and non-renewable energy resources, storage devices, and flexible loads, to operate as a single entity participating in the power market. Much literature has considered frameworks for and the modelling of components and operational systems [7]. To reduce the imbalance between the power generation and consumption due to intermittently generating units, scientifically adopting a bidding strategy for a VPP to elicit maximum benefit is a decidedly primary issue.

Electric vehicles (EV) represent mobile charging loads, as either plug-in electric vehicles or plug-in hybrid electric vehicles. Due to EVs' ability to stabilize the grid and provide significant battery storage capacity without upfront capital cost to grid infrastructure, research on the integration of vehicle-to-grid (V2G) energy storage units and cooperation with intermittent wind/PV in a VPP is a future social focus [8,9], but it's presently unrealistic. However, research on vehicle-to-grid aggregators for frequency regulation [10] and operational modeling of electric vehicle charging stations [11] provides novel ideas about distributed energy storage consumption.

The cloud energy storage (CES) systems presented in [1,2] in 2017 centralize all distributed energy storage devices from consumers into the cloud service center as a virtual energy storage capacity, belonging to the energy storage units of such VPP. The framework of the CES system for power grids is presented in [12] and built by the consumer, the CES operator, the energy storage supplier, and the distribution grid. With the information and cash flow calculated through distributed computing, the CES operator, as a centralized agency, is an intermediary service provider in contact with the consumers, the storage

supplier, and the grid. For the centralized system, considering the hierarchical architecture of the power grid, cloud–edge intelligence [13–15] for wide-area load frequency control, substation simulation and protection control, and load modeling and management, is suitable for application in CES systems. In recent years, distributed storage coordination control strategies [16] and application–research scenarios on large-scale battery energy storage systems [17] are two common fields of energy storage system research.

As for renewable energy resources and EVs, the power grid cannot afford a large volume of electric vehicles (EVs) charging at peak load time because they greatly influence the load curve [18]. To deal with the peak load regulation and demand response for a VPP supporting EVs, three solutions are: optimization dispatch [19,20], control strategies [21], and trade mechanisms [22]. Distributed computing is the main solution for solving the distributed EVs integrated into the power market, e.g., multi-agent intelligent computing [23] and distributed online algorithm [24]. Blockchain represents realistic market and dispatch mechanisms for EVs [25–29] by its technical characteristics e.g., transparent, untampered with, privacy protection, and smart contract, allowing the tracing of EV charging and the extension of its leasing activity.

In recent three years, blockchain technology has led to a complete set of distributed energy trading and supply systems by connecting energy producers and consumers directly, greatly reducing the transactional cost of electricity and improving transactional efficiency. This enables power producers, transmission grid operators, distribution grid operators, and retail energy service providers to trade at different levels, simplifying the complex multilevel structure of current power systems. Blockchain technology can also deploy its tamper-proof characteristics in identifying the certification of carbon power and renewable energy power, directly recording renewable power, and providing the convenience of credible transactions of renewable energy. Building a power-trading platform with blockchain technology will be a technological upgrade to the current power-trading market.

Blockchain, a distributed database technology, has changed some major application fields of the internet of things (IoT) network over the last five years [30,31]. Introduced by Satoshi Nakamoto's Bitcoin in 2008 [32] as a peer-to-peer (P2P) system for distributed computing and decentralized data sharing, blockchain is composed of a distributed series of blocks, linked together by their hash values, that has the characteristics of decentralization, time traceability, autonomy, openness, and tamper-proof information by using time stamps, asymmetric cryptography, distributed consensus, and flexible programming technique. The application domains of blockchain technologies in IoT, e.g., the internet of vehicles, the internet of energy, the internet of cloud, fog computing, etc., are surveyed in [33]. Blockchain techniques, applied in the Chinese energy internet, is surveyed in [34,35]. The architecture and functionality of the blockchain groups in the intelligent distributed electrical energy systems presented in [36,37] is a realizable model for future implementation. At present, realistic engineering samples are so few that most research focuses on the blockchain's framework design concerning power system planning [37–41] and market transaction infrastructure [42–48].

### **3. Planning Method and Modelling**

This section presents the architecture design for a CES and then builds a load model of the charging and discharging of distributed energy storage.

### *3.1. Planning Method*

#### 3.1.1. Architecture Design for CES

Figure 1 shows the architecture design for the CES market trade for local and crossregion power grid. Figure 2 is the unified model architecture for trade participators, a flowchart, and trade codes. The business participators are as follows:

(1) Consumers: The consumers include the users with wind or PV resources, small commercial users (e.g., load or electric vehicles), distributed generation (e.g., wind farms or PV stations), and the power plant. Distributed generation represents a certain scale

of "wind farm /PV station and battery storage (BS)". Small commercial users include the common industrial/commercial loads or batteries. Fossil-fuel power plants output power to charge energy storage devices through the distribution grid. They send fixed addresses—an ID number—and administrator information as block data.


Local consumption of renewable energy is better than long-distance transmission to the power grid. The cost of centralized labor management of operation and maintenance is so high that the smart distributed management system needs user-end intelligence.

(**b**)

**Figure 1.** The architecture of a market trade business: (**a**) energy path system; (**b**) market trade business model architecture.

**Figure 2.** Unified model architecture for trade participators, flowchart and trade code.

3.1.2. Planning of the CES System in a Power Grid

To build the CES system in a realistic power system, this paper presents the planning ideas as follows:

The cloud energy storage operator only needs to build a software platform that has a basic distribution grid with energy storage units. During the planning process, the most important concern is the power/energy flow, information flow, and cost flow. Due to the separation of current dispatching and transaction platforms, the cost flow can be in the market transaction platform. The realistic terminal devices that use these three kinds of flows are the consumers' electrical appliances and all kinds of energy storage device, which are connected by the distribution grid. There are two kinds of energy storage devices, namely the fixed energy storage stations and distributed mobile energy storage devices. The fixed energy storage stations, because of their large capacities and high cost, need to be built by the power grid or by some significant entrepreneurial investment. Mobile batteries feature location flexibility, without a demand to be located, as they can be anywhere. Thus, no realistic building construction cost in the CES system that only the mobile battery vehicles and the charge/discharge interface stations need to be invested instead.

The CES operator is responsible for the following tasks: building an information cloud detailing all service providers and energy storage suppliers, receiving consumers' charge/discharge capacities and energy demands, receiving electric vehicles' charging demands and locations and navigating them to the nearest charging stations, optimizing dispatch suggestions, load shifting from the peak to the valley, smoothing output power, tracking planned output power, ancillary services, primary frequency control in wind/PV and storage systems, meter monitoring, solving the problem of abandoning wind and PV power in order to build a friendly power supply, and improving system operational flexibility. These tasks can be set as functional modules in the desktop platform to be chosen by consumers.

The aim is to operate economically, reliably, with high quality, and efficiently. Costs needs to consider initial investment in the platform and the operation of charging stations based on users' reported installation of system capacity and energy demands. Then the cost is converted into the service price (currency/kWh). The platform needs to manage payment of all functional modules.
