Distributed Energy Resources Management System (DERMS) and Its Coordination with Transmission System: A Review and Co-Simulation

Abstract

Ever-increasing penetration of distributed energy resources (DERs) in the power grids, alongside their numerous benefits, brings new challenges that call for enhanced solutions in the field of control and management of power grids. The majority of the available research have considered either distribution or transmission grids in their studies. In this paper, a comprehensive review of the effects of DERs on the distribution and transmission grids is performed. The focus of this paper is on hierarchical management methods in order to categorize different approaches and highlight the gaps. Moreover, a review is conducted in the field of the newly introduced distributed energy resources management system (DERMS) concept. A DERMS can facilitate the hierarchical energy management procedure due to its functionalities and broad capabilities. Hence, its implementation in energy management and its impact on the power grid will be assessed with the aid of a co-simulation platform that considers both transmission and distribution grids.

1. Introduction

In recent years, fossil fuel challenges and pollutant gases crises on the one hand, and numerous advantages of distributed energy resources (DERs) [1] on the other hand, created a growing trend of DERs’ penetration in power systems. Furthermore, DERs are utilized in microgrids (MGs) and multi-microgrids to satisfy the need for higher reliability and power quality [2,3]. Consequently, researchers have focused on solutions to facilitate the integration of DERs into power systems to deal with obstacles. In this regard, recent advances in the power electronics interfaces pave the way for replacing conventional power generation with DERs [4,5].

By integrating DERs into a conventional passive distribution network, the flow of power will be bidirectional, resulting in an active distribution grid [6]. Predictably, this process is followed by serious challenges. Numerous studies are focused on proposing solutions for the problems raised by DERs. These remedial actions target challenges related to power systems’ planning [7,8,9] and operation [10,11,12], uncertainty of DERs [13,14,15], and mal-operation of protection systems [16,17,18,19].

The mentioned issues are complicated enough to make distribution system operators (DSOs) unable to respond as they have in the passive networks [20]. Hence, new solutions are required to overcome the difficulties of future grids. In this regard, the majority of the proposed methods can be categorized into two categories: (a) hardware-based and (b) software-based remedies. The former tries to resolve the raised issues using advanced and intelligent hardware assets. For instance, power electronic interfaces are introduced for DERs to facilitate their connection to the power grid and enhance their capabilities [21,22]. On the other hand, in the second category, software solutions are provided based on the communication infrastructure of smart grids and using the intelligent assets introduced in the first category.

This paper will focus on the software-based solutions by conducting an extensive literature review. The goal is to analyze the newly introduced concept of the distributed energy resources management system (DERMS) and its functionalities facing management and control challenges of DERs. Furthermore, its commonalities and disparities with available energy management techniques will be investigated. Additionally, the presence of the transmission network in the energy management problem will be reviewed. After analyzing the concept and highlighting available gaps, a method will be proposed and simulated to address the challenges of the management process of DERs.

The remainder of this paper is organized as follows. In Section 2, a comprehensive review of the field of study of this paper is conducted. The theoretical aspects of the proposed method including mathematical formulations are examined in Section 3. Next, the proposed method is evaluated using simulations and the results are presented in Section 4. Finally, the paper is concluded in Section 5.

2. Literature Review

This section provides an overview of relevant research around the scope of this paper. A summary of the research material selection process is illustrated in Figure 1. According to this figure, a total number of 153 items are selected out of the initial 297 studies (Figure 1a). Additionally, the total number of papers per keyword is shown in Figure 1b. Moreover, due to the correlation between DERMS and other keywords, the co-occurrence map of the DERMS keyword considering 136 results obtained from the Web of Science database is illustrated in Figure 1c. According to this figure, there are two main research areas, namely planning and operation. Additionally, the optimization and strategy keywords have the most co-occurrence with the DERMS keyword.

2.1. Distributed Energy Resources Management System

A DERMS is a software platform to manage and control DERs and demand response (DR) in a more efficient way. The main functionalities of DERMS include aggregation, simplification, optimization, and translation of the different languages of DERs and DR to the upstream entity [23]. For instance, it enables end-customers (e.g., DERs and controllable loads) to offer their services to the upper authorities [24]. A real-time application of a commercially available utility DERMS solution called EcoStruxure DERMS is discussed in [25] in order to show the capability of these software solutions in turning challenges into opportunities.

Besides DERMS benefits and advantages, its concept is newly introduced and most of the functionalities and definitions are not fully clear. A systematic review of the available DER management solutions was carried out in [20] to characterize the functions of different hierarchy levels in DERMS. Moreover, the authors enumerate the features of centralized and decentralized management strategies.

According to the available literature, aggregator DERMS and utility DERMS are the two types of DERMS that exist. The authors in [26,27] highlighted the differences between them. Accordingly, the main goal of an aggregator DERMS is to collect different small-scale DERs, behind-the-meter DERs for instance, and represent them as a larger DER to the system operator. Hence, aggregator DERMS can participate in market operations, demand response, and load shedding without being aware of the system’s model or constraints. On the other hand, utility DERMS is able to manage aggregator DERMS, demand response (DR) programs, medium/large-scale DERs, and other resources with regard to the system constraints [28]. A schematic of the hierarchical structure of the DERMS framework is presented in Figure 2. According to this figure, DSO is at the top and can utilize a distribution management system (DMS) or an advanced distribution management system (ADMS) to manage the network under its control. Afterward, utility DERMS is supposed to be in direct connection with DSO while controlling aggregator DERMS, medium/large-scale DERs, virtual power plants (VPPs), and MG controllers. Finally, small-scale DERs, behind-the-meter DERs, and MGs are at the lowest level.

DERMS has been studied by researchers to demonstrate its capabilities. In this regard, the impact of the different penetration levels of the residential and small commercial photovoltaic (PV) systems on the distribution system in terms of technical and economic aspects is investigated in [29]. Three possible solutions including traditional infrastructure upgrades, autonomous volt-var controls, and DERMS are considered. While volt-var control results in the lowest cost, the DERMS solution presents the higher hosting capacity for the grid. Hosting capacity is the number of DERs that a distribution grid can reliably integrate without significant grid upgrades as defined in [30]. Moreover, a novel approach to mitigate sustained high voltage and large voltage fluctuations due to the higher penetrations of DERs is presented in [31]. Authors in this paper have proposed a DERMS platform that is able to generate and send curtailment signals to the customer site controller during higher voltages. In another study, the transient over/under voltages were addressed in the presence of large PVs, electric vehicle (EV) charging stations, and energy storage systems (ESSs) using a reactive power demand response DERMS in [32]. A new distribution-based interactive operation planning method is described in this paper that overcomes the complexity of centralized planning by assuming the variable information related to the problem (e.g., desired demand or generation) to be submitted by the distributed entity itself. This will be possible by exchanging information with other entities (at the same level) or aggregator/system operators (at the higher level) that will send information, e.g., electricity prices. Another DERMS platform consisting of two modules responsible for pre-analysis and minimization is designed in [33]. The pre-analysis module tries to construct the controllable source initial dispatch curves and handle the energy constraints for the storage system. On the other hand, the second module attempts to minimize the objective function, which is the microgrid’s operation cost, using the pattern search optimization technique.

An application of DERMS involving the power system’s technical violations, overloading in this case, is presented in [34] and utilizes priority-based optimal power flow to curtail DERs. The superiority of the proposed method is compared with the “last-connected, first-curtailed” approach, and great benefits in terms of curtailed DER capacity and technical aspects are obtained.

The application of the DSO–DERMS interaction in the critical load restoration process is investigated in [35] by considering control coordination between DERs and ESSs. Additionally, uncertainties related to DERs, e.g., wind turbines (WTs) and loads are taken into account in this research. The authors were able to overcome the aggregation challenges of large-scale DERs/ESSs and the complexity of communication and control during the load restoration process by utilizing DERMS and its coordination with DSO.

2.2. Hardware-in-the-Loop

Every DERMS platform needs to be tested before its implementation. Due to the limitations of testing proposed solutions in the presence of real power systems, real-time simulators are one the best alternatives. Hence, comprehensive tests must be completed to evaluate the coordination between DERMS, other management systems (e.g., ADMS), DERs, and conventional local controllers (e.g., capacitor banks and voltage regulators) [36]. A hardware-in-the-loop (HIL) test platform using the ePHASORsim power flow tool of the OPAL-RT as the power system simulator, as well as the cloud-based IoT hardware controller is presented in [37]. PV generation and energy storage DERs are considered in this study. The objective of the proposed DERMS solution is to manage DERs while avoiding reverse power flow and reducing peak demand at the point of connection to the upper grid. Another HIL platform is formed to evaluate the voltage regulation performance of the DERMS in the presence of hardware inverters [36,38,39] and grid-edge devices [40]. A co-simulation approach is utilized to simulate the distribution grid using the quasi-steady-state power flow solver, OpenDSS, and OPAL-RT real-time simulator. Additionally, the authors in [38] have expanded the proposed HIL testbed in [36] by increasing the hardware inverter devices with the goal of voltage regulation. The Hierarchical Engine for Large-scale Infrastructure Co-Simulation (HELICS) is utilized to coordinate the co-simulation procedure, merge all the software and hardware, and synchronize them. A detailed introduction of a platform to test a DERMS framework is provided in [41]. The authors have developed the modeling environment using the GridAPPS-D platform and different application programming interfaces (APIs). GridAPPS-D is an open-source standards-based platform for advanced data-driven distribution system applications developed by the U.S. Department of Energy at Pacific Northwest National Laboratory (PNNL) [42]. Moreover, the OpenDSS and Opal-RT co-simulation testbed is utilized in [43] to simulate an actual distribution feeder and perform voltage regulation considering coordinated control across the following: advanced distribution management system (ADMS), DERs, and DERMS. Another HIL test bed using IEEE 2030.5 standard is designed and provided in [44] to investigate the application of DERMS and analyze cybersecurity aspects of the future power grids. The mentioned platform consists of OPAL-RT’s OP4510 as the real-time simulator for DERs and the power grid, gateways, and a DERMS cloud server. A summary of DERMS application in distribution systems’ energy management is presented in

Table 1. Summary of the application of DERMS in the energy management problem.

References	Idea	Objectives	DERs Included	Test System
[29]	Impact of high penetration of small-scale DERs	Hosting capacity expansion, cost minimization, power loss minimization	Residential and small-scale commercial PV	Distribution grid
[31]	DERMS vs. line upgrade option to overcome problems of high penetration DER	Mitigate high voltage and large voltage fluctuations	PV and ESS	Distribution grid
[32]	Application of DERMS to aggregate reactive power from numerous small-scale air conditioners	Overcome transient over/under voltages	PV, EV charging stations, ESS	Distribution grid
[33]	A two-module DERMS platform by considering different electricity pricing policies	Minimizing operational costs of the MG	ESS and diesel generator	MG
[34]	Application of DERMS to curtail the most effective DERs to mitigate overloading	Minimizing DER power curtailment	-	Sub-transmission/distribution grids
[35]	Application of DERMS in the critical load restoration process and its coordination with DSO	Regulate frequency and voltage during restoration	PV, WT, ESS	Distribution grid
[37]	DERMS HIL test platform	Preventing reverse power flow at the substation during high DER generation	PV and ESS	Distribution grid
[36,38,39]	DERMS HIL test platform in the presence of hardware inverters	Voltage regulation	PV	Distribution grid
[40]	DERMS HIL test platform in the presence of grid-edge devices	Voltage regulation	PV	Distribution grid
[43]	DERMS HIL test platform in the presence of hardware inverters and local hardware controllers	Voltage regulation	PV	Distribution grid
[44]	DERMS HIL test platform to analyze cybersecurity aspects	-	ESS, PV, WT	Distribution grid

.

2.3. Energy Management

Although the DERMS concept is newly introduced, some of its functionalities were developed and utilized previously. In this regard, a comprehensive review of the control strategies in MGs and MG communities was carried out in [45] by taking into account planning and scheduling programs, categorizing optimization methods, and investigating the application of artificial intelligence (AI) solutions. Additionally, the authors in [46] focused on the active and reactive power flows, known as the tertiary control, to propose a coordinated energy management model for a power system consisting of multiple MGs and a distribution network (DN). In this research, MGs and DN are considered different entities with different objectives that will result in a stochastic bi-level problem. While the first stage problem (DSO) is solved based on the forecasted generation of DERs, the second stage (each MG) tries to modify the results of the first stage by means of the actual outputs of DERs. As another research area, the economical operation of power systems has drawn the attention of researchers. Accordingly, ESS is employed in a three-step energy management method for a MMG system in [47]. According to this method, each MG will place its bid (for buying or selling electricity) based on the electricity price. The central energy management system (EMS) will then decide whether to draw electricity from the upper grid or from MGs with excess generation. It will also determine when to charge and discharge the ESS to minimize total operational costs.

In addition to economic aspects, technical terms can be considered as the objective of energy management solutions. In this regard, a bi-level energy management solution is discussed in [48] to manage energy transactions between the distribution system and the clustered MGs. The first level of this study deals with the operation of the distribution system and minimizes fluctuations in power exchange, voltage deviations, and power losses. Meanwhile, the second level is in charge of the MGs, which aims to minimize operation costs and air pollution. Consequently, voltage regulation, minimizing power loss and operation cost, as well as operation security of the DN can be considered as the objectives of the energy management framework [49]. Moreover, energy management can be discussed in a variety of areas. Authors in [50] developed an EMS that is capable of forecasting PV generation output. The EMS will charge EVs that are parked at the workplace for long hours based on forecasted values and dynamic tariffs. This will ensure minimized total cost and stress on the grid while the maximum PV utilization is achieved. Moreover, the authors benefited from an energy management method to achieve a real-time load management system in [51] that facilitates active load control in an islanded MG. Without a shadow of a doubt, the communication infrastructure plays an important role in the energy management process. Hence, to reduce the burden on the communication system and enhance the energy management outcome, the authors proposed an event-triggered communication scheme in [52] that samples the required data when a predefined triggering function is satisfied. An overview of different energy management objectives and their respective contributions is presented in

Table 2. Summary of the energy management.

Reference	Objectives and Contributions	DERs Included	Test System
[46]	Minimizing operation costs, considering stochastic DERs, and bi-level formulation	PV, WT, microturbine	DN + MGs
[47]	Minimizing operation costs, considering two hierarchy levels, and energy exchange between MGs	ESS	DN + MGs
[48]	Minimizing fluctuations in power exchange, voltage deviations, and power losses in the distribution level Minimizing operation costs and air pollution in MG level	PV, WT, ESS	DN + clustered MGs
[49]	Voltage regulation, minimizing power loss and operational cost, and operational security of the network	PV, WT, ESS, soft open point (SOP)	DN + MGs
[50]	Minimizing cost and stress on the grid and maximizing PV utilization in the presence of EVs	PV, EVs	Charging station
[51]	Real-time load management system	PV, ESS	MG

.

2.4. Simultaneous Study of Transmission and Distribution Grids

As viewed from a transmission system perspective, the distribution system is considered as a load. On the other hand, in the distribution system studies, the transmission system is considered as a voltage source connected to the distribution substation [53]. Considering the increasing rate of integration of DERs into DNs, major differences are likely to be seen in the flow of power and its direction, which requires special analysis and methods for each power system level. To make network studies such as optimal power flow, energy management, and unit commitment more comprehensive, the whole electrical system must be considered in the modeling [53,54]. To overcome the complexity of simultaneous modeling of transmission and distribution systems, researchers have proposed solutions that are applicable to both transmission and distribution systems [55]. In this regard, demand response (DR) was modeled as a problem of co-optimization between distribution and transmission systems in [56]. This model addresses voltage problems in the transmission system by deploying resources in the distribution system. A non-iterative decoupled method for the coordinated robust OPF problem in transmission and distribution networks has been developed in [57] that minimizes the exchanged information between the networks to guarantee robustness. Other approaches for solving the power flow (PF) and optimal power flow (OPF) problems of coupled transmission and distribution systems have also been proposed in the literature [58,59,60,61,62,63,64,65].

Coordinated economic dispatch (ED) of transmission and distribution grids is another problem that researchers tried to solve by means of the heterogeneous decomposition [66], multi-parametric programming [67], analytical target cascading [68], primal-dual gradient algorithm [69], and distributed approximate dynamic programming [70]. In addition, integrated transmission and distribution networks are considered in restoration [71,72], reserve scheduling [73], power system risk assessment [74], voltage stability assessment [75,76], and expansion planning [77].

Co-simulation is another approach that has been proposed in the literature. It has been applied in a wide range of studies including but not limited to electrical machines [78,79], cyber–physical systems [80,81], smart grids [82,83,84,85], energy efficiency of buildings [86,87,88], and communication [89,90,91,92]. Additionally, transmission and distribution systems have been modeled simultaneously under the co-simulation concept in the literature. To this end, the impact of bulk volt/VAR control on the transmission system has been studied in [93] by means of a co-simulation framework with a tight coupling protocol. The authors in [94] have simulated a transmission system using a three-sequence solver in MATLAB. The other part of the co-simulation platform, the distribution network, is then modeled and solved using OpenDSS. In addition, a co-simulation master algorithm is developed in MATLAB to synchronize the simulations by exchanging solutions of the transmission network (bus voltages and angles) and solutions of the distribution grid (active and reactive power flows). To validate the co-simulation results, they have been compared with those of the standalone T&D model in DIgSILENT PowerFactory commercial software. Additionally, a co-simulated platform for transmission-distribution-communication-market framework is introduced in [95] using HELICS and is compared with the other co-simulation frameworks, e.g., High Level Architecture (HLA) and Functional Mock-up Interface (FMI), which demonstrates its superiority. Moreover, dynamic co-simulation of transmission and distribution systems is proposed with a focus on parallel and series computation in [96] and frequency response of DERs in [54]. Other co-simulation platforms have focused on the electricity market challenges [97] and coupling the protocols of the two simulation platforms [98].

After analyzing the available literature in the field of DERMS and energy management, we summarized the available challenges and research gaps in

Table 3. Summary of the research gaps identified from the literature review.

Number	Concluded Gap	Subject Area
1	Application of utility DERMS and aggregator DERMS and their differences	DERMS
2	Neglecting the broad capability of DERMS and its numerous functions	DERMS
3	Neglecting the impact of DERs on transmission networks	Co-simulation of transmission and distribution networks
4	Neglecting DERMS in the simultaneous studies of transmission and distribution networks	DERMS and co-simulation of transmission and distribution networks

. According to this table, it can be concluded that researchers have focused on distribution networks, neglecting the impact of DERs on transmission networks. On the other hand, the co-simulation studies have not considered the functionality of DERMS in coordinating transmission and distribution networks. Considering the large-scale power grids of the future with high penetration of DERs on the distribution side and their interaction with the transmission grid, a coordination method has to be introduced to fill the gaps. Hence, this paper, after a comprehensive review and identifying the gaps, focuses on modeling and implementing a co-simulation platform that considers DERMS functionality in coordinating the transmission and distribution grids. Thus, the consideration of a transmission grid in a DERMS framework is proposed in this paper for the first time.

3. Proposed Method

In this part of the paper, a coordination problem between transmission and distribution grids is solved using a hierarchical approach. The flowchart of the proposed method is presented in Figure 3; following a contingency at the transmission level, e.g., generator or transmission line outage, the maximum deliverable active power is calculated by the transmission system operator (TSO) and sent to the distribution system’s agent: utility DERMS. At the next step, utility DERMS will pass the maximum power setpoint to the aggregator DERMS. We consider TSO to be the highest level in this hierarchical structure. Utility DERMS and aggregator DERMS are at the next levels, respectively. As the lowest level of the structure, DERs communicate with aggregator DERMS to send their available production and receiving set points.

At each iteration, an economic dispatch (ED) problem is solved by the aggregator DERMS with a focus on dispatching the available active power resources. The objective function and related constraints of the ED problem are as follows:

$$\min OF={Cost}_{operation},$$

(1)

$${Cost}_{operation}={Cost}_{i,P}^{DG}+{Cost}_{i,Q}^{DG}+C_i^{Up}+C_i^{Down},$$

(2)

$${Cost}_{i,P}^{DG}=a_i^{DG}\times z_i^{DG}+b_i^{DG}\times P_i^{DG},$$

(3)

$${Cost}_{i,Q}^{DG}=a_i^{DG}\times z_i^{DG}+b_i^{DG}\times Q_i^{DG},$$

(4)

$$C_i^{Up}=\left(z_i^{DG}-z_{i,initial}^{DG}\right)U_i^{DG},$$

(5)

$$C_i^{Down}=\left(z_{i,initial}^{DG}-z_i^{DG}\right)D_i^{DG},$$

(6)

$$C_i^{Up}\ge 0,$$

(7)

$$C_i^{Down}\ge 0,$$

(8)

$$\sum {P^{TR}+P}_i^{DG}+P_i^{PV}+P_i^{WT}={\sum }_{for all loads}{P}_i^{Load},$$

(9)

$$P^{TR}\le P_{max}^{TR},$$

(10)

$$Q^{TR}\le Q_{max}^{TR},$$

(11)

where $OF$ is the objective function of the ED problem and ${Cost}_{operation}$ is the total operation cost of the aggregator DERMS. ${Cost}_{i,P}^{DG}$, ${Cost}_{i,Q}^{DG}$, $C_i^{Up}$, and $C_i^{Down}$ are the costs related to generated active power or $P_i^{DG}$, generated reactive power or $Q_i^{DG}$, startup, and shutdown (for the dispatchable DERs) of the $i^{th}$ unit, respectively. $a_i^{DG}$ and $b_i^{DG}$ are the cost coefficients of the $i^{th}$ unit. $z_i^{DG}$ and $z_{i,initial}^{DG}$ are binary variables indicating the current and initial state of the $i^{th}$ unit, respectively. $U_i^{DG}$ and $D_i^{DG}$ are the startup and shutdown costs of the $i^{th}$ unit, respectively. Moreover, $P^{TR}$ and $Q^{TR}$ are the actual transferred active and reactive powers from the upstream transmission grid. $P_{max}^{TR}$ and $Q_{max}^{TR}$ are the maximum active and reactive powers allowed to be transferred from the transmission network. $P_i^{PV}$ and $P_i^{WT}$ are the output active powers of the $i^{th}$ PV and WT units, respectively. $P_i^{Load}$ is the total demanded active power of the load $i$.

Moreover, the technical constraints related to the limits of the DERs outputs are indicated in the following:

$$z_i^{DG}\times P_{i,min}^{DG}\le P_i^{DG}\le z_i^{DG}\times P_{i,max}^{DG},$$

(12)

$${\left(P_i^{DG}\right)}^2+{\left(Q_i^{DG}\right)}^2\le {\left(S_i^{DG}\right)}^2,$$

(13)

$${PF}_i^{DG}=\frac{P_i^{DG}}{\sqrt{{\left(P_i^{DG}\right)}^2+{\left(Q_i^{DG}\right)}^2}},$$

(14)

$${PF}_{i,min}^{DG}\le {PF}_i^{DG},$$

(15)

where $P_{i,min}^{DG}$ and $P_{i,max}^{DG}$ are the lower and upper limits of the generated active power. $S_i^{DG}$ represents the apparent power of the unit $i$. Additionally, ${PF}_i^{DG}$ indicates the power factor of the $i^{th}$ unit located at the bus $i$. Furthermore, ${PF}_{i,min}^{DG}$ is the minimum allowable power factor for the $i^{th}$ dispatchable unit.

The non-dispatchable PV and WT units have technical equations that are listed in the following:

$$P_i^{PV}\le S_i^{PV},$$

(16)

$$P_{i,max}^{PV}={\eta }_i^{PV} A_i^{PV}{I}^{PV},$$

(17)

$$P_i^{PV}\le P_{i,max}^{PV},$$

(18)

where $S_i^{PV}$ is the maximum capacity of the $i^{th}$ PV unit. $P_{i,max}^{PV}$ represents the maximum available output power of the $i^{th}$ PV unit considering the efficiency of the unit equal to ${\eta }_i^{PV}$, its surface area equal to $A_i^{PV}$, and solar irradiance represented by $I^{PV}$.

The technical constraints for the WT units are as follows:

$$P_i^{WT}\le S_i^{WT},$$

(19)

$$P_i^{WT}=\frac{1}{2}{\rho }_i{A}_i^{WT}{v}_i^3,$$

(20)

where $S_i^{WT}$ is the maximum capacity of the $i^{th}$ WT unit. ${\rho }_i$ and $v_i$ are the air density and wind speed at the location of the $i^{th}$ turbine, respectively. Additionally, $A_i^{WT}$ is the cross-sectional area through which the wind passes for the $i^{th}$ turbine.

The mathematical formulation of the demand response procedure is provided in the following:

$$P_i^{DR}=k_i\times P_i^{Total,DR},$$

(21)

$$P_i^{DR}\ge P_{i,min}^{DR},$$

(22)

$${0\le k}_i\le 1,$$

(23)

where $P_i^{DR}$ is the amount of decreased active power of load $i$ by utilizing the demand response program. $k_i$ is the percentage of reduction for the $i^{th}$ participated load in the DR with the total demanded active power equal to $P_i^{Total,DR}$. Moreover, $P_{i,min}^{DR}$ is the minimum active power that is required to participate in the DR.

The results of the ED are then sent back to the utility DERMS. At this stage, utility DERMS will focus on the voltage level of the network under its supervision and will try to perform a reactive power dispatch using the power flow equations indicated below:

$$P^{TR}+P_i^{PV}+P_i^{DG}+P_i^{WT}-P_i^{Load}-{\sum }_{all lines}{P}_{ij}^{Line}=0,$$

(24)

$$Q^{TR}+Q_i^{DG}-Q_i^{Load}-{\sum }_{all lines}{Q}_{ij}^{Line}=0,$$

(25)

$$P_{ij}^{Line}=G_{ij}^{Line}\left(V_i^2-V_{i}V\mathit{cos}{\theta }_{ij}\right)-B_{ij}^{Line}\left(V_i{V}_j\mathit{cos}{\theta }_{ij}\right),$$

(26)

$$Q_{ij}^{Line}=-B_{ij}^{Line}\left(V_i^2-V_i{V}_j\mathit{cos}{\theta }_{ij}\right)-G_{ij}^{Line}\left(V_i{V}_j\mathit{cos}{\theta }_{ij}\right),$$

(27)

where $P_{ij}^{Line}$ and $Q_{ij}^{Line}$ are the flowed active and reactive powers from node $i$ to node $j$, respectively. $P^{TR}$ and $Q^{TR}$ are the transferred active and reactive powers from the transmission grid, respectively. $P_i^{Load}$ and $Q_i^{Load}$ are the active and reactive power demands of the $i^{th}$ node. $G_{ij}^{Line}$ and $B_{ij}^{Line}$ are the conductance and susceptance of the line connecting node $i$ to node $j$, respectively. Moreover, $V_i$ is the voltage magnitude of the $i^{th}$ node, whereas ${\theta }_{ij}$ is the difference between angles of the voltages of the nodes $i$ and $j$.

Other constraints related to the imported power from the upstream transmission grid and the tap changer equations are depicted in the following:

$$V_{min}\le \left|V_i\right|\le V_{max},$$

(28)

$${\left(P^{TR}\right)}^2+{\left(Q^{TR}\right)}^2\le {{(S}^{TR})}^2,$$

(29)

$$I_{ij}^2={(P}_{ij}^{Line}+P_{ji}^{Line})\frac{{{(G}_{ij}^{Line})}^2+{{(B}_{ij}^{Line})}^2}{G_{ij}^{Line}},$$

(30)

$$I_{ij,t}^2\le I_{ij,max}^2,$$

(31)

where $V_{min}$ and $V_{max}$ are the minimum and maximum acceptable voltages of the nodes, respectively. $S^{TR}$ is the capacity of the substation transformer. $I_{ij}$ is the current flow from node $i$ to node $j$ and $I_{ij,max}$ is the maximum allowable limit for it.

The technical equations related to the tap changer of the substation transformer are as follows:

$$V^{TR}={\sum }_{for all steps}\left(k_m^{Tap}{\gamma }_m^{Tap}\right),$$

(32)

$${\sum }_{for all steps}\left(k_m^{tap}\right)=1,$$

(33)

where $V^{TR}$ is the secondary voltage of the substation transformer. $k_m^{Tap}$ is a binary variable that is equal to 1 for the selected tap and 0 for the others. ${\gamma }_m^{Tap}$ is the voltage level of the $m^{th}$ tap.

Afterward, the final setpoints will be sent back to the aggregator and the problem-solving procedure is finished unless there is at least one bus whose voltage violates the voltage limits. In this situation, the violated bus number(s) will be sent to the aggregator DERMS to repeat the ED problem solving considering the violated bus(es). This procedure continues until the voltages of all buses are in the acceptable range or the maximum number of iterations is reached.

4. Results and Discussion

In this section, the proposed method is applied to the modified IEEE-33 bus test system. The simulated distribution network is connected to the 10th bus in the IEEE-30 bus transmission network, which is considered the upstream transmission network. The schematic of the distribution test system is illustrated in Figure 4. As can be seen, three PV generating units, two wind turbines, and three diesel generators (DGs) are added. The summary of the added generating units is reported in

Table 4. Summary of the added generating units to the original test system.

Type of Generation	Bus Number	Capacity (MVA)	Active Power Limits (MW)		Cost Coefficient
			Min	Max	a	b
PV	8	0.2	0	0.2	-	-
	20	0.3	0	0.3	-	-
	24	0.2	0	0.2	-	-
DG	12	4	0.5	4	28	92
	22	2	0.5	2	31	110
WT	18	3	0	3	-	-
	30	2.1	0	2.1	-	-
DR decreased power	32	-	0	0.63	-	-

. Additionally, the load factor of the simulated network is considered to be 3 to maximize the impact of the proposed method. As a result, the total active and reactive demanded powers of the grid are 11.145 MW and 6.9 MVAR, respectively.

In addition, a tap changer is considered at bus 1 with 7 taps. Moreover, demand response (DR) capability is considered at bus 32 due to the high demand power of this bus.

The schematic of the implemented co-simulation platform is illustrated in Figure 5. As shown in this figure, several tools were used to simulate the proposed method. MATPOWER was utilized to simulate the transmission network. Moreover, the distribution network, considering the utility and aggregator DERMS, was modeled using a mixed-integer linear programming approach by means of the CPLEX solver in GAMS. In order to demonstrate the performance of the proposed method, the following studies were performed.

4.1. Case Study 1

This case study was developed in order to initialize the normal operation condition of the distribution system where there are no contingencies in the transmission network. The outputs of the generating units are reported in

Table 5. Outputs of the generators, DR, and the transferred power for case study 1.

Source of Power	Bus Number	Outputs		Utilization (%)
		P (MW)	Q (MVAR)
Transmission system	1	4.18	6.24	-
PV	8	0.2	0	100
	20	0.3	0	100
	24	0.2	0	100
DG	12	0.54	0.34	15.95
	22	0	0	0
WT	18	3	0	100
	30	2.1	0	100
DR decreased power	32	0	0	0

.

As reported in this table, the renewable units are utilized completely due to their lower cost. Between the two DG units, DG12 is used due to its lower generation cost. Additionally, the remaining demanded power, which is 4.18 MW, is provided from the transmission grid.

4.2. Case Study 2

In this case study, a contingency is considered in the transmission network that enforces a 2 MW limit on the maximum transferred power to the distribution network. As discussed before, this limit is taken into account by the proposed DERMS framework. After eight iterations, the final setpoints are approached as reported in

Table 6. Outputs of the generators and the transferred power for case study 2.

Type of Generation	Bus Number	Outputs		Utilization (%)
		P (MW)	Q (MVAR)
Transmission system	1	1.15	9	-
PV	8	0.2	0	100
	20	0.3	0	100
	24	0.2	0	100
DG	12	2.33	−1.1	64.41
	22	1.63	−1.01	95.85
WT	18	3	0	100
	30	2.1	0	100
DR decreased power	32	0.22	0.11	34.92

.

It can be seen that the transferred active power from the upper grid is lower than the maximum power limit enforced by the upstream network. To make the network operable, the utility DERMS has turned on the DG22, despite its higher generation cost. The voltage levels of the network buses are shown in Figure 6. As is illustrated, all the voltage levels are within the acceptable range (0.95–1.05 p.u.).

As is shown, bus 18 and bus 33 have the highest and lowest voltage levels, respectively. To demonstrate the function of the proposed method, an analysis of the voltage levels of these two buses for each iteration is discussed here.

As illustrated in Figure 7, during the first iteration, an over-voltage is detected at bus 18. Thus, aggregator DERMS tries to overcome the problem by altering the output power of the nearest dispatchable generator, i.e., DG12. Lowering the output of DG12 will also decrease the voltage of bus 33, which is one of the critical buses of the system in terms of low voltage value. It must be mentioned that the DG22 is shut down due to its higher operation cost. To compensate for the lower voltage of bus 33, aggregator DERMS turns on the DG22 and gets the required power from this unit. Subsequently, the operation cost of the network is not the minimum value obtained in the previous iteration and will increase, but on the other hand, the voltage levels are within the acceptable range. The total operation cost of the grid during the simulation is shown in Figure 8. After three iterations, the optimum operating point is not achievable without activating DR. Hence, DR is also activated; consequently, the load power of bus 32 is decreased by %34.92 at the final iteration to achieve the optimum setpoints considering the economic and technical aspects.

A comparison of the bus voltages of the transmission grid and demanded active powers for the two case studies is shown in Figure 9. Due to the active power limits for the studied distribution system (located at bus number 10 in the transmission network), the reactive power demand at bus 10 is increased and as a result, a small drop in voltage levels of this bus and neighboring buses is observed. On the other hand, the enforced active power limit by TSO is satisfied and the normal operation of the transmission network is guaranteed.

5. Conclusions

It is becoming more and more important to employ advanced control and management techniques to deal with the rapid growth of DERs in power systems. DERMS is a newly introduced software solution aimed at enhancing control and management of DERs. Hence, in this paper, a comprehensive review of DERMS as well as energy management strategies were performed. Furthermore, the utility DERMS and aggregator DERMS concepts are investigated and clarified. This review highlights the main applications of DERMS, its objectives, and research gaps. Additionally, previous research on transmission grids has been examined. Consequently, the transmission system and its coordination with DERMS have not been studied in the available literature. Hence, in the next part of this paper, a brief simulation was performed to test the proposed coordination method between the transmission system, utility DERMS, and aggregator DERMS. The co-simulation platform creates a hierarchical structure to manage the interaction between the transmission and distribution networks. By performing simulations, the proposed hierarchical structure is shown to perform well in coordinating the interaction between the two networks.

The future steps in implementing DERMS in real power systems include overcoming some obstacles. Aside from definitions and regulatory frameworks, transmission and distribution networks need to be considered in a more complex environment containing a higher number of DERs with different objectives. DERMS can also be studied in the presence of electricity markets and virtual power plants.