Next Article in Journal
Impact of Time Resolution on Curtailment Losses in Hybrid Wind-Solar PV Plants
Next Article in Special Issue
Distributed Secondary Control for Battery Management in a DC Microgrid
Previous Article in Journal
Machine Learning in Solar Plants Inspection Automation
Previous Article in Special Issue
Surrogate DC Microgrid Models for Optimization of Charging Electric Vehicles under Partial Observability
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

DC Soft Open Points for Resilient and Reconfigurable DC Distribution Networks

by
Husam A. Ramadan
1,2,* and
Spyros Skarvelis-Kazakos
2
1
Electrical Engineering Department, Faculty of Engineering, Minia University, Minia 61519, Egypt
2
School of Engineering and Informatics, University of Sussex, Brighton BN1 9RH, UK
*
Author to whom correspondence should be addressed.
Energies 2022, 15(16), 5967; https://doi.org/10.3390/en15165967
Submission received: 12 June 2022 / Revised: 16 July 2022 / Accepted: 21 July 2022 / Published: 17 August 2022
(This article belongs to the Special Issue Control and Optimization in a DC Microgrid)

Abstract

:
This paper introduces the concept, theory of operation and applications of soft open points for direct current networks (DCSOPs). The DCSOP is based on a bidirectional DC–DC converter actively controlled to behave like a normal conductor. Unlike the normal conductor, the DCSOP can transfer the electric power between nodes at different voltage levels. With this advantage, the DCSOP can effectively control the power flow direction. Thus, DCSOPs can play a vital role in the reconfiguration of DC distribution networks. The operation and control of the DCSOP device was investigated, both in transient and steady-state conditions. Then, a DCSOP was integrated into a DC microgrid model to validate its ability to change the power flow through the modelled feeders. In addition, a set of reliability indicators was calculated for the DC microgrid under different reconfiguration scenarios. It was shown that reliability is improved when the DCSOP device implements network reconfiguration. Finally, an agent-based framework for controlling the DCSOP in a DC microgrid is presented. A fundamental implementation was created for reconfiguring a DC microgrid with a DCSOP controlled by an agent, proving that the agent-based system can effectively control the DCSOP device for reconfiguration and voltage regulation.

1. Introduction

1.1. Need for DC Network Reconfiguration

Network reconfiguration can be achieved with normally open points (NOPs), which are network connections that are deliberately left open, unless they are required. These can be operated manually or using automation, and they can include different isolation mechanisms. The concept of soft open points/soft normally open points (SOPs/SNOPs) is a subset of the NOP where the connection is facilitated by power electronics. The usefulness of SOPs in distribution networks has been demonstrated in the literature. The literature on soft open points and new power electronics technology for improving the resilience of electrical distribution systems in the future is comprehensively reviewed in [1]. They can be implemented as active network management (ANM) tools, helping to keep parameters such as voltage within statutory and operational limits [2]. Consequently, SOPs can also be used for targeted improvements in network operation, such that distributed generation (DG) penetration is maximised [3]. Other benefits that can be gained by using SOPs include improved feeder load balancing and the minimisation of power losses [4,5,6]. The majority of the benefits stem purely from the fact that the networks are reconfigured [7]. However, as shown in [4], existing network reconfiguration techniques on their own were not as effective as they were when combined with the utilisation of SOPs. When SOPs are employed, care must be taken to ensure that feeder automation and protection coordination are not adversely affected, although SOPs can potentially help with identification of faults [8,9]. In AC systems, SOPs can be modelled either as a back-to-back (B2B) voltage source converters (VSC), or as a unified power flow controller (UPFC) [10]. These AC SOP models are characterised by their ability to control reactive power, which is a key consideration in AC network operation.
On the other hand, DC networks do not have reactive power control requirements, but other parameters become more influential, such as voltage levels. The control of multi-terminal DC systems can be more difficult than their AC counterparts, precisely due to the lack of reactive power capability. This becomes especially important during disturbances [11].
DC network studies have been dominated by high-voltage direct current (HVDC) transmission, but hybrid AC/DC systems can also be employed at the medium voltage (MV) distribution level [12]. The benefits of DC distribution, especially at the low voltage (LV) level, have been demonstrated in the literature. One of the key potential benefits is the minimisation of power losses [13], especially at the household level, by avoiding unnecessary AC/DC conversion [14]. Other benefits of using DC distribution systems have been thoroughly researched and demonstrated in the literature [15,16]. In comparison to AC systems, DC distribution systems have lower harmonic content. This benefit is becoming clear with the widespread installation of electric vehicle-charging stations [17,18]. Furthermore, DC distribution systems integrate easily with a wide range of energy storage systems (ESS) and renewable energy sources (RES) [19]. In addition, because of the ease of deployment of energy storage and capacitors, they can provide better dynamic behaviour with the ability to supply pulsed power loads, which improves resiliency and ride-through capability in the event of blackouts and major disturbances in the AC grid. The practical viability of DC networks is a complex aspect, and one of the key considerations is stability, which has different characteristics in AC and DC networks [20].
Multi-terminal SOPs can be employed to help with feeder load balancing in AC systems [12], which indicates the potential for this method to also support DC network stability. The dominant models of AC SOPs require rectification and inversion of the AC waveform through a DC link. In a DC system, this requirement is not there; hence, AC/DC or DC/AC conversion is redundant. Still, there is a need to implement appropriate controls to maintain the benefits of SOPs in a DC system, as opposed to a much simpler automated switch such as a normally open point (NOP).

1.2. Reliability Assessment of Reconfigurable DC Microgrids

The reliability of electricity distribution networks is a very important part of power system planning. A number of methodologies are being used to assess reliability. These mainly relate to reliability indices. The earliest methods have been implemented using probabilistic methodologies, such as Monte Carlo [21]. Test systems have also been put together to aid the development of further, more advanced methods [22]. Reliability assessment is inherently a risk assessment exercise and is based on probabilities of faults and other things that can go wrong. They are a measure of the vulnerability of the power system to potential threats [23,24]. The aim of the reliability assessment is to evaluate the adequacy of the system to supply the load, as well as the security of that supply. To that end, there are numerous reliability indicator metrics that are used in the industry, which are both deterministic and probabilistic. These include, amongst others [23,24]:
  • Loss of load expectation (LOLE),
  • Loss of load probability (LOLP),
  • Expected energy not supplied (EENS),
  • System average interruption frequency index (SAIFI),
  • Customer average interruption frequency index (CAIFI),
  • System average interruption duration index (SAIDI),
  • Customer average interruption duration index (CAIDI).
Other indicators can include cost-based assessment, such as the value of lost load (VOLL) [24]. In that case, the impact of the lack of reliability on the operational cost is quantified.
Reliability methodologies have been used widely in the power industry. In the UK, the Energy Networks Association (ENA) provides a number of documents, the most relevant of which are the Engineering Recommendation P2.7 [25], as well as the associated Engineering Reports 130 and 131 [26,27]. These reliability standards are considered very important for the development of future electricity networks, especially considering the challenges that are presented by the development of power systems and the increasing number of threats, such as climate change and cybersecurity [28].
The use of direct current (DC) networks instead of alternating current (AC) networks for electricity has mostly been employed in high voltage transmission, via high-voltage DC (HVDC) systems. The reliability analysis of such systems can be performed using the same indicators, such as SAIFI, SAIDI, CAIDI and EENS [29].
However, low-voltage distribution networks are more complex and can be analysed locally. The hybrid usage of AC and DC distribution networks has been considered in [30,31], and the benefits have been quantified. The authors of [31,32] also considered the concept of microgrids, which is a local segment of the distribution network, and considered the most reliable topologies. The local generation located in microgrids can potentially be used to improve the adequacy and security of the supply. Another important consideration is that the reliability of power systems depends on the reliability of its constituent components. It has been found that some of the weakest points in a local distribution network with power electronics are the capacitors and the power electronic switches [33]. On the other hand, replacing power electronics more frequently means that the implementation of the solutions is updated in accordance with technological innovations.
Finally, the topology and connections of a small network can significantly affect its reliability. Relevant methods have been developed in the past, in order to assess the reliability of a network based on its structure [34,35]. In such methods, the network is segmented and the reliability indices are calculated based on these segments. This is important when considering the reconfiguration of the network, as has been shown in [36] and [37].

1.3. Agent-Based Control of Reconfigurable DC Microgrids

Multi-agent systems have emerged from the field of computer science and have found widespread application in the area of Smart Grids. Agent-based systems have been proposed for a number of applications in power systems. The first widely recognised attempt to utilise agent-based concepts in microgrids involved the coordination and control of distributed energy resources (DER) [38]. Since then, a series of concepts, approaches and typical applications on power systems has been put forward [39].
The use of multi-agent systems has since extended to the implementation of emissions control in virtual power plants (VPP) [40], as well as the control of electric vehicle (EV) charging, discharging and aggregation [41]. A methodology has also been proposed for developing a multi-agent system that is capable of coordinating resources across multiple energy carriers, such as electricity, gas and heat [42].
However, power systems also require control beyond the coordination of the interconnected energy resources, such as real-time operational parameters. In a microgrid, the control of current, voltage and power flow in real time is vital to its stable operation. Hence, agent-based systems have been proposed for the real-time control of microgrids, also focusing on microgrids where a DC bus is present [43].
Looking at wider system impacts, the use of agent-based technology has been proposed even earlier, in the field of cascading failures and self-healing networks [44]. This approach implies the use of agents who are able to detect an adverse event, assess the vulnerability of the system, and reconfigure and be in charge of the restoration of the system [45]. This is a high-level approach that may lead to an autonomous self-healing network [46].
In terms of DC networks, since these are not widespread, their study is more limited. Still, there have been several propositions for the use of agent-based systems. One of these concerns hybrid medium-voltage AC/DC networks, where the agents are used to support the frequency of the AC system and allow the interconnection of weak and strong grids [47]. Hybrid AC/DC systems with energy storage have also been considered, where a multi-agent system implements an event-driven microgrid control management methodology [48].
Agent-based systems have also been employed in purely DC networks for controlling heterogeneous storage devices, such as super-capacitors, where limited communications were required [49]. Supervisory control was also implemented using agents in isolated microgrids [50]. Several techniques have been employed for the simulated study of such DC microgrids, such as hardware-in-the-loop (HiL) [51]. In the latter case, the benefits of DC microgrids in terms of demand response have also been quantified.
Finally, one key application of multi-agent systems in DC microgrids is for regulating the voltage and contributing to its real-time operation [52,53]. This is especially important because voltage control in DC networks is slightly more complicated than in AC networks due to the lack of reactive power.

1.4. Contributions to Knowledge

In order to address the need for DCSOP control, this research proposes a design for a DCSOP, based on the model of a virtual conductor that has been introduced by the authors in previous research [54,55]. The main contribution of this paper is the definition of the DCSOP as a bidirectional DC/DC power electronics converter that behaves like a normal conductor. It is proposed as a fundamental component for power routing in DC networks, since it can transfer power between nodes at different voltage levels. In turn, it allows connecting two normally open points in the DC distribution networks; hence, the DC network becomes flexible and reconfigurable.
In order to further investigate the value of DC network reconfiguration with DCSOPs, the segmentation methodologies mentioned in Section 1.2 are employed to study the reliability of a DC microgrid under different reconfiguration scenarios. This provides a system-wide benefit assessment of the DCSOP reconfiguration technology.
The third major contribution of this paper is proposing an agent-based framework for controlling the DC network reconfiguration devices based on signals and measurements taken across the network. This extends the benefits of active network management (ANM) to the field of DC networks.

2. Materials and Methods

2.1. DC Soft Open Point (DCSOP)

2.1.1. Implementation of the DCSOP Using a Bi-Directional DC–DC Converter

The DCSOP is based on the virtual conductor (VC) design. This virtual conductor is technically a bidirectional DC–DC converter with a specific control strategy, as shown in Figure 1 and Figure 2. The main target of this control strategy is to keep the voltage difference between the converter terminals constant at a certain value. Since DC power flow can be controlled through voltage differences between two buses, the ability of the DCSOP to control voltage can be utilised to control power flows. Thus, this VC can play a vital role as a DCSOP in DC distribution networks, since it can connect two normally open points as shown in in Figure 3. The DCSOP can be implemented based on two power electronic circuit topologies, isolated or non-isolated, as shown in Figure 4 and Figure 5. Employing the isolated topology can offer the advantage of electrical isolation between the two normally open points. However, the electrical power is still transferred between them via the DCSOP.

2.1.2. Seamless Averaged Model for the DCSOP

The fundamental bi-directional DC–DC circuit configuration of the DCSOP is as shown in Figure 6. It employs a DC–DC converter to connect two different load/supply units. This model has been implemented in the case study simulations in Section 3 below. The voltage difference between the converter sides is controlled by the duty ratio of the main switch ( S a ) and of the complementary switch ( S b ). The two bidirectional DC current sources, I 1 and I 2 , have internal resistances, r i and r o , respectively. There are three storage components:   C 1 , C 2 and L .
The inductor parasitic resistance r l and the MOSFET turn-on resistance r s are included in the model. The transfer function for this converter has been derived based on the state-space averaging method. Considering the state-space vector ( t ) = [ v c 1 ( t ) v c 2 ( t ) i L ( t ) ] , and the input vector u = [ I 1   I 2 ] ; the state-space averaged DC model is shown in (1):
0 = A [ I L V 1 V 2 ] + B [ I 1 I 2 ]
where:
A = [ r s + r l L D L 1 L D C 1 1 r i C 1     0 1 C 2 0 1 r o C 2 ] ,   B = [ 0 0 1 C 1 0 0 1 C 2 ]
By solving (1), the following expressions are given:
I L = r i D I 1 + r o I 2 r i D 2 + r o + r l + r s
V 1 = r i I 1 ( r o + r l + r s ) D r o r i I 2 r i D 2 + r o + r l + r s
V 2 = D r o r i ( I 1 D I 2 ) r o ( r l + r s ) I 2 r i D 2 + r o + r l + r s
where D is the duty cycle.
The state-space averaged AC model is shown in (5).
d d t | i ^ L v ^ 1 v ^ 2 | = A · | i ^ L v ^ 1 v ^ 2 | + C · | I L V 1 V 2 | · d ^
where:
A = [ r s + r l L D L 1 L D C 1 1 r i C 1     0 1 C 2 0 1 r o C 2 ]   ,   C = [ 0 1 L 0 1 C 1 0     0 0 0 0 ]
By solving (5), the following Equations (6)–(8) are obtained:
v ^ 1 d ^ = D C 1 · i ^ L d ^ 1 C 1 · I L s + 1 C 1 · r i
v ^ 2 d ^ = 1 C 2 s + 1 C 2 · r o   · i ^ L d ^
i ^ L d ^ = ( s + 1 C 1 · r i ) · ( s + 1 C 2 · r 0 ) V 1 L D · I L C 1 L · ( s + 1 C 2 · r o ) ( s + r l + r s L ) · ( s + 1 C 1 · r i ) · ( s + 1 C 2 · r o ) + D 2 ( s + 1 C 2 · r o ) L · C 1 + ( s + 1 C 1 · r i ) L · C 2
Using Equations (6)–(8), the control-to-voltage difference transfer function G d v ( s )   can be obtained as in Equation (9), where v ^ = v ^ 1 v ^ 2 .
G d v ( s ) = v ^ 1 v ^ 2 d ^ = [ D C 1 ( s + 1 C 1 r i ) 1 C 2 ( s + 1 C 2 r o ) ] · i ^ L d ^ I L C 1 ( s + 1 C 1 r i )

2.2. Reliability Assessment Method

2.2.1. Reliability Analysis Sets

This work applies a reliability assessment method that was developed by the authors of [34] and also used in [35], in order to study the effect of DC network reconfiguration using DCSOP on reliability. Reliability indices have been examined under different reconfiguration scenarios to quantitatively evaluate the impact of these scenarios.
The reliability assessment methodology is based on the concept of segmentation, where the power system is split into segments. These segments are assumed to be representative of protection systems and their potential to disconnect and isolate sections of the power system during fault conditions.
The method is simple and straightforward, and does not require specialised computer models to be created. In the case of the small network that this paper describes, the calculations are simple enough to be performed manually.
This section describes the sets required to analyse the reliability of a certain load node. The links between these sets are described in Figure 7.
The “S” segment is defined as the load node that is interrupted.
The “L” set contains all segments of the network, which may cause loss of power to segment “S” if they fail. The “L” set can be divided into subsets “SSL” and “NSSL”.
The “SSL” set is defined with respect to a continuous power route between segment “S” and the power source node. It comprises those segments that can be isolated from the above route. In contrast, the “NSSL” set contains these segments that cannot be isolated, i.e., that are integral to the power route between segment “S” and the source. “NSSL” can also be divided into subsets “SL” and “NSL”.
The “SL” set contains components that are not integral to supply of segment “S”, which can be supplied in another way in case of a disconnection in the “SL” subset. Conversely, the “NSL” subset contains components that are integral to the supply of segment “S” and should not be removed. “SL” is further divided into the “SAF” and “NSAF” subsets.
The “SAF” and “NSAF” subsets are very similar to “SL” and “NSL”, although in this case, they refer to the failure of components rather than their removal, and the temporary restoration of supply to “S”.
Finally, the “SAF” set is divided into subsets “SF and “NSF”. In this case, these sets refer to the restoration of power to segment “S” without violating the system constraints.
Another important parameter that is defined is the downtime for segment “S”, referred to as D T s , and given by Equation (10) [35]:
D T s = i ϵ NSL NSAF NSS FR i   REP i   + i ϵ SSL SF FR i   S O T i  
where:
FR i : failure rate for segment “i”.
REP i   : repair time for segment “i”.
S O T i   : switch operation time to re-supply segment “S” because of the failure of segment “i”.

2.2.2. Reliability Indices

The above downtime calculation was used in the reliability indices below as a measure of the “sum of customer interruption durations”. The following two reliability indices have been evaluated during this study [23,24]:
  • SAIDI (system average interruption period index) is the average interruption period per customer served.
    SAIDI = s u m   o f   c u s t o m e r   i n t e r r u p t i o n   d u r a t i o n s t o t a l   n u m b e r   o f   c u s t o m e r s
  • CAIDI (customer average interruption period index) is the average interruption period for those customers interrupted during a year.
    CAIDI = s u m   o f   c u s t o m e r   i n t e r r u p t i o n   d u r a t i o n s t o t a l   n u m b e r   o f   c u s t o m e r   i n t e r r u p t i o n s

2.3. Agent-Based Model

This section describes the agent-based model that was developed for the purpose of proving the concept of agent-based control of DCSOP within a DC microgrid. The types of agents considered were the microgrid/network agent, the DCSOP agent and the generator/load agent. These are described below.

2.3.1. Microgrid/Network Agent

The microgrid agent is defined as the overarching agent, which has the following capabilities:
  • Calculate power flows and operational parameters based on the different measurement points in the microgrid/network.
  • Communicate with DCSOP agents and other agents, and provide information and knowledge gained from overseeing the operation of the network.
  • Send direct requests to particular equipment agents to act in specific ways.
  • Communicate with other DCSOP, load, power flow or other network agents in order to establish joint decision making.
  • Communicate with other, adjacent microgrid agents, in case there is any synergy that can be gained, such as load reduction to prevent load shedding.

2.3.2. DCSOP Agent

The DCSOP agent is defined as an agent that has the following capabilities:
  • Communicates with other agents (e.g., load agents) and receives messages from them on various conditions in the network.
  • Makes decisions regarding the state of the network and whether there is a need to act or not.
  • Communicates with other DCSOP, load, power flow or other network agents in order to establish joint decision making.
  • Sends commands to the DCSOP device, enabling it to transition from the ON to the OFF state, and vice versa.
  • Receives measurements on different parameters from the DCSOP in order to check on the state of the device.

2.3.3. Generator/Load Agent

The generator/load agent is defined as an agent that has the following capabilities:
  • Communicates with other agents (e.g., DCSOP or microgrid agents) and sends messages to them regarding various parameters associated with their generator or load.
  • Makes decisions regarding the local state of the network and whether there is a need to act or not (e.g., generator disconnection, load shedding).
  • Communicates with other DCSOP, load, power flow or other network agents in order to establish joint decision making.

2.3.4. Multi-Agent System Structure

The structure of the multi-agent system including the agents described above is shown in Figure 8.

2.3.5. Agent Implementation in MATLAB/Simulink Stateflow

The DCSOP and load agents were implemented in the Stateflow programming environment of MATLAB/Simulink. Screenshots of the implementations that were tested in this work are shown in Figure 9. The implementation was minimal and enabled us to test the concept of controlling the DCSOP within the context of a DC microgrid.
The load agent is able to measure the level of the load at a particular point and decide on the classification (low/high). It then transmits the classification to the DCSOP agent.
The DCSOP agent receives the classification of the load from the load agent. If the load is classified as low, the DCSOP agent decides to transition the DCSOP device to OFF, i.e., to open the circuit. If the load is classified as high, the DCSOP agent decides to transition the DCSOP device to the ON state, i.e., to close the circuit.

3. Results and Discussion

3.1. DCSOP Characterisation Case Studies

In order to investigate the performance of the DCSOP, four cases were analysed in MATLAB/Simulink, as shown in Table 1. The first case was the stress test, in which the DCSOP was tested for a very stiff change in the direction of the power flow to ensure a rigid and reliable controller performance. The second case was to test the ability of the DCSOP to control the direction of the current/power flow through the network. In the third case, DCSOP was integrated into a DC microgrid to show its effect on the power distribution through the microgrid. The last case replaced the NOP in the microgrid of the previous case with a short circuit in order to compare its effect with that of the DCSOP.

3.2. DCSOP Stress Test

In this test, the reliability of the DCSOP controller was investigated. Thus, a repeated and a very aggressive step change in the power flow direction was applied to the DCSOP configuration shown in Figure 10, as shown in Figure 11. The results in Figure 12 show that the controller succeeded in keeping the voltage difference between the DCSOP terminals at a constant value (50 V) in spite of the change in power flow direction.

3.3. Control of Current Flow Direction

The purpose of the DCSOP controller is to keep the voltage difference between the DCSOP terminals constant at a certain value. Therefore, the voltage at one terminal of the DCSOP will be higher than the other terminal by a particular value. In turn, the current will flow from the terminal with the higher voltage to the other terminal. The current flow direction can thus be controlled. In order to test this functionality, a circuit configuration, as shown in Figure 13, was implemented and the DCSOP connected between Buses 2 and 5. Once the DCSOP caused the voltage at Bus 5 to be higher than that of Bus 2, the current flowed in the direction shown by the red arrow (from Bus 5 to Bus 2). When the DCSOP controlled the voltage at Bus 2 to be higher than that at Bus 5, the current flowed in the opposite direction. The results in Figure 14 confirm this advantage of including the DCSOP in a DC distribution system.

3.4. Replacing NOP with DCSOP in a DC Microgrid

The performance of the DCSOP was investigated through integrating it into a DC microgrid. This DC microgrid model, shown in Figure 15, was constructed based on the CERTS AC microgrid described in [56], with the key difference that it was converted to DC. It consists of four feeders, ten buses, seven loads, five generators and two NOPs. The DCSOP was tested by connecting it to NOP1. Table A1 includes the data of this DC microgrid. The results, in Figure 16 and Figure 17, as well as Table 2 and Table 3, show the currents through the feeders and from the generators both before and after connecting the DCSOP. It can be seen that these currents were severely altered by connecting the DCSOP, and this means that the DCSOP can have a strong effect when employed in the reconfiguration of DC distribution systems. In particular, it can be observed that the current across Buses 2 and 6 was dramatically reduced, avoiding potential overloading in that line. Likewise, the output of generators 3 and 4 was distributed more equally.

3.5. Replacing NOP with a Normal Conductor in a DC Microgrid

In order to compare the performance of the DCSOP with that of a normal conductor, the DC microgrid model was modified by closing NOP1 with a normal conductor instead of a DCSOP. The results in Figure 18 and in Figure 19 show the similarity of the performance for the DCSOP and the normal conductor, except for a slight difference in some branch currents. Table 4 and Table 5 show a comparison between the performance of the DCSOP and the normal conductor in terms of the change of branch and generator currents before and after the connection. Based on this comparison, it can be concluded that the DCSOP can behave like a normal conductor in connecting the NOPs. However, it has additional merits, such as controlling the current direction and connecting two nodes with different voltage levels.

3.6. Impact of DCSOP on Reliability of Reconfigurable DC Microgrids—Case Studies

Four reliability study cases were investigated, and these are described in this section. The network used in these study cases is the microgrid defined by CERTS, as seen in [56]. The microgrid model in [56] is an AC microgrid, whereas in this study the model was converted to a DC microgrid, considering the same or equivalent component ratings. The component failure rates were considered to be uniform, as they are very hard to estimate due to the lack of data on DC components. In any case, they are of minor importance, since the key assessment in this study relates to the network structure of the microgrid rather than the individual components.
The microgrid was reconfigured in three different ways, by linking two pairs of buses in all combinations. The link was performed with a DCSOP, which the authors will be describing in a different publication. The essence of this DCSOP is that it acts as a controllable power electronics switch that can be operated remotely, and that can link two DC buses in different parts of the network. Figure 20 shows the network, including the locations that were considered for the installation of the DCSOPs. The four cases and their results are described in the following subsections.

3.6.1. Case R.1: Base Case, No DCSOP

In this case, the reliability analysis was conducted for the original DC system without considering any DCSOPs in either Location A or B.
The two reliability indices were calculated as follows:
  • CAIDI: 1.269,
  • SAIDI: 1.0497.

3.6.2. Case R.2: One DCSOP, Location A

In this case, the reliability analysis was conducted for the original DC system with the addition of one DSCOP between Bus 4 and Bus 6 (Location A in Figure 20). It can be observed that both indices dropped, particularly CAIDI.
The two reliability indices were calculated as follows:
  • CAIDI: 0.961,
  • SAIDI: 0.964.

3.6.3. Case R.3: One DCSOP, Location B

In this case, the reliability analysis was conducted for the original DC system with the addition of one DSCOP between Bus 9 and Bus 6 (Location B in Figure 20). This type of reconfiguration seems to be more effective, as the SAIDI index dropped further. The CAIDI index is almost the same as Case R.2, although marginally higher.
The two reliability indices were calculated as follows:
  • CAIDI: 0.964,
  • SAIDI: 0.7461.

3.6.4. Case R.4: Two DCSOPs, Locations A and B

In this case, the reliability analysis was conducted for the original DC system with the addition of two DSCOPs. This reconfiguration case yielded the most significant improvements, since both indices dropped significantly. The CAIDI index in particular is almost half that of the original Case R.1.
The two reliability indices were calculated as follows:
  • CAIDI: 0.658,
  • SAIDI: 0.656.

3.6.5. Summary and Comparison of Reliability Cases

A comparison of all four cases is presented in Table 6. The percentage improvement against the base case (Case R.1) is also provided. It can be seen that Cases R.2–R.4 present a significant reduction in both indices.

3.7. Testing of Agent-Based Control of DCSOP for DC Network Reconfiguration

The sequence diagram of the interaction between the two agents can be seen in Figure 21a. The interactions demonstrate that when the load agent decides that the load is low, then it signals to the DCSOP agent to transition to the OFF state. Hence, in the next iteration, the DCSOP agent transitions to the OFF state. Conversely, when the load agent decides that the load is high, it signals to the DCSOP agent to transition to the ON state. Hence, in the next iteration, the DCSOP agent transitions to the ON state.
The measurements from the DC microgrid can be seen in Figure 21b. It can be seen that the DCSOP transitions between the ON and OFF state several times throughout the simulation. The current that passes through the DCSOP is non-zero when it is in the ON state and zero when it is in the OFF state. Its value when it is in the ON state depends on the power flows in the circuit, hence the precise value of the load.
The switching of the DCSOP caused significant transient effects, but the transient performance of the system was not considered in this particular study. One important observation is that the steady-state voltage across the load remains relatively stable, despite the fluctuations of the load. This can be explained by the fact that when the load is high, the DCSOP is on; hence, the power is being routed towards that load from more than one source. Voltage transients can also be observed, which are most likely due to the switching operation of the DCSOP and the reactive elements of the circuit.
The above demonstrates that a DCSOP controlled by agent-based systems can be used for voltage regulation in a DC microgrid.

4. Conclusions

This paper presented the novel concept of the DC soft open point as a tool for reconfiguring DC distribution networks. The model of the DCSOP was presented, along with controller schematics and a state space analysis. Four characterisation case studies were simulated to evaluate the operational performance of the DCSOP within a DC microgrid. The results show that the proposed DCSOP concept can be used to effectively reconfigure a DC network. In some cases, the power flow distribution was altered with beneficial reductions in the loading of some lines. With careful coordination, such a reconfiguration can prevent the overloading of DC lines and other equipment, and enable the optimal utilisation of assets. It has also been proven that the DCSOP offers additional functionality and control to the system operator, by:
  • Being able to control the power flows, and
  • Being able to connect two buses operating at different voltages.
In addition, the effectiveness of the DC network reconfiguration for improving reliability was evaluated. An established method for reliability evaluation based on segmenting power networks was used. Four reliability case studies were set up. The first one was a DC conversion of the CERTS microgrid testbed, without any reconfiguration. The remaining case studies considered reconfiguring the DC microgrid with a DC soft open point. Two reconfiguration points were considered and the scenarios involved all combinations. The reliability indices, CAIDI and SAIDI, were calculated for all cases. It was found that the most effective way of improving reliability is to reconfigure the network with two DCSOPs. In that case, the reliability indices were almost halved. The reason for this is likely that there are multiple alternative routes for the power to reach the load in the case of a failure in the network.
Finally, an agent-based framework was presented for the control of a DCSOP power electronics device by an intelligent agent. The purpose of this framework was to enable DC microgrid control and reconfiguration. The framework was implemented in MATLAB/Simulink and a test case presented, using an AC microgrid converted to DC. One DCSOP device was placed between two buses of the microgrid and an agent was attached to it. The operation of this fundamental agent-based system was demonstrated, and this will be used as the basis for further development of more elaborate agent-based systems. The results show that the DCSOP transitions between ON and OFF states, as instructed by the agent, based on the load levels at a particular bus. The outcome of that is that the steady-state voltage of that particular load does not drop unnecessarily when the load increases. This is due to the fact that by turning the DCSOP device on, alternative routes are provided for the power to reach that load. Future extensions of this work include the study of the effects of power electronics converters and their coordination with the existing protection devices. In addition, multi-objective analysis can be conducted with the other power quality factors, such as harmonic distortions and efficiency.

Author Contributions

Conceptualization H.A.R. and S.S.-K.; methodology H.A.R.; software, H.A.R.; validation, H.A.R. and S.S.-K.; formal analysis, H.A.R. and S.S.-K.; investigation, H.A.R. and S.S.-K.; resources, H.A.R. and S.S.-K.; writing—original draft preparation, H.A.R. and S.S.-K.; writing—review and editing, H.A.R. and S.S.-K.; visualization, H.A.R. and S.S.-K.; funding acquisition, S.S.-K. and H.A.R. All authors have read and agreed to the published version of the manuscript.

Funding

This project received funding from the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement No. 796743.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Appendix A

Table A1. Data of the simulated DC microgrid.
Table A1. Data of the simulated DC microgrid.
BranchesR/ΩLoadsRating/KWGeneratorsRating/V
Bus 1–Bus 21.32Load 1100G1400
Bus 2–Bus 31.32Load 260G2400
Bus 3–Bus 41.32Load 360G3400
Bus 4–Bus 51.32Load 460G4400
Bus 2–Bus 61.32Load 530G5400
Bus 6–Bus 71.32Load 630
Bus 2–Bus 81.32Load 760
Bus 8–Bus 91.32
Bus 8–Bus 101.32

References

  1. Fuad, K.S.; Hafezi, H.; Kauhaniemi, K.; Laaksonen, H. Soft Open Point in Distribution Networks. IEEE Access 2020, 8, 210550–210565. [Google Scholar] [CrossRef]
  2. Giannelos, S.; Konstantelos, I.; Strbac, G. Option value of Soft Open Points in distribution networks. In Proceedings of the 2015 IEEE Eindhoven PowerTech, Eindhoven, The Netherlands, 29 June–2 July 2015; 2015; pp. 1–6. [Google Scholar] [CrossRef]
  3. Bloemink, J.M.; Green, T.C. Increasing distributed generation penetration using soft normally-open points. In Proceedings of the IEEE PES General Meeting, Providence, RI, USA, 25–29 July 2010; pp. 1–8. [Google Scholar]
  4. Cao, W.; Wu, J.; Jenkins, N.; Wang, C.; Green, T. Benefits analysis of Soft Open Points for electrical distribution network operation. Appl. Energy 2016, 165, 36–47. [Google Scholar] [CrossRef]
  5. Li, P.; Ji, H.; Yu, H.; Zhao, J.; Wang, C.; Song, G.; Wu, J. Combined decentralized and local voltage control strategy of soft open points in active distribution networks. Appl. Energy 2019, 241, 613–624. [Google Scholar] [CrossRef]
  6. Ismail, M.; Hassane, E.; Hassan, E.M.; Tijani, L. Power losses minimization in distribution system using soft open point. In Proceedings of the 2020 1st International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET), Meknes, Morocco, 16–19 April 2020; pp. 1–5. [Google Scholar]
  7. Yang, Y.; Zhang, S.; Pei, W.; Sun, J.; Lu, Y. Network reconfiguration and operation optimisation of distribution system with flexible DC device. J. Eng. 2019, 2019, 2401–2404. [Google Scholar] [CrossRef]
  8. Aithal, A.; Long, C.; Cao, W.; Wu, J.; Ugalde-Loo, C.E. Impact of soft open point on feeder automation. In Proceedings of the 2016 IEEE International Energy Conference (Energycon), Leuven, Belgium, 4–8 April 2016; pp. 1–6. [Google Scholar]
  9. Aithal, A.; Li, G.; Wu, J.; Yu, J. Performance of an electrical distribution network with Soft Open Point during a grid side AC fault. Appl. Energy 2018, 227, 262–272. [Google Scholar] [CrossRef]
  10. Cao, W.; Wu, J.; Jenkins, N. Feeder load balancing in MV distribution networks using soft normally-open points. In Proceedings of the IEEE PES Innovative Smart Grid Technologies, Europe, Istanbul, Turkey, 12–15 October 2014; pp. 1–6. [Google Scholar]
  11. Olowookere, O.; Skarvelis-Kazakos, S.; Habtay, Y.; Woodhead, S. Fault ride through during loss of converter in a 4-VSC based HVDC transmission. In Proceedings of the IEEE PES Innovative Smart Grid Technologies Conference Europe, Istanbul, Turkey, 12–15 October 2014; pp. 1–6. [Google Scholar]
  12. Ji, H.; Wang, C.; Li, P.; Zhao, J.; Song, G.; Ding, F.; Wu, J. An enhanced SOCP-based method for feeder load balancing using the multi-terminal soft open point in active distribution networks. Appl. Energy 2017, 208, 986–995. [Google Scholar] [CrossRef]
  13. Bloemink, J. Distribution Level Power Electronics: Soft Open-Points (SOPs). Ph.D. Thesis, Imperial College, London, UK, 2013. [Google Scholar]
  14. Daniel, S.; Skarvelis-Kazakos, S.; Jain, P. Local smart DC networks and distributed storage for reducing and shifting peak load. In Proceedings of the 22nd International Conference on Electricity Distribution (CIRED), Stockholm, Sweden, 10–13 June 2013. [Google Scholar]
  15. Gopinathan, S.; Rao, V.S.; Kumaravel, S. Family of Non-Isolated Quadratic High Gain DC-DC Converters based on Extended Capacitor-Diode Network for Renewable Energy Source Integration. IEEE J. Emerg. Sel. Top. Power Electron. 2022. [Google Scholar] [CrossRef]
  16. Iovine, A.; Carrizosa, M.J.; De Santis, E.; Di Benedetto, M.D.; Pepe, P.; Sangiovanni-Vincentelli, A. Voltage Regulation and Current Sharing in DC Microgrids with Different Information Scenarios. IEEE Trans. Control Syst. Technol. 2021. [Google Scholar] [CrossRef]
  17. Heydari-doostabad, H.; O’Donnell, T. A Wide-Range High-Voltage-Gain Bidirectional DC–DC Converter for V2G and G2V Hybrid EV Charger. IEEE Trans. Ind. Electron. 2022, 69, 4718–4729. [Google Scholar] [CrossRef]
  18. He, P.; Khaligh, A. Comprehensive Analyses and Comparison of 1 kW Isolated DC–DC Converters for Bidirectional EV Charging Systems. IEEE Trans. Transp. Electrif. 2017, 3, 147–156. [Google Scholar] [CrossRef]
  19. Rahimi, R.; Habibi, S.; Ferdowsi, M.; Shamsi, P. Z-Source-Based High Step-Up DC-DC Converters for Photovoltaic Applications. IEEE J. Emerg. Sel. Top. Power Electron. 2021. [Google Scholar] [CrossRef]
  20. Kwasinski, A.; Onwuchekwa, C.N. Dynamic behavior and stabilization of DC microgrids with instantaneous constant-power loads. IEEE Trans. Power Electron. 2011, 26, 822–834. [Google Scholar] [CrossRef]
  21. Billinton, R.; Li, W. Reliability Assessment of Electric Power Systems Using Monte Carlo Methods; Plenum Press: New York, NY, USA, 1994. [Google Scholar]
  22. Billinton, R.; Jonnavithula, S. A test system for teaching overall power system reliability assessment. IEEE Trans. Power Syst. 1996, 11, 1670–1676. [Google Scholar] [CrossRef]
  23. Kjølle, G.; Jakobsen, S.H.; Baldursson, F.M.; Galant, S.; Haarla, L. State of the Art on Reliability Assessment in Power Systems Including Socio-Economic Impact. GARPUR Deliverable D1.1. 2014. Available online: https://www.sintef.no/globalassets/project/garpur/deliverables/garpur-d1.1-state-of-the-art-on-reliability-assessment-in-power-systems.pdf (accessed on 1 November 2021).
  24. Heylen, E.; Deconinck, G.; Hertem, D.V. Review and classification of reliability indicators for power systems with a high share of renewable energy sources. Renew. Sustain. Energy Rev. 2018, 97, 554–568. [Google Scholar] [CrossRef]
  25. Energy Networks Association. Engineering Recommendation; Energy Networks Association: London, UK, 2019. [Google Scholar]
  26. Energy Networks Association. Engineering Report 130, Application Guide for Assessing the Capacity of Networks Containing Distributed Generation; Energy Networks Association: London, UK, 2014. [Google Scholar]
  27. Energy Networks Association. Engineering Report 131, Analysis Package for Assessing Generation Security Capability—Users’ Guide; Energy Networks Association: London, UK, 2012. [Google Scholar]
  28. Strbac, G.; Kirschen, D.; Moreno, R. Reliability standards for the operation and planning of future electricity networks. Found. Trends Electr. Energy Syst. 2016, 1, 143–219. [Google Scholar] [CrossRef]
  29. An, W.; Mou, M.; Wei, C.Z.; Liu, K.; Deng, S.R.; Zhao, W.F. Reliability evaluation and comparison for different topologies of VSC-HVDC distribution networks using analytical and simulation methods. In Proceedings of the 12th IET International Conference on AC and DC Power Transmission (ACDC 2016), Beijing, China, 28–29 May 2016; pp. 1–6. [Google Scholar]
  30. Moreno, R.; Chen, Y.; Strbac, G. Evaluation of benefits of coordinated DC & AC flexible transmission systems with probabilistic security and corrective control. In Proceedings of the IET International Conference on Resilience of Transmission and Distribution Networks (RTDN), Birmingham, UK, 22–24 September 2015; pp. 1–6. [Google Scholar]
  31. Sabouhi, H.; Doroudi, A.; Fotuhi-Firuzabad, M.; Bashiri, M. Hybrid AC/DC microgrids flexible reliability index by using the axiomatic design concept. IET Gener. Transm. Distrib. 2020, 14, 5456–5462. [Google Scholar] [CrossRef]
  32. Nguyen, M.Y.; Yoon, Y.T. A Comparison of microgrid topologies considering both market operations and reliability. Electr. Power Compon. Syst. 2014, 42, 585–594. [Google Scholar] [CrossRef]
  33. Kaipia, T.; Peltoniemi, P.; Lassila, J.; Salonen, P.; Partanen, J. Power electronics in smartgrids—Impact on power system reliability. In Proceedings of the CIRED Seminar 2008: SmartGrids for Distribution, Frankfurt, Germany, 23–24 June 2008; pp. 1–4. [Google Scholar]
  34. Broadwater, R.P.; Shaalan, H.E.; Oka, A.; Lee, R.E. Distribution system reliability and restoration analysis. Electr. Power Syst. 1994, 29, 203–211. [Google Scholar] [CrossRef]
  35. Zhu, D. Electric Distribution Reliability Analysis Considering Time-Varying Load, Weather Conditions and Reconfiguration with Distributed Generation. Ph.D. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA, 2007. [Google Scholar]
  36. Khraiwish, A.S. Effect of distributed generation and feeder sectioning devices on the reliability of the electric distribution systems. Dirasat Eng. Sci. 2008, 35, 26–34. [Google Scholar]
  37. Rouina, A.; Aboubou, A.; Abdelaziz, A.; Elhabachi, A. A novel assessor for distribution system reliability. Courr. Du Savoir N 2012, 13, 109–115. [Google Scholar]
  38. Dimeas, A.L.; Hatziargyriou, N.D. Agent based control for microgrids. In Proceedings of the 2007 IEEE Power Engineering Society General Meeting, Tampa, FL, USA, 24–28 June 2007; pp. 1–5. [Google Scholar]
  39. McArthur, S.D.; Davidson, E.M.; Catterson, V.M.; Dimeas, A.L.; Hatziargyriou, N.D.; Ponci, F.; Funabashi, T. Multi-agent systems for power engineering applications—part I: Concepts, approaches, and technical challenges. IEEE Trans. Power Syst. 2007, 22, 1743–1752. [Google Scholar] [CrossRef]
  40. Kazakos, S.S.; Rikos, E.; Kolentini, E.; Cipcigan, L.M.; Jenkins, N. Implementing agent-based emissions trading for controlling virtual power plant emissions. Electr. Power Syst. Res. 2013, 102, 1–7. [Google Scholar] [CrossRef]
  41. Unda, I.G.; Papadopoulos, P.; Kazakos, S.S.; Cipcigan, L.M.; Jenkins, N.; Zabala, E. Management of electric vehicle battery charging in distribution networks with multi-agent systems. Electr. Power Syst. Res. 2014, 110, 172–179. [Google Scholar] [CrossRef]
  42. Kazakos, S.S.; Papadopoulos, P.; Unda, I.G.; Gorman, T.; Belaidi, H.; Zigan, S. Multiple energy carrier optimisation with intelligent agents. Appl. Energy 2016, 167, 323–335. [Google Scholar] [CrossRef]
  43. Aung, H.N.; Khambadkone, A.M.; Srinivasan, D.; Logenthiran, T. Agent-based intelligent control for real-time operation of a microgrid. In Proceedings of the 2010 Joint International Conference on Power Electronics, Drives and Energy Systems & 2010 Power, New Delhi, India, 20–23 October 2010; pp. 1–6. [Google Scholar]
  44. Liu, H.; Chen, X.; Yu, K.; Hou, Y. The control and analysis of self-healing urban power grid. IEEE Trans. Smart Grid 2012, 3, 1119–1129. [Google Scholar] [CrossRef]
  45. Amin, M. Toward self-healing energy infrastructure systems. IEEE Comput. Appl. Power 2001, 14, 20–28. [Google Scholar] [CrossRef]
  46. Jung, J.; Liu, C.C. Multi-agent technology for vulnerability assessment and control. In Proceedings of the 2001 Power Engineering Society Summer Meeting, Vancouver, BC, Canada, 15–19 July 2001; Volume 2, pp. 1287–1292. [Google Scholar]
  47. Musa, A.; Sabug, L.; Ponci, F.; Monti, A. Multi-agent based intelligent frequency control in multi-terminal dc grid-based hybrid ac/dc networks. IET Renew. Power Gener. 2018, 12, 1434–1443. [Google Scholar] [CrossRef]
  48. Khatibzadeh, A.; Besmi, M.; Mahabadi, A.; Haghifam, M.R. Multi-agent-based controller for voltage enhancement in AC/DC hybrid microgrid using energy storages. Energies 2017, 10, 169. [Google Scholar] [CrossRef]
  49. Morstyn, T.; Hredzak, B.; Agelidis, V.G. Cooperative multi-agent control of heterogeneous storage devices distributed in a DC microgrid. IEEE Trans. Power Syst. 2016, 31, 2974–2986. [Google Scholar] [CrossRef]
  50. Hamad, A.A.; El-Saadany, E.F. Multi-agent supervisory control for optimal economic dispatch in DC microgrids. Sustain. Cities Soc. 2016, 27, 129–136. [Google Scholar] [CrossRef]
  51. Yoo, C.; Choi, W.; Chung, I.; Won, D.; Hong, S.; Jang, B. Hardware-in-the-loop simulation of DC microgrid with multi-agent system for emergency demand response. In Proceedings of the 2012 IEEE Power and Energy Society General Meeting, San Diego, CA, USA, 15 December 2012; pp. 1–6. [Google Scholar]
  52. Zhang, R.; Hredzak, B. Distributed finite-time multiagent control for DC microgrids with time delays. IEEE Trans. Smart Grid 2019, 10, 2692–2701. [Google Scholar] [CrossRef]
  53. Hamad, A.A.; Farag, H.E.; El-Saadany, E.F. A novel multiagent control scheme for voltage regulation in DC distribution systems. IEEE Trans. Sustain. Energy 2015, 6, 534–545. [Google Scholar] [CrossRef]
  54. Ramadan, H.A.; Imamura, Y.; Yang, S.; Shoyama, M. Performance of a virtual conductor in DC power systems using a controlled bidirectional DC-DC converter. Int. J. Innov. Res. Electr. Electron. Instrum. Control. Eng. 2014, 2, 2002–2008. [Google Scholar]
  55. Ramadan, H.A. Modeling and Control of Bidirectional DC-DC Converters for DC Power Systems with Renewable Energy. Ph.D. Thesis, Kyushu University, Fufuoka, Japan, 2015. [Google Scholar]
  56. Lasseter, R.H.; Eto, J.H.; Schenkman, B.; Stevens, J.; Vollkommer, H.; Klapp, D.; Linton, E.; Hurtado, H.; Roy, J. CERTS Microgrid laboratory test bed. IEEE Trans. Power Deliv. 2011, 26, 325–332. [Google Scholar] [CrossRef]
Figure 1. DCSOP based the virtual conductor.
Figure 1. DCSOP based the virtual conductor.
Energies 15 05967 g001
Figure 2. Control block diagram of the DCSOP.
Figure 2. Control block diagram of the DCSOP.
Energies 15 05967 g002
Figure 3. Basic configuration of a distribution network with a DCSOP.
Figure 3. Basic configuration of a distribution network with a DCSOP.
Energies 15 05967 g003
Figure 4. Main circuit topology: isolated bi-directional DC–DC converter based DCSOP.
Figure 4. Main circuit topology: isolated bi-directional DC–DC converter based DCSOP.
Energies 15 05967 g004
Figure 5. Main circuit topology: non-isolated bi-directional DC–DC converter based DCSOP.
Figure 5. Main circuit topology: non-isolated bi-directional DC–DC converter based DCSOP.
Energies 15 05967 g005
Figure 6. Circuit configuration of the proposed bi-directional DC–DC converter based DCSOP.
Figure 6. Circuit configuration of the proposed bi-directional DC–DC converter based DCSOP.
Energies 15 05967 g006
Figure 7. Reliability analysis sets as described in [29].
Figure 7. Reliability analysis sets as described in [29].
Energies 15 05967 g007
Figure 8. DCSOP multi-agent system structure.
Figure 8. DCSOP multi-agent system structure.
Energies 15 05967 g008
Figure 9. DCSOP agent (a) and load agent (b) implementation in Stateflow.
Figure 9. DCSOP agent (a) and load agent (b) implementation in Stateflow.
Energies 15 05967 g009
Figure 10. Configuration for the DCSOP stress test.
Figure 10. Configuration for the DCSOP stress test.
Energies 15 05967 g010
Figure 11. Change of the current and power flow direction as a result of a repeated and very aggressive step change in power flow direction.
Figure 11. Change of the current and power flow direction as a result of a repeated and very aggressive step change in power flow direction.
Energies 15 05967 g011
Figure 12. DCSOP terminal voltages as a result of a repeated and very aggressive step change in power flow direction.
Figure 12. DCSOP terminal voltages as a result of a repeated and very aggressive step change in power flow direction.
Energies 15 05967 g012
Figure 13. System diagram for testing the control of the current flow direction.
Figure 13. System diagram for testing the control of the current flow direction.
Energies 15 05967 g013
Figure 14. Current across the DCSOP, showing control of the current flow direction.
Figure 14. Current across the DCSOP, showing control of the current flow direction.
Energies 15 05967 g014
Figure 15. DC version of the CERTS microgrid.
Figure 15. DC version of the CERTS microgrid.
Energies 15 05967 g015
Figure 16. Currents through the DC microgrid feeders before/after connecting the DCSOP.
Figure 16. Currents through the DC microgrid feeders before/after connecting the DCSOP.
Energies 15 05967 g016
Figure 17. Currents from the DC microgrid generators before/after connecting the DCSOP.
Figure 17. Currents from the DC microgrid generators before/after connecting the DCSOP.
Energies 15 05967 g017
Figure 18. Currents through the DC microgrid feeders before/after connecting a normal conductor.
Figure 18. Currents through the DC microgrid feeders before/after connecting a normal conductor.
Energies 15 05967 g018
Figure 19. Currents from the DC microgrid generators before/after connecting a normal conductor.
Figure 19. Currents from the DC microgrid generators before/after connecting a normal conductor.
Energies 15 05967 g019
Figure 20. DC microgrid with DCSOP in locations A and B.
Figure 20. DC microgrid with DCSOP in locations A and B.
Energies 15 05967 g020
Figure 21. (a) Agent states sequence and (b) results from DC microgrid simulation over a period of 0.5 s, as shown in the horizontal axis.
Figure 21. (a) Agent states sequence and (b) results from DC microgrid simulation over a period of 0.5 s, as shown in the horizontal axis.
Energies 15 05967 g021
Table 1. Rationale behind the set of case studies used for evaluating DCSOP operation.
Table 1. Rationale behind the set of case studies used for evaluating DCSOP operation.
DCSOP Operation Case StudyRationale
(C.1) Stress testEvaluate controller performance
(C.2) Control of power flowAssess the ability of the controller to regulate power flow
(C.3) DCSOP in a microgridSimulate the effect of a DCSOP and evaluate potential benefits such as reduced line loading
(C.4) Comparison between DCSOP and simple NOPIdentify differences between DCSOP functionality against a simple DC line equivalent
Table 2. Currents through the DC microgrid feeders before/after connecting the DCSOP.
Table 2. Currents through the DC microgrid feeders before/after connecting the DCSOP.
BranchesCurrents before Connecting the DCSOP, ACurrents after Connecting the DCSOP, A
Bus 1–Bus 272.533.3
Bus 2–Bus 372.533.3
Bus 3–Bus 4072.8
Bus 4–Bus 50−72.8
Bus 2–Bus 6217.5100
Bus 6–Bus 7102.6119.4
Bus 2–Bus 872.533.3
Bus 8–Bus 900
Bus 8–Bus 10242.3242.3
DCSOP0−145.6
Table 3. Currents from the DC microgrid generators before/after connecting the DCSOP.
Table 3. Currents from the DC microgrid generators before/after connecting the DCSOP.
GeneratorsCurrents before Connecting the DCSOP, ACurrents after Connecting the DCSOP, A
Generator 172.533.3
Generator 2322.5356.7
Generator 3−149.8−222.6
Generator 4615.5576.4
Generator 5149.8149.8
Table 4. Comparison between the performance of the DCSOP and the normal conductor in terms of the change of branch currents.
Table 4. Comparison between the performance of the DCSOP and the normal conductor in terms of the change of branch currents.
BranchesCurrents with NOP, ACurrents with a Normal Conductor Closing the NOP, A
Bus 1–Bus 272.537.9
Bus 2–Bus 372.537.9
Bus 3–Bus 4067.6
Bus 4–Bus 50−67.6
Bus 4–Bus 60−135.2
Bus 2–Bus 6217.5113.8
Bus 6–Bus 7102.6117.4
Bus 2–Bus 872.537.9
Bus 8–Bus 900
Bus 8–Bus 10242.3242.3
Table 5. Comparison between the performance of the DCSOP and the normal conductor in terms of the change of generator currents.
Table 5. Comparison between the performance of the DCSOP and the normal conductor in terms of the change of generator currents.
GeneratorsCurrents with NOP, ACurrents with a Normal Conductor Closing the NOP, A
Generator 172.538
Generator 2322.5355.5
Generator 3−149.8−217.4
Generator 4615.5581
Generator 5149.8149.8
Table 6. Comparison of all reliability cases. Case R.1 is the base case.
Table 6. Comparison of all reliability cases. Case R.1 is the base case.
CAIDI% ImprovementSAIDI% Improvement
Case R.11.269-1.0497-
Case R.20.96124.3%0.9648.1%
Case R.30.96424.0%0.746128.9%
Case R.40.65848.1%0.65637.5%
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Ramadan, H.A.; Skarvelis-Kazakos, S. DC Soft Open Points for Resilient and Reconfigurable DC Distribution Networks. Energies 2022, 15, 5967. https://doi.org/10.3390/en15165967

AMA Style

Ramadan HA, Skarvelis-Kazakos S. DC Soft Open Points for Resilient and Reconfigurable DC Distribution Networks. Energies. 2022; 15(16):5967. https://doi.org/10.3390/en15165967

Chicago/Turabian Style

Ramadan, Husam A., and Spyros Skarvelis-Kazakos. 2022. "DC Soft Open Points for Resilient and Reconfigurable DC Distribution Networks" Energies 15, no. 16: 5967. https://doi.org/10.3390/en15165967

APA Style

Ramadan, H. A., & Skarvelis-Kazakos, S. (2022). DC Soft Open Points for Resilient and Reconfigurable DC Distribution Networks. Energies, 15(16), 5967. https://doi.org/10.3390/en15165967

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop