CC BY
15Distributed Ancillary Services in Smart Distribution Grids: Demand, Requirements and Benefits
Marc Richter1'*, Przemyslaw Trojan2, André Naumann1, Przemyslaw Komarnicki3
1 Fraunhofer Institute for Factory Operation and Automation IFF, Magdeburg, Germany
2 Otto von Guericke University, Magdeburg, Germany
3 Magdeburg-Stendal University of Applied Sciences, Magdeburg, Germany
Abstract — The progressing distribution of the electricity supply necessitates redesigning the mechanism for providing ancillary services particularly by the distribution grid. Methods of voltage regulation and congestion management particularly have to satisfy new standards since, although the development of renewables is increasing the number of resources with an impact, these resources' individual contribution is comparatively slight. Taking the state-of-the-art and the basic regulatory conditions in Germany as a point of departure, this paper analyzes the requirements for algorithms and communication systems that provide distributed support to distribution grid operation. A novel mathematical method that prevents voltage range deviations and feeder overloads based on sensitivities is presented and validated in simulations by a case study. An analysis of the communications systems for monitoring and control technologies for distributed energy resources, including the available communication channels, serves as the basis for an evaluation of the suitability of current control mechanisms in the future. The findings of a live field test in a real 110 kV distribution grid corroborate the necessity for coordinated grid support by distributed energy resources and demonstrate the limits of current methods.
Index Terms—Ancillary services, distribution network, active/reactive distributed power control, renewables integration, communication standards for distributed energy resources, live test.
* Corresponding author. E-mail: [email protected]
http://dx.doi.org/10.25729/esr.2018.04.0001 Available online February 28, 2019.
© 2018 ESI SB RAS and authors. All rights reserved.
I. Using distributed energy resources to support the grid
A. Introduction
Electrical grids and distribution grids in particular are currently undergoing a transition. Distributed sources, flexible loads and (stationary and mobile) storage systems will affect their operation in the future [1]. The growing number of distributed energy resources [2] will be operated primarily based on market factors [3]. This is the case in Germany in particular. Resources will supply energy at times when it is not necessarily expedient in terms of benefit to the grid and sometimes even detrimental to grid stability [4]. Distribution grids will have to be made smarter [5] and be able to use distributed electricity generation, loads and storage systems optimally for the current grid situation [6]. Distribution grid control centers will have to coordinate optimal operation and the requisite data exchange between control centers and distributed resources will have to be integrated [7], [8]. Good observability of the distribution grid will also be essential.
This paper presents the requirements that smart distribution grids ought to meet and methods for implementing them technically. Optimization algorithms and concepts for linking information systems of distributed resources are also presented. Although some aspects of observability are examined, readers are primarily referred to other literature.
B. State-of-the-Art Power System
An analysis of the situation in Germany reveals a rated generating capacity of 183.6 [9] at a peak demand of 82 GW. Around 94 GW (51%) of this rated capacity comes from renewable and thus distributed energy resources alone. They are supplemented by distributed energy resources operated with fossil fuels. The trend is toward an increasing share of renewables, thus making generation even more important on the distribution grid level [10]. For comparison, the rated capacity of storage systems
Technology Active (♦) / reactive (•) power support SC power in multiples of Sr Island Black operation start
Photovoltaic converters ----------------\ 1.1 -1.5 No No
p----------m I « i i
Wind turbines i i ♦......----------------"! 1.1 No No
; ---------- : 1 1
Biogas CHP plants ♦------L> *-------------• ! • i i >2.0 Yes Yes
Hydro power plants (2) ! (3) +------*-3-I—* ¡»-------------« ; » i i ■ i > 2.0 Yes Yes
ms ......... s ......... min .........15min .........1h ......►
dynamic steady-state
(1) Converter-based inertia emulation -Capability including start up
(2) Generator inertia — ■ — Requires power from RES
(3) Average start-up time (cold) ——. In operation
■ Instantaneous reserve
Fig. 1. Overall structure of the proposed aggregator.
is relatively low, the largest number being conventional storage systems such as pumped-storage power stations with 37 GWh in Germany [11], [12]. While technologies such as compressed air energy storage and battery systems are in use, their numbers are insignificant at present [13], [14], [15], [16]. Around 34000 battery energy storage systems with PV systems were installed in Germany in 2015. Assuming an average storage system has a capacity of 15 kWh, the total capacity of PV energy storage systems is approximately 510 MWh [12], equivalent to merely a fraction of energy storage capacities.
Distributed resources injecting power into the grid are primarily operated to maximize energy output and commercial profit. Benefit to the grid thus plays a subordinate role and grid operators can only take action to prevent equipment overloads in critical situations. In Germany, such actions are regulated in Section 13 of the Energy Industry Act [17]. This law stipulates that grid operators first have to perform appropriate switching operations, take market-related actions or activate additional reserves in order to ensure an electrical grid's reliability. Only when such actions are insufficient, grid operator may take-grid-related actions to curtail (the active power of) energy resources. Market factors are subordinate in this second step.
A grid operator usually takes action in the context of grid stability management, solely reducing the active power injected by resources [18]. Other potential capabilities of resources, e.g. supplying reactive power for voltage backup or microgrid capability, are not utilized at present. There is no regulatory and economic framework to offer system operators an incentive to provide resources for
this. Moreover, sufficient measurements for system state estimation and thus suitable dynamic control algorithms for integration of distributed resources beneficial to the grid are not always available for grid operation.
C. Analysis of Ancillary Service Capabilities of Distributed Energy Resources
Relevant requirements ought to be listed first in order to assess distributed resources' capabilities. In principle, resources should be able to contribute to providing ancillary services [19] [20]. Their contribution is relevant on every level of the electricity supply and has increasingly to be covered by the growing number of distributed resources in the distribution grid. The typical features of each resource technology are crucial to capability. An overview of researchable capabilities beside storage systems as in [21] is presented in Fig. 1.
RES power plants are already able to supply active and reactive power within a very short time. Synthetic inertia emulation even enables converters to contribute instantaneous reserves [22]. Regardless of a resource size, they are not necessarily conditional on the availability of incoming power since power can be drawn from the intermediate circuit. Conventional generator systems still obtain this power from their inertia. Frequency support as well as grid restoration capabilities are nevertheless the domain of non-volatile resources alone [23], [24].
Reactive power is normally injected within a few grid cycles [24] regardless of its generation [25]. Increased use of converters decreases short-circuit power in the electrical grid. Protection necessitates usually limiting short-circuit current to 1.1 - 1.5 times the nominal current for [26], [27]. Only generator resources are able to supply much more
nominal current [28].
Other services, e.g. supplying short-circuit current, improving power quality, and supplying microgrid capability and black-start capability, will presumably gain importance in the future. This will also generate new opportunities for the use of renewable and distributed energy resources, appropriate control algorithms having to be created and suitably adapted to the regulatory framework.
ii. Active and Reactive power Adjustment Beneficial to the Grid
A Three-Level Approach
An analysis can be performed on three different levels corresponding to the increasing contribution to ancillary services to utilize distributed resources capability to benefit the grid. On the simplest level, a distributed resource can be controlled so that it selects an optimized operating point at its connection point from the available measurements. To do this, the resource's local control system must continuously record voltage and frequency values at the connection point. These values are used to adjust the resource's operation to attain active and reactive power values that maximize system support based on the situation at the resource's connection point.
On the second level, the resource is run not only optimized, given the local connection point, but also with the goal of benefitting the grid in the section of the distribution grid with which it is connected. For voltage regulation, this can mean that voltage level is monitored along the grid branch (or ring) and distributed resources are controlled locally and regionally so that the voltage level has an optimal value within the grid branch (or ring). The active power supply is controlled in keeping with the requirements of the primary substation. Local and regional grid segments are incorporated for restoration of supply and microgrid capability to ensure a stable supply to these sections in the event of a malfunction.
The control of distributed resource on the third level goes beyond a local and regional analysis and is utilized to contribute to ancillary services in coordination with the primary grid. Direct or indirect communication must be ensured to transfer setpoints required for active and reactive power between primary grid operators and the particular resource, possibly in the form of a schedule. Whereas only the surrounding grid up to the primary substation is relevant for the operation of distributed resource on the other two levels, this does not constrain contributions to ancillary services on this level. Coordination with other distributed resources is essential to obtain optimal results since this maximizes the contributions to ancillary services [29], [30]. This requirement and the resultant requirements of requisite information and communications technologies (ICT) and coordination additionally makes this the most complex option of all.
The third level, which contributes most to global system
stability, always ought to be striven for to stabilize operation. Moreover, the requirements of the two lower levels are taken into account during operation on these levels so that local and regional stability as well as the contribution to global ancillary services can be implemented.
B. Iterative Control Algorithm Using Sensitivity Analysis
An optimization algorithm that provides distributed support to the electrical grid is presented here. It ascertains an optimal setpoint value for individual resources in order to provide an ideal value for voltage response in the grid and to ensure that no equipment is overloaded. It is based on a method based on standard network theory, which depicts the distribution of flows of different resources between loads and lines. Approaches based on demand response using sensitivity matrices have been studied in [31]. Priority is given to renewable energy sources. Network structure parameters can employ diversity factors to identify the power diversity. This method can be used for economic dispatch and power flow analysis.
The active and reactive power adjustment necessary in each case is determined based on the Jacobian matrix J determined with the Newton-Raphson method for power flow studies. This matrix depicts the correlation between the active and reactive power buses and the electrical system state variables (voltage magnitude U and angle ff):
API = UPe JPU AQj \Jq0 JQU
The following correlation results for the influence of purely
'] H
,\ lauj
(3)
Fig. 2. Iterative optimization method for grid support
Fig. 3. Distribution grid structure for the field test
AO = Oqu - /QeJpeJpu) • A ^ = JR • AU (2)
The influence of reactive power control AQ in the grid area analyzed can consequently be ascertained by inverting JR according to (X).
AUJ - AQ (3)
The coefficients in the matrix Jr1 indicate which buses have the greatest sensitivity to other buses, a higher value indicating a greater correlation. An analogous approach can be applied to active power control.
The optimization function is flowcharted in Fig. 2. The algorithm uses the Jacobian matrix in every iteration to verify the voltage values. In the event of voltage problems at the system buses, the reactive power adjustment is retrieved selectively until the grid malfunction has been eliminated. This algorithm runtime has proven practicable in tests since it completes the calculation in 0.5 s. This applies to the analyzed grid computed on a standard computer. It can be scaled up for larger systems. This also makes the algorithm suitable for operative use by CHPs controlled with parameters for a control strategy considering forecasts in less than one second [32].
C. Simulation Evaluation
Since thresholds were not exceeded during the entire period, the control algorithms did not automatically activate. Different operating points were therefore created in an 18-bus test system (see Fig. 3) to verify the algorithm developed to regulate voltage. At the start of every scenario, every dispatchable generator (at buses 4, 8 and 11, designated G4, G8 und G11) is assumed to supply 10 Mvar of reactive power. The outcomes of the simulations are presented in Table 1. Supplying reactive power stabilizes high voltage more efficiently. Adjusting active power would additionally entail undesired actions in the electricity market.
The simulation results presented in Table I (scenarios 1 to 6) confirm the optimization algorithm effectiveness. In scenario 1, generator G8 is selected to support the grid actively by supplying reactive power. The reactive power
setpoints are adjusted to 1.13 Mvar. The voltage at bus 1 has already returned to the feasible value of 1.1 pu after the third iteration. Similar effects are observable in scenarios 2, 3 and 4. In the latter, generators with low sensitivities are additionally included for support since the technical limits of G8 have been reached. The number of iterations needed to calculate new setpoints rises to eleven loops. Scenarios 5 and 6 confirm the method effectiveness even when voltage is unduly low. Increase in reactive power at G4 and G8 also eliminates the stability problem.
D. Remote Control and Monitoring of Distributed Energy Resources
1) Introduction
Distributed energy resources integrated in an existing grid have to meet standard technical specifications and communications interface specifications, thus ensuring that they can be monitored with sufficient accuracy and time resolution, and controlled when necessary [33]. The interface of communications systems has to be designed accessibly so that different systems are interoperable [34]. Using open standards and standard interfaces is expedient [35], [36]. Practicable methods for this are presented below.
2) IEC 60870 and IEC 61850 Interface
Established and advanced remote control protocols such as IEC 60870 and IEC 61850 improve distributed resources' interface [37], [38]. This enables every relevant measurement to be continuously transmitted to a control center [39]. Any setpoint value desired can also be implemented within the limits of a particular resource technical capability. This means that both active and reactive power setpoint values can be varied dedicatedly in fine steps [40]. For instance, power electronic converters in advanced wind and PV systems can contribute substantially and flexibly to the local supply of reactive power. This can only be activated for grid operation with a proper communications interface, though. Although advanced and large resources have such interfaces, grid operators do not take full advantage of them at present [41]. Moreover, since the number of resources equipped with interfaces is very limited, conventional solutions still have to be employed.
Table 1. Simulation results
Before optimization After optimization
Scenario Min. bus Max. bus QG4 Qg8 QG11 Min. bus Max. bus Iteration
voltage voltage in Mvar in Mvar in Mvar voltage voltage
in pu. in pu. in pu. in pu.
1 Ui7 = 1.028 U1 = 1.102 10.00 1.13 10.00 U17 = 1.028 U1 = 1.100 3
2 Ui7 = 1.028 U1 = 1.107 10.00 -22.65 10.00 U17 = 1.028 U1 = 1.100 7
3 Ui7 = 1.028 U1 = 1.108 10.00 -29.64 10.00 U17 = 1.028 U1 = 1.099 8
4 Ui7 = 1.028 U1 = 1.114 3.49 -30.00 -10.00 U17 = 1.028 U1 = 1.099 11
5 Ui = 0.899 U17 = 1.028 10.00 18.25 10.00 U1 = 0.901 U17 = 1.028 2
6 U1 = 0.885 U17 = 1.028 40.00 30.00 10.00 U1 = 0.900 U17 = 1.028 7
3) Grid Stability Management Interface
Since the majority of distributed resources can only be controlled by a grid stability management interface, upgrades of installed equipment must be allowed. Such an upgrade enables the transmission of current measurements so that previously "blind" control systems of resources also deliver current measurements by wireless control signal and the grid operator is immediately notified of the resource's response. Even though this interface only facilitates relatively rough control of active power, while not permitting any control of reactive power, it improves the use of distributed resources to apply the aforementioned algorithm.
The additional technology that has to be installed to implement this control system and to transmit measurements basically consists of an additional current and voltage transformer measurement logger and an interface to the existing grid stability management interface, which normally consists of four main switching contacts. Every contact corresponds to one of four stages (100% - 60% - 30% - 0%). Conventional control systems (programmable logic controllers) provided by remote control interfaces can both log measurements and connect switches. MODBUS/TCP, IEC 60870, IEC 61850 or DNP3 can be used as the communications protocol for a remote control interface, thus enabling the grid operator to access resource parameters directly. A biomass CHP plant serves as an example in the design for upgrading existing resources diagrammed in Fig. 4.
4) Adding RTUs
Retrofitting with an RTU is a more flexible but also more expensive way of equipping an existing resource with a suitable remote monitoring and control system than that described in section 3). An additional RTU provided by such interfaces as IEC 61850 and IEC 60870 is subsequently integrated in the distributed resource's process control system. Normally, this is done by using Profibus, CAN bus or EtherCAT to link the retrofit RTU with the distributed resource's existing process control system. This makes the retrofitted RTU a gateway that exchanges defined data (measurement and setpoint values) between the resource's process equipment and the grid operator's remote control system. The RTU must be carefully configured since, in the worst case, direct access to the resource control system can damage the resource if the RTU has been configured incorrectly. The grid operator has to configure and test the IEC 61850 or IEC 60870 interface based on the resource features and the specifications.
E. Selecting Appropriate Communication Channels
1) Dedicated Lines through Grid Operators
A line installed by the electrical grid operator and dedicated to transmitting the resource measurement data and setpoint values can be used to connect distributed resources. This is a state-of-the-art approach to interfacing
the communications system of substations but is inconsistently employed for distributed energy resources. The advantage of this approach is its provision of a very reliable interface explicitly established for this purpose, which is highly reliable and available. The relatively high installation costs are a drawback.
2) Public Network DSL or ISDN Line
The advantage of using existing lines of operators of public telecommunications networks such as DSL or ISDN is that they are already virtually ubiquitous and can be used to connect distributed resources. A landline telecommunications infrastructure may inadequately cover rural regions where the majority of distributed resources are installed. While ISDN lines and DSL suffice for the quantities of data normally transmitted, they may also be used for other public communication purposes, thus causing congestion. A factor viewed critically by grid operators in particular is property rights to the respective lines and thus also monopolies on lines. This can cause trouble in the event of a malfunction. Data security is more serious. Since public telecommunications networks are physically linked directly with the public Internet, actions have to be taken to prevent unauthorized access to the resource's or the grid operator's data network. Complex security measures and gateways have to be planned between the grid operators' internal network and the public network.
3) Mobile Broadband
Along with landline public telecommunications networks, public cellular networks can also be employed to monitor and control distributed resources. Since this does not necessitate installing additional lines to the particular resource, the capital expenditures are relatively low. This communication channel is the most unreliable in practice though. Cellular coverage may be inadequate in rural regions and is heavily dependent on weather conditions. Packets may be transmitted incompletely or the transmission bandwidth may not suffice at times, depending on the connection quality. Since this telecommunications network is public, the security constraints and associated additional security measures for landline public telecommunications networks also apply (see section 2)).
Fig. 4. Interface design.
4) Selections in Comparison
The relevant features of the different technologies are compared in Table II.
5) Future Communication Channels
Additional communication channels can be expected to gain importance in the future. Power-line communication systems, for instance, can transmit data from a distributed resource to the nearest substation where other technologies forward them. This approach could grow more prevalent in the future [42], [43], [44], [45].
Efforts are also made to regulate radio frequency bands dedicated to transmit data for electrical grid operation [46]. Frequencies of 400 MHz and 450 MHz that provide a channel bandwidth sufficient to transmit a sufficient payload are especially relied on here. The band frequency should be low enough to expand physically (with appropriately low free-space path loss) to connect to respective distributed resources reliably.
III. Practical Limits of Active Power Adjustments
A. Field Test Environment
A 110 kV voltage section of an 18-bus distribution grid (see Fig. 3) served as the field test area. The grid is highly saturated with RES (160 MW of wind and 40 MW of photovoltaic power) and has a peak load of220 MW. An 80 MW wind farm was used to test actions using active power to stabilize voltage. Two high voltage transformers connect it with the grid. The large electrical distance to the 110/380 kV connection point (bus 14) minimizes influences from the high voltage system.
Voltages were recorded directly at the wind farm connection point U4 to analyze the effects of active power control on the rest of the grid and additionally at substation U2 approximately 20 km away. The wind farm is represented by generator G4.
B. Findings
The recorded curves of the incoming power supplied by the wind farm and the related voltage values at the wind farm U2 and at the adjacent station U2 are presented in Fig. 5.
The active power was curtailed when the load was low for monetary reasons. The control signal was sent at t1 = 30 min. Instruments verified the complete cessation
Fig. 5. Active power and voltage curve in the field test.
of active power injection after about 4 min. The curtailed power was 8.1 MW and accompanied by a voltage drop of 0.3 kV at both the connection point and the nearby substation. The grid stability management interface was reactivated at t2 = 40 min.
C. Evaluation
A correlation k [47], [48] of the voltage curves with the active power curve can be discerned. The mathematical correlation factors of the active power PG4 with the voltage at the wind farm connection point U4 or the active power PG4 with the nearby substation voltage U2 at bus 2 are as follows:
Z: = I0G4,1 - Pga) p4,1 - U4)
£"=I(PG4,¿ - Pg¡Y ■ - Ü¡f
S"=i(JW - fb¡) (u2,i - üd
£"=i(pG4,¿ - Pg¡)2 ■ 2"=I(^2,Í - Ül)2
0,84 (4)
0,73 (5)
Factoring in a potential range of the correlation factor from -1 (completely negative correlation) to 0 (no correlation) to +1 (completely positive correlation), the field test demonstrated that:
1. the active power control influences voltages both at the connection point and in the nearby grid,
2. the voltage effect diminishes greatly as distance increases, and
3. although an active power variation-voltage causality is
Table 2. Telecommunication technologies in comparison.
Parameter Dedicated network operator lines Public lines (ISDN, DSL) Public wireless (GPRS, UMTS, LTE)
Availability (space) Limited High Variable
Availability (time) Constant Very high Variable (weather-dependent)
Reliability Very high High Low
Cost High Medium Low
Security Very high Low Medium
present, its impact is unduly low.
Consequently, the standard grid stability management interface is "not suited" for limiting active power. On the one hand, economic losses ensue since available renewable power cannot be injected into the electrical grid. Therefore, additional battery storage systems could be used to minimize impacts [49]. On the other hand, the simulation tests have demonstrated that smart and coordinated reactive power control by distributed resources has a greater impact on the grid while keeping costs the same.
iv. conclusion
The capabilities of renewable energy plants and distributed energy resources to play a role in certain ancillary services was scrutinized and the technical, communication and regulatory requirements for this were analyzed. Capabilities to eliminate voltage problems by adjusting reactive power in selected resources while minimizing the impact on ongoing grid operation and not taking market-related actions were presented in simulations. A field test demonstrated the technical limits of conventional curtailment of renewable energy plants. The moderate influence of unilateral control of active power on voltage as well as the shortfall renewable energy at equal cost caused by it was decisive.
An increasing response to critical situations directly at the local level and with more intensive coordination (among distribution grid operators as well) will therefore be expedient in the future. This approach will guarantee a globally optimized result while minimizing the impact on grid operation and components [50]. Advanced renewable energy plants already have the technical capabilities to implement fine operating points while employing advanced communications standards [51], [52]. More intensive monitoring of low voltage levels and the use of available communication channels to interface field devices are therefore an important prerequisite for a more dynamic and accurate control system [53]. Germany requires few regulatory changes to implement such a system. Part of the standards have already been incorporated in amendments (e.g. to the 2017 German Renewable Energy Act [19]) and regulations (e.g. 2015 Regulation of Ancillary Services by Wind Turbines [54]).
Acknowledgment
This paper was written as part of the research project SECVER funded by the German Federal Ministry for Economic Affairs and Energy.
Mathematical symbols k Correlation factor J Jacobian matrix
P Active Power Q Reactive power S Apparent power
t Time
U Voltage
Abbreviations CHP Combined heat and power DER Distributed energy resource DSL Digital subscriber line DSO Distribution system operator EEG Erneuerbare-Energien-Gesetz (German
Renewable Energy Sources Act) GPRS General Packet Radio Service ISDN Integrated Services Digital Network LTE Long Term Evolution RCR Remote Control Relay SDLWindV
Verordnung zu Systemdienstleistungen durch Windenergieanlagen (Regulation of Ancillary Services by Wind Turbines) TSO Transmission system operator UMTS Universal Mobile Telecommunications System
References
[1] J. Wang, "Smart grids, renewable energy integration, and climate change mitigation - Future electric energy systems," Applied Energy, pp. 1-3, 2012.
[2] J. Wu, "Integrated Energy Systems," Applied Energy, pp. 155-157, 2016.
[3] J. Winkler, M. Pudlik, M. Ragwitz and B. Pfluger, "The market value of renewable electricity - Which factors really matter?," Applied Energy, pp. 464-481, 2016.
[4] N. Ederer, "The market value and impact of offshore wind on the electricity spot market: Evidence from Germany," Applied Energy, pp. 805-814, 2015.
[5] N. I. Voropai and B. Bat-Undraal, "Multicriteria reconfiguration of distribution network with distributed generation," Journal of Electrical and Computer Engineering, 2012.
[6] C. F. Calvillo, A. Sanchez-Miralles and J. Villar, "Optimal planning and operation of aggregated distributed energy resources with market participation," Applied Energy, pp. 340-357, 15 11 2016.
[7] J. Yan, Y. Zhai, P. Wijayatunga, A. M. Mohamed and P. E. Campana, "Renewable energy integration with mini/micro-grids," Applied Energy, pp. 241-244, 2017.
[8] S. B. Myungchin Kim, "Decentralized control of a scalable photovoltaic (PV)-battery hybrid power systems," Applied Energy, pp. 444-455, 2016.
[9] Bundesnetzagentur, "Kraftwerksliste,"
Bundesnetzagentur, 2017. [Online]. Available: https://www.bundesnetzagentur.de/cln_1411/ DE/Sachgebiete/ ElektrizitaetundGas/
Unternehmen_ Institutionen/Versorgungssicherheit/ Erzeugungskapazitaeten/Kraftwerksliste/ kraftwerksliste-node.html. [Zugriff am 24 April 2017].
[10] H. U. Heinrichs and P. Markewitz, "Long-term impacts of a coal phase-out in Germany as part of a greenhouse gas mitigation strategy," Applied Energy, pp. 234-246, 2017.
[11] J. L. Peter Stenzel, "Concept and potential of pumped hydro storage in federal waterways,"Applied Energy, pp. 486-493, 2016.
[12] J. Schmidt-Curreli, A. Knebel und L. Lawrenz, "Energiewendeatlas Deutschland 2030," Agentur für erneuerbare Energien e.V., Berlin, 2016.
[13] P. A. Lombardi and F. Schwabe, "Sharing economy as a new business model for energy storage systems," Applied Energy, pp. 485-496, 2017.
[14] C. Guo, L. Pan, K. Zhang, C. M. Oldenburg, C. Li und Y. Li, "Comparison of compressed air energy storage process in aquifers and caverns based on the Huntorf CAES plant," Applied Energy, pp. 342-356, 2016.
[15] R. Madlener and J. Latz, "Economics of centralized and decentralized compressed air energy storage for enhanced grid integration of wind power," Applied Energy, pp. 299-309, 2013.
[16] M.-A. Hessami and D. R. Bowly, "Economic feasibility and optimisation of an energy storage system for Portland Wind Farm (Victoria, Australia)," Applied Energy, pp. 2755-2763, 2011.
[17] Bundesrepublik Deutschland, "Gesetz über die Elektrizitäts- und Gasversorgung (Energiewirtschaftsgesetz - EnWG)," Berlin, 2017.
[18] V. M. Michael Waite, "Modeling wind power curtailment with increased capacity in a regional electricity grid supplying a dense urban demand,"
Applied Energy, pp. 299-317, 2016.
[19] Bundesrepublik Deutschland, "Gesetz für den Ausbau erneuerbarer Energien (EEG)," Berlin, 2017.
[20] VDE, "TAB Hochspannung (VDE-AR-N 4120)," VDE, Frankfurt, Germany, 2015.
[21] S. Balischewski, C. Wenge, P. Komarnicki and M. Wolter, "Optimized operation of energy storages for primary control reserve," in Proc. of 2018 NEIS Conference on Sustainable Energy Supply and Energy Storage, Hamburg, Germany, 2018.
[22] F. M. Gonzalez-Longatt, "Activation schemes of synthetic inertia controller on full converter wind turbine (type 4)," in Proc. of 2015 Power & Energy Society General Meeting, Denver, CO, USA, 2015.
[23] A. Hoke, M. Shirazi, S. Chakraborty, E. Muljadi and
D. Maksimovic, "Rapid Active Power Control of Photovoltaic Systems for Grid Frequency Support," IEEE Journal of Emerging and Selected Topics in Power Electronics, 2016.
[24] Sandia National Laboratories, "Reactive Power Interconnection Requirements for PV and Wind Plants
- Recommendations to NERC," Sandia National Laboratories, Albuquerque, NM, 2012.
[25] SMA, "Technical Information: Q at Night," SMA, Kassel, Germany.
[26] D. Turcotte and F. Katiraei, "Fault Contribution of Grid-Connected Inverters," in Proc. of IEEE Electrical Power Conference, Montreal, Quebec, Canada, 2009.
[27] E. Muljadi, N. Samaan, V. Gevorgian, J. Li and S. Pasupulati, "Short Circuit Current Contribution for Different Wind Turbine Generator Types," in Proc. of 2010 IEEE PES General Meeting, Providence, RI, USA, 2010.
[28] WA notstromtechnik GmbH, "StromerzeugerLexikon," [Online]. Available: http:// www.stromerzeuger-lexikon.de/lexikon/ kurzschlusleistung. [Zugriff am 24 April 2017].
[29] D. Wang, S. Parkinson, W. Miao, H. Jia, C. Crawford and N. Djilali, "Hierarchical market integration of responsive loads as spinning reserve," Applied Energy, pp. 229-238, 2013.
[30] P. A. Ostergaard, "Ancillary services and the integration of substantial quantities of wind power," Applied Energy, pp. 451-463, 2006.
[31] O. Homaee, A. Zakariazadeh, S. Jadid and P. Siano, "A new approach for real time voltage control using demand response," Applied Energy, vol. 117, pp. 157166, 2014.
[32] M. Short, T. Crosbie, M. Dawood and N. Dawood, "Load forecasting and dispatch optimization for decentralied co-generation plant with dual energy storage," Applied Energy, vol. 186, pp. 304-320, 2017.
[33] B. M. Buchholz and Z. A. Styczynski, Smart Grids
- Fundamentals and Technologies in Electricity Networks, Berlin Heidelberg: Springer-Verlag, 2014.
[34] G. Ghatikar, S. Mashayekh, M. Stadler, R. Yin and Z. Liu, "Distributed energy systems integration and demand optimization for autonomous operations and electric grid transactions," Applied Energy, vol. 167, pp. 432-448, 2016.
[35] W. Shi, X. Xie, C.-C. Chu and R. Gadh, "Distributed Optimal Energy Management in Microgrids," IEEE Transactions on Smart Grid, pp. 1137-1146, 05 2015.
[36] G. Zapata, A. Salazar, D. Moreno and R. Garcia, "Supervision of a distributed energy resources generation system using IEC and ISA standards," Robotics and Automation (CCRA), IEEE Colombian
Conference, 2016.
[37] N. Higgins, V. Vyatkin, N.-K. C. Nair and K. Schwarz, "Distributed Power System Automation With IEC 61850, IEC 61499, and Intelligent Control," IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), pp. 81-92, 2011.
[38] T. Dethlefs, C. Brunner, T. Preisler, O. Renke, W. Renke and A. Schröder, "Energy Service Description for Capabilities of Distributed Energy Resources," Lecture Notes in Computer Science, 6 01 2016.
[39] C. Wenge, A. Pelzer, A. Naumann and P. Komarnicki, "Wide area synchronized HVDC measurement using IEC 61850 communication," in 2014 IEEE PES General Meeting | Conference & Exposition, National Harbour, MD, USA, 2014.
[40] N. Etherden, V. Vyatkin and M. H. J. Bollen, "Virtual Power Plant for Grid Services Using IEC 61850," IEEE Transactions on Industrial Informatics, pp. 437447, 2016.
[41] S. Alepuz, S. Busquets-Monge, J. Bordonau and J. Gago, "Interfacing Renewable Energy Sources to the Utility Grid Using a Three-Level Inverter," IEEE Transactions on Industrial Electronics, pp. 15041511, 10 2006.
[42] F. R. Yu, P. Zhang, W. Xiao und P. Choudhury, "Communication systems for grid integration of renewable energy resources," IEEE Network, 2011.
[43] Y.-s. Son, T. Pulkkinen, K.-d. Moon und C. Kim, "Home energy management system based on power line communication," IEEE Transactions on
Consumer Electronics, 2010.
[44] U. Braun, "Bridging the gap with Broadband Powerline (BPL) technology," in Innovative Smart Grid Technologies (ISGT Europe), Manchester, UK, 2011.
[45] L. Karg und M. Wedler, "E-Energy Abschlussbericht - Ergebnisse und Erkenntnisse aus der Evaluation der sechs Leuchtturmprojekte," B.A.U.M Consult GmbH, München, Berlin, 2014.
[46] D. Marger, J. Riedl, J. Tusch, C. Wietfeld, I. Wolff und W. Zeittler, "Kommunikationsnetz für das Smart Grid - Positionspapier des VDE zur weiteren Ausgestaltung der Energiewende," VDE Verband der
Elektrotechnik Elektronik Informationstechnik e.V., Frankfurt am Main, 2015.
[47] K. Pearson, "Notes on the History of Correlation," Biometrika, pp. 25-45, 1920.
[48] J. Hauke and T. Kossowski, "Comparison of values of Pearson's and Spearman's correlation coefficients on the same sets of data," Quaestiones Geographicae, 2011.
[49] S. Balischewski, I. Hauer, M. Wolter, C. Wenge, P. Lombardi and P. Komarnicki, "Battery storage
services that minimize wind farm operating costs: A case study," in Proc. of 2017 IEEE PES Innovative Smart Grid Technologies Conference Europe, Torino, Italy, 2017.
[50] M. Bae, H. Kim, E. Kim, A. Y. Chung, H. Kim and J. H. Roh, "Toward electricity retail competition: Survey and case study on technical infrastructure for advanced electricity market system," Applied Energy, pp. 252-273, 2014.
[51] T. J. Hammons, "Integrating renewable energy sources into European grids," International Journal
of Electrical Power & Energy Systems, pp. 462-475, 2008.
[52] P. N. Vovos, A. E. Kiprakis, A. R. Wallace and G. P. Harrosin, "Centralized and Distributed Voltage Control: Impact on Distributed Generation Penetration," IEEE Transactions on Power Systems, pp. 476-483, 2007.
[53] S. U. Taha, C. Ozansoy and A. Zayegh, "Modeling of a Centralized Microgrid Protection System and Distributed Energy Resources According to IEC 61850-7-420," IEEE Transactions on Power Systems, pp. 1560-1567, 08 2012.
[54] Bundesrepublik Deutschland, "Verordnung zu Systemdienstleistungen durch Windenergieanlagen (SDLWindV)," Berlin, 2015.
Marc Richter holds a PhD in electrical engineering from the Otto von Guericke University of Magdeburg. He is currently an expert scientist in the Convergent Infrastructures business unit at the Fraunhofer Institute IFF. His main research interest is in demand side management and electric energy system studies. In recent years, Marc Richter has also focused on the development and operation of innovative technologies, such as digital twins and artificial intelligence, in modern power systems.
mц Przemyslaw Trojan received Iiis PhD degree in electrical engineering L ^Jft. from Otto von Guericke University M Magdeburg, Germany in 2013 and 2018 Vcf I respectively. He is currently research ^rtHPj^ assistant at the Otto von Guericke University Magdeburg at the Chair of Electric Power Networks and Renewable Energy Sources. His research interests are agent-based approaches for power system management, modelling of power systems, real-time simulations, and communication technologies.
André Naumann received his PhD at Otto von Guericke University Magdeburg in 2012. He is now group manager at Fraunhofer Institute for Factory Operation and Automation IFF. His special field of interest include protection systems in electrical energy systems as well as substation automation and communication technologies for energy systems. He is a member of the German Association for Electrical, Electronic and Information Technologies (VDE), active in the standardization process for Power Network Operation in the DKE German Commission for Electrical
Przemyslaw Komarnicki earned a dual Master's of Science from Wroclaw University of Technology in Poland and Otto von Guericke University of Magdeburg in Germany in 2004. Since 2008, he has been the manager of the Electric Power Systems Group in the Convergent Infrastructures business unit of the Fraunhofer Institute IFF. Parallel to his work, he earned
