virtual power plant
Content
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Virtual Power Plant as a bridge between Distributed Energy Resources and Smart Grid
Slobodan Lukovic, Igor Kaitovic, Marcello Mura, Umberto Bondi Faculty of Informatics - ALaRI University of Lugano Lugano, Switzerland {lukovics, muram, bondi}@alari.ch, [email protected] Abstract
The liberalization of energy markets, especially in correlation with the Smart Grid concept development, requires adjusted legislation, new business models, energy stock exchanges establishment and many other advanced instruments. Realization of these features necessitates novel concepts to support such changes in the power system while granting security and reliability of supply. Such evolution poses new challenges to ICT (Information and Communication Technologies) to bridge the gap between increased complexity of deregulated market and on the other side expected rapid growth of number of players in power systems. Increasing presence of Distributed Energy Resources (DER) implementations constitutes a further source of complexity. Bearing in mind ongoing and possible scenarios we aim to determinate the place and role of the novel Virtual Power Plants (VPP) concept, related to the Smart Grid structure. At the same time we introduce an innovative modeling approach as an instrument to determine actors and highlight their actual roles and interactions from different point of view, trying to pave the way for development of a common understanding platform for variety of stakeholders. The effectiveness of the proposed modeling concept is shown through a number of UML models representing system level description of VPP at different levels of abstraction.
eralization and decentralization are effecting and changing traditional production, distribution and consumption patterns. On the other hand the power system as a complex and sensitive entity needs modular integration of reliable components. The Smart Grid [13, 17] concept appears to be a promising response to these needs. It is fairly mature strategy that aims at leveraging energy efficiency management using the most advanced ICT, and it concerns power systems at all levels. There are many issues to be tackled by future Smart Grids; in the presented paper we focus at, and try to relate two of them: • Efficient integration of Distributed Energy Resources (DER) in an open energy market • Development of a common understanding platform based on unified modeling concept Market deregulation aims at involving a greater number of players in energy generation, transmission and distribution, and this trend results in decentralization of power flows’ control. The increased utilization of distributed generation (mostly based on renewable resources) and its efficient integration in the electricity system is considered to be a technological driver for stable, secure and sustainable energy supply in the future. On the other side, achieving efficient control over greater number of generators (especially those with low controllability) and other accompanying elements coupled with ’intelligent devices’ deployed in each of them, would result in overwhelming data generation, transmission and processing. At the same time, different Market applications also need large amounts of data for a variety of purposes (predictions, real-time pricing, stock exchange etc). All this data are eventually coming from thousands or even millions of field devices (i.e. embedded systems) employed to measure, monitor and control generators, substation elements, loads etc.
1. Introduction
Awareness to global climate changes caused by different kinds of pollution on the one side and increasing concern about sustainable energy supply at the other side have created a global sensitivity to energy production as well as an interest in finding pollution-free and sustainable solutions. The energy market is rapidly evolving and processes of lib-
978-0-7695-3869-3/10 $26.00 © 2010 IEEE
1
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Keeping supply secure and reliable in such conditions, in particular considering distributed generation, requires improved communication among all players in the system in order to match market requirements with technological capabilities of underlying structures. In order to avoid possible incompatibility between market requirements and technical feasibility, we need novel concepts and approaches to provide a common language between the different actors and thus bridge market (financial, legal, commercial etc. issues) needs and technical capabilities. Therefore, in order to successfully answer aforementioned challenges (coming from new energy market requirements and security supply must) Smart Grid has to give a response to huge communicational and computational requirements. A possible solution could be in clustering of certain power system elements in terms of their geographical, technological and/or commercial characteristics. A Virtual Power Plant (VPP) [18, 4] as a concept to aggregate distributed resources (including also storages and consumers/prosumers) and present them to the rest of the energy system as a single technical and commercial entity can be seen as a solution for this need. Such a module/cluster inside a wider Smart Grid environment should have certain autonomy and it would represent a kind of distributed computing but also trading node/unit. Being remarkably interdisciplinary topic, VPP requires holistic approach that highlights this problem from many different sides. In order to relate variety of aspects of this concept we introduce modeling methodology. So, we use VPP to show solutions for DER integration in power systems, and in particular we focus on reflection of future market requirements at such concept. We aim at determining VPP functional requirements posed by market needs and their impact on enabling ICT technologies. Moreover, in a case study of metering infrastructure we show, using again modeling approach as a an instrument, how commercial/market requirements impact embedded devices evolution as one of key actors to enable efficient execution of higher level applications. As an demonstration instrument we introduce and at the same time promote for wider acceptance - UML (Unified Modeling Language) based modeling approach. The choice of UML seams logical as it is established as industrial de-facto modeling standard; it is also widely accepted in commercial and finance community and through Common Information Model (CIM) it is already introduced to many experts in power system field. The paper is organized as follows: Section 2, after reviewing some relevant work on power system evolution we provide an overview of ongoing research on VPP. In Section 3, a conceptual, general layered structure of energy system is presented and explained. Section 4 details relevance and potential of system-level modeling approach. In Section 5 we provide deeper insight on VPP structure with
Business models
INPUTS:
Economic Dispatch (ED) Load Shedding (LS)
EMS
Voltage Var Control (VVC)
text text
Automatic Generation Control (AGC)
Load Management (LM)
Production costs
PC
System parameters status
Set points
RTU
for generator
SCADA
PC
RTU
for consumer
Energy storage
RTU
for generator
RTU
for generator
RTU
for generator
RTU
for generator
Figure 1. Traditional power system structure special regard to commercial issues, while Section 6 gives a general system level description of future smart-metering devices as case study for a certain group of intelligent devices involved in smart-grid concept. Finally, Section 7 presents conclusions and future work.
2. State of the art
Energy market is evolving from centralized to liberalized one [10]. This shift impacts entire power system and it is naturally followed by a number of open issues that are pending resolution. One of main instruments to cope with such challenges is underlying ICT system. ICT structure in contemporary energy systems is built to support centralized generation based on generationtransmission-distribution-consumption concept. The communication between software control centers and generation units, substation elements etc. is performed through integration of Remote Terminal Units (RTUs) and Intelligent Electronic Devices (IEDs) over SCADA system. On top of SCADA software extension known as Energy Management System (EMS) is managing the optimizing grid utilization in a centralized fashion (as shown in Figure 1) [16, 9]. Changes in this paradigm are introduced by several factors, among which we may mention: • Production is coming closer to the end consumers (a trend imposed by increased distributed generation) • Massive insertion of heterogeneous, non-controllable renewable resources
2
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
• Growing environmental concerns that call for higher energy efficiency • Introduction of new business models • Increased need for Demand Side Management, real time pricing etc. The smart-grid [13, 17] emerge as a promising concept to tackle these trends. Its development and realization attracts huge research efforts. A fairly large number of issues ranging from standardization to concrete solutions for secure and reliable integration of distributed generation [6] have emerged to be a fundamental concern for both academia and industry. Researchers are addressing smart-grid issues from different aspects; seamless integration of distributed, mostly renewable resources is one of key issues. A novel concept to answer this challenge is Virtual Power Plant (VPP) [18, 22, 4]. Different issues are considered in VPP (each of them involves ICT in some way): • Distributed Energy Management Systems (DEMS) [16, 3] • VPP as trading entity [12] • Interoperability [19] • Security issues [14] • Communication interfaces [8] • Different implementation possibilities [12] In order to develop fully modular and scalable power system absolute standard harmonization must be achieved. This problem is addressed by many industrial and academic institutions and huge progress has been done in this field. Standardization of information models and communication protocols of data and control flow going from field devices up to high level Control Center (CC) applications has been well established (refer to Figure 2) [20, 8] .
Figure 2. Standards harmonization Implementation of such advanced features (as real-time pricing, DMS etc.) will require increased control over the grid and tight interaction with all actors in energy market (from energy stocks and utilities to the end-consumer) which further implies massive implementation of robust ICT structure. The front-end of this structure will be software applications that take care of variety of market parameters enabling high level services (e.g. stock exchange). The back-end of the structure will be represented by diverse embedded devices that handle direct measuring and monitoring and control of adequate parameters but also provide some local data storing, processing and transfer. In order to support bidirectional flows and increased generation attached at distribution level extensions of physical infrastructure (Power Flow layer in our model) are a must. But in this work we refrain from this topic, discussing only concepts at two upper levels. Above analysis shows that market actors, underlying ICT structures and power flows involve increasingly complex interactions due to growing number of players and requirements. Such systems require structuring and harmonization. A very simplified model (see Figure 3) represents this structure and in the sequel of the work we aim to relate these three basic components and their connecting points. ICT as an enabling technology for smart-grids plays a key role in implementing required functionalities. At one side it provides high-level applications used by Market actors, at other side it provides measuring, monitoring and control of field devices using embedded devices. Between those two ends, there is information flow determined by involved stakeholders and their needs. Determining adequate ICT structure that manages in an optimal way (considering balance between reliability and implementation cost) required information flows is a crucial task for successful realization of smart-grids. The design analysis starts from
3. Power system as layered structure
Fundamental role of future energy systems (i.e. smartgrids) is to leverage economical and environmental efficiency versus security and reliability of supply. Achieving of this aim will require support for bidirectional power flows, advanced pricing schemes, new business models, legislative adjustments and so forth, we refer here to all these characteristics as Market. Well established Market is a key to provide customer-centric power system that implements closer interaction with costumers through Demand Side Management (DMS).
3
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Market level
S W Applications
Financial, commercial and legislative framework, business models…
T structure EMS IC
S martMeter / IED
ICT
ICT technologies (information model/ flow, control… ) Physics of the system (generators, loads, storages, power-lines, transformers...)
Market
VPP Power flow
S mart Metering
VPP management Distribution Generation Storage
Figure 3. Power system - layered structure determination of stakeholders involved and their requirements. According to those needs an appropriate information flow is defined. This actually means what and when and to whom certain data is needed to be sent. Accordingly ,enabling technologies are then determinated in terms of adequate choice of: • Embedded devices - that are front line to power lines, generators, transformers and other electro-mechanic and physical elements of the grid • Communication infrastructure - that transfers right information in right time, in a reliable and secure manner • Control software - that manages monitoring and control of the grid in an optimal way according to obtained parameters from the grid and requirements from the market In Section 4 we give basis for a similar analysis of VPP concept with spacial focus on commercial aspects.
Distribution Network
Load
Figure 4. VPP - Context Diagram In this work we use VPP system, as an element of smartgrids, to show modeling strategy applicable for variety of similar power system concepts. Modeling methodology starts with Context Diagram, that represents in a very general manner main actors involved in certain concept (one such VPP context diagram is given in Figure 4). The next step consists of developing of an Abstract Model, that consists of different Use Cases that represent certain functionalities. Sequence and Class diagrams and derived from those use-cases and they are further refined in direction and to level of granularity that depends on the considered aspect of the concrete problem.
5
VPP as distributed generation integrator vs. smart-grid module
4
General abstract model of VPP
Smart-grids require modular and scalable approach to power system architecture. Distribution of monitoring and control operations from central point to local ”nodes” is one of instruments to cope with increased communicational and computational needs (see Figure 5) [13]. Physical decentralization of generation is followed by commercial paradigm shift. We can consider VPP as a mean to integrate heterogeneous distributed renewable resources into smart-grid framework in both technological and market/commercial terms. Align with its dual role, VPPs are considered as Technical (TVPP) and Commercial (CVPP) [12, 18], detailed explanations of these concepts are given in Sections 5.1.1 and 5.1.2. VPP could be an important link in achieving end-to-end interoperability. At one side it is directly attached to the rest of smart-grid and on the other side it directly manages field devices employed in the grid. An additional step in VPP development could be integration with Building Automation
Standards harmonization is one of main instruments to achieve full end-to-end interoperability. This process leads to unified instrument of representing power system elements and their interaction (see Figure 2). A widely accepted CIM standard [5, 7] aims at representing all software, hardware and electronic components in UML-like fashion and in such a way facilitate communication between applications in inter and intra utility information exchange. UML is de-facto industrial standard for system level modeling and it is also widely accept in modeling of different financial and commercial issues. These facts give a solid base to methodologies, based on this modeling language, to become a common understanding platform for communication among different smart-grid stakeholders (technical/financial/legal experts, authorities, entrepreneurs etc.). Representing entire power system in an unified yet simple manner could facilitate realization of advanced concepts by enabling efficient communication among involved players.
4
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
TODAY
TSO CC Transmission
FUTURE
Distribution
DSO Control Centre DSO Control Centre
SubStation Control Centre
SubStation Control Centre
Middle Voltage Low Voltage
CVPP/TVPP
VPP to show an abstract model of exact VPP functionality. Figure 6 presents, again in a very simple way considered aspect of VPP and actors involved in it. Entire abstract model would consists of use cases which represent all functionalities of the system. Further step in modeling strategy goes towards development of class and sequence diagrams. These two types of diagrams can fully describe the system from certain aspect showing exact components and their relationships. Prior to presenting concrete models some more detailed analysis of involved actors must be provided. All needed explanations are given in the following sections. 5.1.1 Commercial VPP functionalities
Figure 5. VPP inside smart-grid Systems which would lead to brining market capabilities from the market level directly to the home appliances. The structure of VPPs and exact roles of its components are given in following Sections.
5.1
VPP structure
As presented in Figure 4, we may refer to several groups of stakeholders involved in VPP management. It is important to stress that concrete realization of VPP differs from country to country, mostly due to different regulations, impacting type of stakeholders involved. For this reason we rather refer to groups of stakeholders leaving refinement in this sense to the concrete cases. Essentially, the VPP is represented by ICT layer over existing power system infrastructure (generators, storages, power-lines etc.). In order to better distinguish the role and functionalities of VPP, but also to better structure and encapsulate VPP’s functionalities, ICT structure of VPPs is separated in two complementary entities[18, 4]; TVPP (Technical Virtual Power Plant) ensures that the power system is operated in an optimized and secure way, taking physical constraints and potential services offered by VPP into account; CVPP (Commercial Virtual Power Plant) considers DERs as commercial entities offering the price and amount of energy that it can deliver, optimizing economical utilization of VPP portfolio for the open electricity market [15]. In other words, the purpose of CVPP is to, taking into account all the energy offers and needs, schedules an optimized DER’s utilization. The operational employment of particular DERs in technical terms is managed by the TVPP. So that, in terms of service layers, TVPP is a lower layer providing services to the CVPP as a higher layer. After defining context diagram of VPP (given in Figure 4), the very next step in our modeling methodology consist of describing certain aspect of the concrete context (in this case VPP management). We take commercial operation of
Since different VPPs can be legally owned by different legacy entities, energy trading and price evaluating becomes one of the main concerns of future energy system. Pricedriven market-based approach is used for this purpose, guaranteeing fairness and optimal energy usage depending on the balance of current energy demands and production capabilities. For purpose of encapsulation of market requirements of VPP, CVPP is introduced as an aggregation of capacities of DER units, that optimizes production with actual needs and schedules DGs production (see Figure 7). In achieving its goal CVVP directly or indirectly interacts with following entities [12] (general abstract model of this interaction is given at use case in Figure 6): • DER (Distributed Energy Resource) is considered as balancing responsible resource that is obligatory to plan its production and provide that information to the TVPP • BRP (Balance Responsible Party) represent an energy trading entity with a property to make own production/consumption plan available to be used by the TVPP (in some cases Trading Agent) [21] • TSO (Transmission System Operator) has a main role in maintaining the instantaneous demand and supply balance in the network. • TVPP which role will be detailed in Section 5.1.2 Basic CVPP functionalities would be optimization and scheduling of production based on predicted needs of consumers. In case that actual needs differ from predicted ones, DRR (Demand Response Resources) are introduced to fulfill the gap between production and real consumption. In general, CVPP functions should also include: • Maintenance and submission of DERs’ characteristics and costs
5
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Figure 7. VPP commercial operation - class diagram Figure 6. VPP commercial operation - abstract model (use case) will be involved in production. In case that given plan is infeasible, TVPP reports an error and new optimization has to be done. Both diagrams are simplified and they are used just as a sketch to show the conceptual approach of modeling methodology. Further refinement of developed models goes in direction of modeling of each of entity (actually aspects) in provided class model. Considering model represented in Figure 7, we have chosen AMI (Advanced Metering Infrastructure) as a case study for system level model (see Figure 9) of concrete component of power system, in this case metering structure consisting of embedded devices that are one of basic VPP enabling technologies. This model has an aim to show in finer granularity one certain component/aspect of the system. AMI is not directly involved in VPP commercial operation so it is not presented as an actor in this use case but it is used by CVPP and in such way indirectly takes part in this process. In Section 6 we give description of metering infrastructure employed in VPP and its appropriate model in greater detail.
• Production and consumption forecasting based on weather forecasting and demand profiles • ODM (Outage Demand management) • Building DER bids and submitting them to the market • Daily optimization and generation scheduling • Selling energy provider by DERs to the Market An UML class diagram in Figure 7 shows entities involved (and their relations) in wider context of commercial operation of VPP, it actually relates CVPP and TVPP with other power system components. This diagram is conceived as conceptual and illustrative; it does not propose any implementable solutions. Moreover, for the sake of simplicity but without losing generality, it does not include all the functions that CVPP should provide. It is aimed to be rather general concept, and graphical representation that helps in better understanding certain CVPP’s roles. The real goal is to show applicability and usability of UML models as an instrument in describing such concepts. Furthermore, the sequence diagram shown in Figure 8 serves as an example of an elementary scenario of production optimization and scheduling according to [15]. In order to do to negotiation on the network and to optimize production, CVPP first gathers required consumption and production capacities information. This is done in two separate iterations, one for gathering consumption information from each BRP consumer and other one to gather production capacity information from each DER. Since both, BRP consumers and DERs are connected to the TVPP request, getting the information goes through the TVPP. After negotiation CVPP gives production plan to TVPP that, knowing network topology does production optimization and decides which particular DER
5.1.2
Technical VPP functionalities
TVPP takes care of correct operation of the DER according to requirements obtained from CVPP and system status information. It provides management of the network and execution of ancillary services. It operates using information related to the network topology, operating parameters and so forth. Some of functionalities of Energy Management System (EMS) like ’load management’ and ’load shedding’ are performed by TVPP. As this work focuses more on impact of the market at the VPP we will not go in greater detail of TVPP.
6
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Figure 8. VPP commercial operation - Elementary Sequential Diagram
6
Case study - System level model of smartmetering devices employed in VPP
In this section we show, using UML class model, functionalities of smart devices deployed in VPP and their integration with higher level services (like CVPP, Trading Agent etc.). Smart-meter’s advanced functions such as real-time (or near real-time) sensing, power outage notification, power frequency and VAr (reactive power) monitoring and other features will be provided by Advanced Metering Infrastructure (AMI). The underlying monitoring and control structure is the core to communication, processing, actuation etc. [11]. Communication of smart-meters among themselves as well as with central AMI unit could be realized through various technologies. While PLC (Power Line Communication) provides reliable and secured but in general slow communication, GPRS provides less reliability but greater data throughput. Other solutions like, private (even Ad-Hoc) wireless networks and radio frequency are also possible. All the data gathered in the system are processed and possibly stored in AMI central unit (and units directly attached to it) that is responsible for energy flow monitoring and also control (using active Smart-meters) [1]. Minimal requirement that AMI should fulfill are remote measurements, disconnection and reconnection of particular users [2]. Ultimate goal is connecting the AMI with the Market. At the other side active Smart-meters could also be able of (remote) control/actuation (trough the Smart House Controller) of all the smart house appliances.
Figure 9. Metering Infrastructure - Class Diagram
7
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
According to previously exposed analysis of smart-grids evolution in terms of requirements for future technologies to be employed to sustain these needs, we provide in Figure 9 a simplified system level model of metering devices to support such trends. This model also shows in which way these embedded systems are integrated with higher level applications. In such a way, using UML models refinement we show reflection of the high level functional requirements (in this case Market) through the system down there to the concrete field devices employed.
7
Conclusions and Future Work
Energy market is experiencing fast evolution from centralized to deregulated structures. Along with massive insertion of distributed renewable resources it poses tremendous challenges to future power systems. A VPP concept emerges to be a promising solution to integrate such DERs and interface them as a kind of cluster inside smart-grid. We show place and role of VPP inside given power system structure. At the same time we use VPP structure to demonstrate modeling methodology by which we relate widest context of the problem from high level market applications down to field devices. Using set of UML diagrams appropriate aspects of VPP management are presented at different levels of abstraction by different types of modeling instruments like use cases, abstract model, class and sequence diagrams. Eventually, by developing a UML class model of metering infrastructure as a case study, we show an example of reflection of Market applications at enabling technologies. One of aims of this work is to advocate wider use of UML as common understanding platform among different stakeholders. Together with further standardization in this field, a dedicated UML standards and design patterns dedicated to smart-grids should be developed. This will, even more, improve understanding and easier communication between researchers coming from different areas. Adopted market regulation and business models will greatly impact role and design of smart-devices. More advanced modeling methodologies will enable full top-down representation of power system which could reveal potential conflicts between higher level market application requirements and low level technological capabilities.
References
[1] [2] [3] [4] http://www.aeris.net. http://www.dpi.vic.gov.au/energy. http://www.siemens.com. I. Bel, A. Velenti, J. Maire, J. M. Corera, and P. Lang. Innovative operation with aggregated distributed generation. In 15th Power Systems Computation Conference PSCC, Wien, 2007.
[5] J. P. Britton and A. N. deVos. CIM-based standards and CIM evolution. Power Systems, IEEE Transactions on, 20(5):10– 16, May 2005. [6] L. B. Cardell. Integrating small scale distributed generation into a deregulated market : control strategies and price feedback. Massachusetts Institute of Technology, PhD Thesis, 1997. [7] B. Chadha, R. E. Fulton, and J. C. Calhoun. Design and implementation of a CIM information system. Engineering with Computers, May 2005. [8] J. F. R. Goodman. Utility Communications Architecture (UCA) Object Models for Distributed Energy Resources (UCA-DER). Electricity Innovation Institute and Electric Power Research Institute, Inc., 2003. [9] R. Gilbert, J. Goodin, and K. Nelson. The benefits of a distribution automation system using the utility communications architecture (UCA). In Rural Electric Power Conference, 1998. Papers Presented at the 42nd Annual Conference, 1998. [10] J. Gordijn and H. Akkermans. Business models for distributed generation in a liberalized market environment. Electric Power Systems Research, 77(9), September 2006. [11] D. Hart. Using AMI to realize the Smart Grid. In Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century, IEEE, Wien, 2008. [12] http://www.fenix-project.org/. Fenix Project Bulletin, 2008. [13] http://www.smartgrids.eu/. [14] S. Keemink and B. Roos. Security analysis of Dutch smart metering systems. Universiteit van Amsterdam, 2008. [15] K. Kok, Z. Derzsi, J. Gordijn, M. Hommelberg, C. Warmer, R. Kamphuis, and H. Akkermans. Agent-based electricity balancing with distributed energy resources, a multiperspective case study. In HICSS ’08: Proceedings of the 41st Annual Hawaii International Conference on System Sciences, Washington, DC, USA, 2008. IEEE Computer Society. [16] D. J. Marihart. Communications technology guidelines for ems/scada systems. IEEE Transactions on Power, 16(2), April 2001. [17] S. Massoud Amin and B. F. Wollenberg. Toward a smart grid: power delivery for the 21st century. IEEE Power and Energy Magazine, 3(5), October 2005. [18] D. Pudjianto, C. Ramsay, and G. Strbac. Virtual power plant and system integration of distributed energy resources. Renewable Power Generation, IET, 1(1):10–16, March 2007. [19] K. Schwarz. IEC 68150, IEC 61400-25, and IEC 61970: Information model and information exchange for electric power systems. In Distributech, 2004. [20] K. Schwarz. IEC 61850 also outside the substation for the whole electrical power system. In 15th Power Systems Computation Conference PSCC, Liege, Belgium, DC, USA, 2005. [21] C. Warmer, M. Hommelberg, J. Kok, and I. Kamphuis. Local DER driven grid support by coordinated operation of devices. In Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century. IEEE, 2008. [22] B. Willems. Physical and financial virtual power plants. Discussion Paper, 5(1):1–12, March 2007.
8
Virtual Power Plant as a bridge between Distributed Energy Resources and Smart Grid
Slobodan Lukovic, Igor Kaitovic, Marcello Mura, Umberto Bondi Faculty of Informatics - ALaRI University of Lugano Lugano, Switzerland {lukovics, muram, bondi}@alari.ch, [email protected] Abstract
The liberalization of energy markets, especially in correlation with the Smart Grid concept development, requires adjusted legislation, new business models, energy stock exchanges establishment and many other advanced instruments. Realization of these features necessitates novel concepts to support such changes in the power system while granting security and reliability of supply. Such evolution poses new challenges to ICT (Information and Communication Technologies) to bridge the gap between increased complexity of deregulated market and on the other side expected rapid growth of number of players in power systems. Increasing presence of Distributed Energy Resources (DER) implementations constitutes a further source of complexity. Bearing in mind ongoing and possible scenarios we aim to determinate the place and role of the novel Virtual Power Plants (VPP) concept, related to the Smart Grid structure. At the same time we introduce an innovative modeling approach as an instrument to determine actors and highlight their actual roles and interactions from different point of view, trying to pave the way for development of a common understanding platform for variety of stakeholders. The effectiveness of the proposed modeling concept is shown through a number of UML models representing system level description of VPP at different levels of abstraction.
eralization and decentralization are effecting and changing traditional production, distribution and consumption patterns. On the other hand the power system as a complex and sensitive entity needs modular integration of reliable components. The Smart Grid [13, 17] concept appears to be a promising response to these needs. It is fairly mature strategy that aims at leveraging energy efficiency management using the most advanced ICT, and it concerns power systems at all levels. There are many issues to be tackled by future Smart Grids; in the presented paper we focus at, and try to relate two of them: • Efficient integration of Distributed Energy Resources (DER) in an open energy market • Development of a common understanding platform based on unified modeling concept Market deregulation aims at involving a greater number of players in energy generation, transmission and distribution, and this trend results in decentralization of power flows’ control. The increased utilization of distributed generation (mostly based on renewable resources) and its efficient integration in the electricity system is considered to be a technological driver for stable, secure and sustainable energy supply in the future. On the other side, achieving efficient control over greater number of generators (especially those with low controllability) and other accompanying elements coupled with ’intelligent devices’ deployed in each of them, would result in overwhelming data generation, transmission and processing. At the same time, different Market applications also need large amounts of data for a variety of purposes (predictions, real-time pricing, stock exchange etc). All this data are eventually coming from thousands or even millions of field devices (i.e. embedded systems) employed to measure, monitor and control generators, substation elements, loads etc.
1. Introduction
Awareness to global climate changes caused by different kinds of pollution on the one side and increasing concern about sustainable energy supply at the other side have created a global sensitivity to energy production as well as an interest in finding pollution-free and sustainable solutions. The energy market is rapidly evolving and processes of lib-
978-0-7695-3869-3/10 $26.00 © 2010 IEEE
1
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Keeping supply secure and reliable in such conditions, in particular considering distributed generation, requires improved communication among all players in the system in order to match market requirements with technological capabilities of underlying structures. In order to avoid possible incompatibility between market requirements and technical feasibility, we need novel concepts and approaches to provide a common language between the different actors and thus bridge market (financial, legal, commercial etc. issues) needs and technical capabilities. Therefore, in order to successfully answer aforementioned challenges (coming from new energy market requirements and security supply must) Smart Grid has to give a response to huge communicational and computational requirements. A possible solution could be in clustering of certain power system elements in terms of their geographical, technological and/or commercial characteristics. A Virtual Power Plant (VPP) [18, 4] as a concept to aggregate distributed resources (including also storages and consumers/prosumers) and present them to the rest of the energy system as a single technical and commercial entity can be seen as a solution for this need. Such a module/cluster inside a wider Smart Grid environment should have certain autonomy and it would represent a kind of distributed computing but also trading node/unit. Being remarkably interdisciplinary topic, VPP requires holistic approach that highlights this problem from many different sides. In order to relate variety of aspects of this concept we introduce modeling methodology. So, we use VPP to show solutions for DER integration in power systems, and in particular we focus on reflection of future market requirements at such concept. We aim at determining VPP functional requirements posed by market needs and their impact on enabling ICT technologies. Moreover, in a case study of metering infrastructure we show, using again modeling approach as a an instrument, how commercial/market requirements impact embedded devices evolution as one of key actors to enable efficient execution of higher level applications. As an demonstration instrument we introduce and at the same time promote for wider acceptance - UML (Unified Modeling Language) based modeling approach. The choice of UML seams logical as it is established as industrial de-facto modeling standard; it is also widely accepted in commercial and finance community and through Common Information Model (CIM) it is already introduced to many experts in power system field. The paper is organized as follows: Section 2, after reviewing some relevant work on power system evolution we provide an overview of ongoing research on VPP. In Section 3, a conceptual, general layered structure of energy system is presented and explained. Section 4 details relevance and potential of system-level modeling approach. In Section 5 we provide deeper insight on VPP structure with
Business models
INPUTS:
Economic Dispatch (ED) Load Shedding (LS)
EMS
Voltage Var Control (VVC)
text text
Automatic Generation Control (AGC)
Load Management (LM)
Production costs
PC
System parameters status
Set points
RTU
for generator
SCADA
PC
RTU
for consumer
Energy storage
RTU
for generator
RTU
for generator
RTU
for generator
RTU
for generator
Figure 1. Traditional power system structure special regard to commercial issues, while Section 6 gives a general system level description of future smart-metering devices as case study for a certain group of intelligent devices involved in smart-grid concept. Finally, Section 7 presents conclusions and future work.
2. State of the art
Energy market is evolving from centralized to liberalized one [10]. This shift impacts entire power system and it is naturally followed by a number of open issues that are pending resolution. One of main instruments to cope with such challenges is underlying ICT system. ICT structure in contemporary energy systems is built to support centralized generation based on generationtransmission-distribution-consumption concept. The communication between software control centers and generation units, substation elements etc. is performed through integration of Remote Terminal Units (RTUs) and Intelligent Electronic Devices (IEDs) over SCADA system. On top of SCADA software extension known as Energy Management System (EMS) is managing the optimizing grid utilization in a centralized fashion (as shown in Figure 1) [16, 9]. Changes in this paradigm are introduced by several factors, among which we may mention: • Production is coming closer to the end consumers (a trend imposed by increased distributed generation) • Massive insertion of heterogeneous, non-controllable renewable resources
2
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
• Growing environmental concerns that call for higher energy efficiency • Introduction of new business models • Increased need for Demand Side Management, real time pricing etc. The smart-grid [13, 17] emerge as a promising concept to tackle these trends. Its development and realization attracts huge research efforts. A fairly large number of issues ranging from standardization to concrete solutions for secure and reliable integration of distributed generation [6] have emerged to be a fundamental concern for both academia and industry. Researchers are addressing smart-grid issues from different aspects; seamless integration of distributed, mostly renewable resources is one of key issues. A novel concept to answer this challenge is Virtual Power Plant (VPP) [18, 22, 4]. Different issues are considered in VPP (each of them involves ICT in some way): • Distributed Energy Management Systems (DEMS) [16, 3] • VPP as trading entity [12] • Interoperability [19] • Security issues [14] • Communication interfaces [8] • Different implementation possibilities [12] In order to develop fully modular and scalable power system absolute standard harmonization must be achieved. This problem is addressed by many industrial and academic institutions and huge progress has been done in this field. Standardization of information models and communication protocols of data and control flow going from field devices up to high level Control Center (CC) applications has been well established (refer to Figure 2) [20, 8] .
Figure 2. Standards harmonization Implementation of such advanced features (as real-time pricing, DMS etc.) will require increased control over the grid and tight interaction with all actors in energy market (from energy stocks and utilities to the end-consumer) which further implies massive implementation of robust ICT structure. The front-end of this structure will be software applications that take care of variety of market parameters enabling high level services (e.g. stock exchange). The back-end of the structure will be represented by diverse embedded devices that handle direct measuring and monitoring and control of adequate parameters but also provide some local data storing, processing and transfer. In order to support bidirectional flows and increased generation attached at distribution level extensions of physical infrastructure (Power Flow layer in our model) are a must. But in this work we refrain from this topic, discussing only concepts at two upper levels. Above analysis shows that market actors, underlying ICT structures and power flows involve increasingly complex interactions due to growing number of players and requirements. Such systems require structuring and harmonization. A very simplified model (see Figure 3) represents this structure and in the sequel of the work we aim to relate these three basic components and their connecting points. ICT as an enabling technology for smart-grids plays a key role in implementing required functionalities. At one side it provides high-level applications used by Market actors, at other side it provides measuring, monitoring and control of field devices using embedded devices. Between those two ends, there is information flow determined by involved stakeholders and their needs. Determining adequate ICT structure that manages in an optimal way (considering balance between reliability and implementation cost) required information flows is a crucial task for successful realization of smart-grids. The design analysis starts from
3. Power system as layered structure
Fundamental role of future energy systems (i.e. smartgrids) is to leverage economical and environmental efficiency versus security and reliability of supply. Achieving of this aim will require support for bidirectional power flows, advanced pricing schemes, new business models, legislative adjustments and so forth, we refer here to all these characteristics as Market. Well established Market is a key to provide customer-centric power system that implements closer interaction with costumers through Demand Side Management (DMS).
3
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Market level
S W Applications
Financial, commercial and legislative framework, business models…
T structure EMS IC
S martMeter / IED
ICT
ICT technologies (information model/ flow, control… ) Physics of the system (generators, loads, storages, power-lines, transformers...)
Market
VPP Power flow
S mart Metering
VPP management Distribution Generation Storage
Figure 3. Power system - layered structure determination of stakeholders involved and their requirements. According to those needs an appropriate information flow is defined. This actually means what and when and to whom certain data is needed to be sent. Accordingly ,enabling technologies are then determinated in terms of adequate choice of: • Embedded devices - that are front line to power lines, generators, transformers and other electro-mechanic and physical elements of the grid • Communication infrastructure - that transfers right information in right time, in a reliable and secure manner • Control software - that manages monitoring and control of the grid in an optimal way according to obtained parameters from the grid and requirements from the market In Section 4 we give basis for a similar analysis of VPP concept with spacial focus on commercial aspects.
Distribution Network
Load
Figure 4. VPP - Context Diagram In this work we use VPP system, as an element of smartgrids, to show modeling strategy applicable for variety of similar power system concepts. Modeling methodology starts with Context Diagram, that represents in a very general manner main actors involved in certain concept (one such VPP context diagram is given in Figure 4). The next step consists of developing of an Abstract Model, that consists of different Use Cases that represent certain functionalities. Sequence and Class diagrams and derived from those use-cases and they are further refined in direction and to level of granularity that depends on the considered aspect of the concrete problem.
5
VPP as distributed generation integrator vs. smart-grid module
4
General abstract model of VPP
Smart-grids require modular and scalable approach to power system architecture. Distribution of monitoring and control operations from central point to local ”nodes” is one of instruments to cope with increased communicational and computational needs (see Figure 5) [13]. Physical decentralization of generation is followed by commercial paradigm shift. We can consider VPP as a mean to integrate heterogeneous distributed renewable resources into smart-grid framework in both technological and market/commercial terms. Align with its dual role, VPPs are considered as Technical (TVPP) and Commercial (CVPP) [12, 18], detailed explanations of these concepts are given in Sections 5.1.1 and 5.1.2. VPP could be an important link in achieving end-to-end interoperability. At one side it is directly attached to the rest of smart-grid and on the other side it directly manages field devices employed in the grid. An additional step in VPP development could be integration with Building Automation
Standards harmonization is one of main instruments to achieve full end-to-end interoperability. This process leads to unified instrument of representing power system elements and their interaction (see Figure 2). A widely accepted CIM standard [5, 7] aims at representing all software, hardware and electronic components in UML-like fashion and in such a way facilitate communication between applications in inter and intra utility information exchange. UML is de-facto industrial standard for system level modeling and it is also widely accept in modeling of different financial and commercial issues. These facts give a solid base to methodologies, based on this modeling language, to become a common understanding platform for communication among different smart-grid stakeholders (technical/financial/legal experts, authorities, entrepreneurs etc.). Representing entire power system in an unified yet simple manner could facilitate realization of advanced concepts by enabling efficient communication among involved players.
4
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
TODAY
TSO CC Transmission
FUTURE
Distribution
DSO Control Centre DSO Control Centre
SubStation Control Centre
SubStation Control Centre
Middle Voltage Low Voltage
CVPP/TVPP
VPP to show an abstract model of exact VPP functionality. Figure 6 presents, again in a very simple way considered aspect of VPP and actors involved in it. Entire abstract model would consists of use cases which represent all functionalities of the system. Further step in modeling strategy goes towards development of class and sequence diagrams. These two types of diagrams can fully describe the system from certain aspect showing exact components and their relationships. Prior to presenting concrete models some more detailed analysis of involved actors must be provided. All needed explanations are given in the following sections. 5.1.1 Commercial VPP functionalities
Figure 5. VPP inside smart-grid Systems which would lead to brining market capabilities from the market level directly to the home appliances. The structure of VPPs and exact roles of its components are given in following Sections.
5.1
VPP structure
As presented in Figure 4, we may refer to several groups of stakeholders involved in VPP management. It is important to stress that concrete realization of VPP differs from country to country, mostly due to different regulations, impacting type of stakeholders involved. For this reason we rather refer to groups of stakeholders leaving refinement in this sense to the concrete cases. Essentially, the VPP is represented by ICT layer over existing power system infrastructure (generators, storages, power-lines etc.). In order to better distinguish the role and functionalities of VPP, but also to better structure and encapsulate VPP’s functionalities, ICT structure of VPPs is separated in two complementary entities[18, 4]; TVPP (Technical Virtual Power Plant) ensures that the power system is operated in an optimized and secure way, taking physical constraints and potential services offered by VPP into account; CVPP (Commercial Virtual Power Plant) considers DERs as commercial entities offering the price and amount of energy that it can deliver, optimizing economical utilization of VPP portfolio for the open electricity market [15]. In other words, the purpose of CVPP is to, taking into account all the energy offers and needs, schedules an optimized DER’s utilization. The operational employment of particular DERs in technical terms is managed by the TVPP. So that, in terms of service layers, TVPP is a lower layer providing services to the CVPP as a higher layer. After defining context diagram of VPP (given in Figure 4), the very next step in our modeling methodology consist of describing certain aspect of the concrete context (in this case VPP management). We take commercial operation of
Since different VPPs can be legally owned by different legacy entities, energy trading and price evaluating becomes one of the main concerns of future energy system. Pricedriven market-based approach is used for this purpose, guaranteeing fairness and optimal energy usage depending on the balance of current energy demands and production capabilities. For purpose of encapsulation of market requirements of VPP, CVPP is introduced as an aggregation of capacities of DER units, that optimizes production with actual needs and schedules DGs production (see Figure 7). In achieving its goal CVVP directly or indirectly interacts with following entities [12] (general abstract model of this interaction is given at use case in Figure 6): • DER (Distributed Energy Resource) is considered as balancing responsible resource that is obligatory to plan its production and provide that information to the TVPP • BRP (Balance Responsible Party) represent an energy trading entity with a property to make own production/consumption plan available to be used by the TVPP (in some cases Trading Agent) [21] • TSO (Transmission System Operator) has a main role in maintaining the instantaneous demand and supply balance in the network. • TVPP which role will be detailed in Section 5.1.2 Basic CVPP functionalities would be optimization and scheduling of production based on predicted needs of consumers. In case that actual needs differ from predicted ones, DRR (Demand Response Resources) are introduced to fulfill the gap between production and real consumption. In general, CVPP functions should also include: • Maintenance and submission of DERs’ characteristics and costs
5
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Figure 7. VPP commercial operation - class diagram Figure 6. VPP commercial operation - abstract model (use case) will be involved in production. In case that given plan is infeasible, TVPP reports an error and new optimization has to be done. Both diagrams are simplified and they are used just as a sketch to show the conceptual approach of modeling methodology. Further refinement of developed models goes in direction of modeling of each of entity (actually aspects) in provided class model. Considering model represented in Figure 7, we have chosen AMI (Advanced Metering Infrastructure) as a case study for system level model (see Figure 9) of concrete component of power system, in this case metering structure consisting of embedded devices that are one of basic VPP enabling technologies. This model has an aim to show in finer granularity one certain component/aspect of the system. AMI is not directly involved in VPP commercial operation so it is not presented as an actor in this use case but it is used by CVPP and in such way indirectly takes part in this process. In Section 6 we give description of metering infrastructure employed in VPP and its appropriate model in greater detail.
• Production and consumption forecasting based on weather forecasting and demand profiles • ODM (Outage Demand management) • Building DER bids and submitting them to the market • Daily optimization and generation scheduling • Selling energy provider by DERs to the Market An UML class diagram in Figure 7 shows entities involved (and their relations) in wider context of commercial operation of VPP, it actually relates CVPP and TVPP with other power system components. This diagram is conceived as conceptual and illustrative; it does not propose any implementable solutions. Moreover, for the sake of simplicity but without losing generality, it does not include all the functions that CVPP should provide. It is aimed to be rather general concept, and graphical representation that helps in better understanding certain CVPP’s roles. The real goal is to show applicability and usability of UML models as an instrument in describing such concepts. Furthermore, the sequence diagram shown in Figure 8 serves as an example of an elementary scenario of production optimization and scheduling according to [15]. In order to do to negotiation on the network and to optimize production, CVPP first gathers required consumption and production capacities information. This is done in two separate iterations, one for gathering consumption information from each BRP consumer and other one to gather production capacity information from each DER. Since both, BRP consumers and DERs are connected to the TVPP request, getting the information goes through the TVPP. After negotiation CVPP gives production plan to TVPP that, knowing network topology does production optimization and decides which particular DER
5.1.2
Technical VPP functionalities
TVPP takes care of correct operation of the DER according to requirements obtained from CVPP and system status information. It provides management of the network and execution of ancillary services. It operates using information related to the network topology, operating parameters and so forth. Some of functionalities of Energy Management System (EMS) like ’load management’ and ’load shedding’ are performed by TVPP. As this work focuses more on impact of the market at the VPP we will not go in greater detail of TVPP.
6
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
Figure 8. VPP commercial operation - Elementary Sequential Diagram
6
Case study - System level model of smartmetering devices employed in VPP
In this section we show, using UML class model, functionalities of smart devices deployed in VPP and their integration with higher level services (like CVPP, Trading Agent etc.). Smart-meter’s advanced functions such as real-time (or near real-time) sensing, power outage notification, power frequency and VAr (reactive power) monitoring and other features will be provided by Advanced Metering Infrastructure (AMI). The underlying monitoring and control structure is the core to communication, processing, actuation etc. [11]. Communication of smart-meters among themselves as well as with central AMI unit could be realized through various technologies. While PLC (Power Line Communication) provides reliable and secured but in general slow communication, GPRS provides less reliability but greater data throughput. Other solutions like, private (even Ad-Hoc) wireless networks and radio frequency are also possible. All the data gathered in the system are processed and possibly stored in AMI central unit (and units directly attached to it) that is responsible for energy flow monitoring and also control (using active Smart-meters) [1]. Minimal requirement that AMI should fulfill are remote measurements, disconnection and reconnection of particular users [2]. Ultimate goal is connecting the AMI with the Market. At the other side active Smart-meters could also be able of (remote) control/actuation (trough the Smart House Controller) of all the smart house appliances.
Figure 9. Metering Infrastructure - Class Diagram
7
Proceedings of the 43rd Hawaii International Conference on System Sciences - 2010
According to previously exposed analysis of smart-grids evolution in terms of requirements for future technologies to be employed to sustain these needs, we provide in Figure 9 a simplified system level model of metering devices to support such trends. This model also shows in which way these embedded systems are integrated with higher level applications. In such a way, using UML models refinement we show reflection of the high level functional requirements (in this case Market) through the system down there to the concrete field devices employed.
7
Conclusions and Future Work
Energy market is experiencing fast evolution from centralized to deregulated structures. Along with massive insertion of distributed renewable resources it poses tremendous challenges to future power systems. A VPP concept emerges to be a promising solution to integrate such DERs and interface them as a kind of cluster inside smart-grid. We show place and role of VPP inside given power system structure. At the same time we use VPP structure to demonstrate modeling methodology by which we relate widest context of the problem from high level market applications down to field devices. Using set of UML diagrams appropriate aspects of VPP management are presented at different levels of abstraction by different types of modeling instruments like use cases, abstract model, class and sequence diagrams. Eventually, by developing a UML class model of metering infrastructure as a case study, we show an example of reflection of Market applications at enabling technologies. One of aims of this work is to advocate wider use of UML as common understanding platform among different stakeholders. Together with further standardization in this field, a dedicated UML standards and design patterns dedicated to smart-grids should be developed. This will, even more, improve understanding and easier communication between researchers coming from different areas. Adopted market regulation and business models will greatly impact role and design of smart-devices. More advanced modeling methodologies will enable full top-down representation of power system which could reveal potential conflicts between higher level market application requirements and low level technological capabilities.
References
[1] [2] [3] [4] http://www.aeris.net. http://www.dpi.vic.gov.au/energy. http://www.siemens.com. I. Bel, A. Velenti, J. Maire, J. M. Corera, and P. Lang. Innovative operation with aggregated distributed generation. In 15th Power Systems Computation Conference PSCC, Wien, 2007.
[5] J. P. Britton and A. N. deVos. CIM-based standards and CIM evolution. Power Systems, IEEE Transactions on, 20(5):10– 16, May 2005. [6] L. B. Cardell. Integrating small scale distributed generation into a deregulated market : control strategies and price feedback. Massachusetts Institute of Technology, PhD Thesis, 1997. [7] B. Chadha, R. E. Fulton, and J. C. Calhoun. Design and implementation of a CIM information system. Engineering with Computers, May 2005. [8] J. F. R. Goodman. Utility Communications Architecture (UCA) Object Models for Distributed Energy Resources (UCA-DER). Electricity Innovation Institute and Electric Power Research Institute, Inc., 2003. [9] R. Gilbert, J. Goodin, and K. Nelson. The benefits of a distribution automation system using the utility communications architecture (UCA). In Rural Electric Power Conference, 1998. Papers Presented at the 42nd Annual Conference, 1998. [10] J. Gordijn and H. Akkermans. Business models for distributed generation in a liberalized market environment. Electric Power Systems Research, 77(9), September 2006. [11] D. Hart. Using AMI to realize the Smart Grid. In Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century, IEEE, Wien, 2008. [12] http://www.fenix-project.org/. Fenix Project Bulletin, 2008. [13] http://www.smartgrids.eu/. [14] S. Keemink and B. Roos. Security analysis of Dutch smart metering systems. Universiteit van Amsterdam, 2008. [15] K. Kok, Z. Derzsi, J. Gordijn, M. Hommelberg, C. Warmer, R. Kamphuis, and H. Akkermans. Agent-based electricity balancing with distributed energy resources, a multiperspective case study. In HICSS ’08: Proceedings of the 41st Annual Hawaii International Conference on System Sciences, Washington, DC, USA, 2008. IEEE Computer Society. [16] D. J. Marihart. Communications technology guidelines for ems/scada systems. IEEE Transactions on Power, 16(2), April 2001. [17] S. Massoud Amin and B. F. Wollenberg. Toward a smart grid: power delivery for the 21st century. IEEE Power and Energy Magazine, 3(5), October 2005. [18] D. Pudjianto, C. Ramsay, and G. Strbac. Virtual power plant and system integration of distributed energy resources. Renewable Power Generation, IET, 1(1):10–16, March 2007. [19] K. Schwarz. IEC 68150, IEC 61400-25, and IEC 61970: Information model and information exchange for electric power systems. In Distributech, 2004. [20] K. Schwarz. IEC 61850 also outside the substation for the whole electrical power system. In 15th Power Systems Computation Conference PSCC, Liege, Belgium, DC, USA, 2005. [21] C. Warmer, M. Hommelberg, J. Kok, and I. Kamphuis. Local DER driven grid support by coordinated operation of devices. In Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century. IEEE, 2008. [22] B. Willems. Physical and financial virtual power plants. Discussion Paper, 5(1):1–12, March 2007.
8
