Sie sind auf Seite 1von 10

Smart grid power system control

in distributed generation environment

Pertti Jrventausta*, Sami Repo*, Antti Rautiainen*
Jarmo Partanen**

*Tampere University of Technology, P.O.Box 692, 33101 Tampere,

**Lappeenranta University of Technology, P.O.Box 20, 53851 Lappeenranta

Abstract: This paper discusses on general aspects of smart grids and focuses on some smart grid features
at distribution level like interconnection of distributed generation and active distribution management,
using automated meter reading (AMR) systems in network management and power quality monitoring,
application of power electronics in electricity distribution, plug-in vehicles as part of smart grids, and
frequency based load control as examples of interactive customer gateway.
Keywords: smart grid, electricity distribution, distributed generation, active distribution management,
automated meter reading, plug-in vehicles, frequency based load control

1. INTRODUCTION There are much research, and many visions and concepts for
future power delivery system, like super grid, smart grid,
The energy markets are in transition and there are many micro grid, intelligent grid, active network, power cell etc.
drivers for creating a new kind of power delivery system for Some of them focus on transmission level functions (e.g.
the future. There are many drives and needs as follows: integration of large-scale wind power or utilization of
The penetration of distributed generation (DG), especially FACTS -devices) as some cover low voltage level and
based on Renewable Energy Sources (RES), will continue customer interface (e.g. large-scale advanced AMR). The
due to environmental reasons. concepts have many common features but also some
The European and North American vision is to have differences, but they are not analysed more deeply here. The
common electricity market areas with a high penetration main aims to fulfil the above needs are still same. A shared
of distributed power generation. vision of smart grid, or corresponding concept, is an
Efficient use of energy at customer level and intelligent important issue in order to develop commercially successful
demand response has become an essential issue. and useful products for future power delivery system. This
Power quality (supply reliability and voltage quality) vision should be shared by network companies, product
requirements will increase due to public and regulatory vendors and network customers.
actions and at the same time failure rates are expected This paper focuses mainly on general aspects of smart grids
increase due to the climate change. at distribution level and gives some examples on studied and
There is a need, due to economical reasons, to increase developed smart grid features.
the utilization rate of existing network. The traditional
way of developing a distribution network would be the 2. CHALLENGES FOR DISTRIBUTION SYSTEM
investment on passive wires which would lead to
decrement of utilization rate. Electricity distribution networks create a market place for
Many components of existing networks are becoming into small-scale power producers (i.e distributed generation) and
end of their lifetime. They need replacement or for customers (i.e users of electricity). Here, the role of
continuation of their lifetime in safe and controlled way. distribution networks is of great significance. For example in
Regulation of network companies will tighten up while Finland, about a half of the total price of electricity for small
companies want to ensure profitability of their business. customers and over 90 % of all interruptions come from the
This will mean rationalization of network management distribution process. There are many challenges for
both in short- and in long-term perspective. distribution system to enhance its functionality as the real
The risk of major disturbances is increasing, both the market place, as follows:
probability and consequences. The reason for increased improving the capability to serve the increasing amount
probability is the complexity of power network and the of distributed generation,
increased failure rate due to climate change. The enabling the electricity market development at the
consequences are increasing due to societys higher customer level e.g. for enhancing market-based demand
dependency on the power supply. response and customer-oriented services,
safe and cost-efficient operation of distribution networks Smart grids can be characterized as follows (partly based on
in all circumstances. Chuang 2008):
interactive with consumers and markets
Traditionally power generation, distribution network adaptive and scalable to changing situations
management and loads have been considered as quite optimized to make the best use of resources and
independent processes. Along with increasing amount of equipment
distributed generation the traditional approach is being
proactive rather than reactive, to prevent emergencies
gradually changing. Considerable amount of renewable
self-healing grids with high level of automation
energy resources represents distributed generation, but also
active energy resources like loads, storages and plug-in integrated, merging monitoring, control, protection,
hybrid vehicles will be increased. One of the main barriers maintenance, EMS, DMS, AMI, etc.
for the penetration of active resources at distribution network having plug-and-play features for network equipment
level is the complexity of the interconnection process. From and ICT solutions
network management point of view the increasing amount of secure and reliable
DG is often considered with reluctance as it brings the Traditional grid includes centralized power generation, and at
complexity of transmission network to distribution network distribution level one-directional power flow and weak
level. The main reason for the complexity is caused by the market integration. Smart grids include centralized and
present methods for managing the distribution networks as distributed power generation produced substantially by
well as the features of different active resource components renewable energy sources. They integrate distributed and
themselves, which are not sufficiently developed to enable active resources (i.e. generation, loads, storages and
easy interconnection. So far loads and customers have been electricity vehicles) into energy markets and power systems.
passive from network point of view. By making the customer Smart grids can be characterized by controllable multi-
connection point more flexible and interactive the demand directional power flow.
response functions (e.g. by real-time pricing, elastic load
control) are more achievable and the efficient use of existing Smart metering has been seen as an essential part of the
network and energy resources by market mechanisms can be vision of smart grids. Remote readable energy meter is being
improved. developed to be an intelligent equipment (i.e. interactive
customer gateway) including in addition to traditional energy
3. GENERAL FEATURES OF SMART GRIDS metering also different kind of new advanced functions based
on local intelligence. This gateway opens possibilities for
Smart grid concept has different aspects as shown in Fig. 1. It network companies, energy traders and service providers to
includes novel solutions of infrastructure for future power offer new kind of added-value services to end-customers.
distribution, e.g. use of power electronics and DC. Active
resources (i.e. distributed generation, loads, storages and The concept of smart grids may be characterized by words
electricity vehicles) actually change the traditional passive like flexible, intelligent, integration and co-operation. gGrids
distribution network to be an active one. New network are flexible because they utilize controllable resources
solutions and active resources call for novel ICT solutions for throughout the network. Respectively the passive network
network operation and asset management providing has flexibility by network capacity i.e. network itself may
intelligence to active networks. Smart grids enable active handle all probable loading conditions. Intelligence is simply
market participation of customers and also have effect on investments on protection, controllability and information
changes in business environment. Smart grids are customer- and telecommunication technologies instead of pure passive
driven marketplaces for DG and consumers. lines, cables, transformers and switchgears.

Drives, e.g: ICT solutions active market participation DG and existing controllable resources like direct load
mitigation of climate
network automation and
energy management
changes in business
control, reactive power compensation and demand side
efficiency of energy and
network resouces
asset management
new software tools
environment (e.g. service
purchasing, de-regulation,
integration provide a good potential as a controllable
quality of network service regulation) resources for the smart grids. The integration of DG and
Smart Grids flexible loads in distribution network will benefit the network
Intelligent management and when managed appropriately. The traditional passive network
Energy market

operation of smart grids


management or "fit & forget" principle in DG connection

Smart Grids

needs to be changed into active network management. The

Active resources integration of DG and other active resources into a
distribution system is a requirement in order to fully exploit
the benefits of active resources in network management. With
Future infrastructure of proper management of active resources the overall system
power distribution
performance may be improved from presently used practices.
requirements of SG large scale cabling Interactive interfaces
various network
DC and power
electronics in
for active resources
electrical conversion
electric vehicles One important control task in power systems is to maintain
enabling new distribution and network
customers balance between power production and consumption which
technologies new materials connection of DG
means keeping power systems frequency at an appropriate
Fig. 1. Aspects of smart grids level. This process is becoming more and more challenging
due to increase of penetration level of intermittent power network safety and stability. However the consequences of
production, for example wind and solar power. In recent stability issues for the whole power system and also for DG
years there have also been many serious frequency instability owners and other distribution network customers are
related wide-area power system blackouts in Europe and becoming more important when the disconnection of DG
USA, and their costs, both economical and social, are high. units may cause system wide stability or local power quality
4.2 ADINE project for active distribution management
4.1 Impacts of distributed generation on distribution network
A joint demonstration project called ADINE financed by the
The production of electricity close to consumers will reduce 6th Framework Programme of the European Commission,
the transfer of electricity. This will also affect network losses. Priority 6.1 Sustainable Energy Systems is in progress
Network losses may also increase when a large DG unit, e.g. (ADINE, Repo 2008). The partners of the project are ABB
wind farm, is located far from consumption and the electrical Oy Distribution Automation, AREVA Energietechnik
distance of transferred electricity increases compared to GmbH, AREVA T&D Ltd, ComPower AB, Lund University,
situation without a DG unit. Tampere University of Technology, and Technology Centre
Hermia Ltd as the coordinator. The project is under execution
The intermittent (non-dispatchable, uncertain and between October 2007 and September 2010.
uncontrolled) production into passive network does not
benefit network rating. The loadability of distribution The aim of the ADINE project is to develop new methods for
network is determined by voltage profile (decrease or rise), distribution network management including DG. When
power quality and thermal ratings. The intermittent distribution network is managed according to the ANM (i.e
production in weak rural distribution network may cause active network management) method the interactions of
voltage rise problems. The dimensioning of network becomes different active network devices can be planned and
quite challenging when there are different size and type of controlled to benefit the operation and stability of the
DG units along the network. The worst case planning network. One feature of this project is that it develops and
principle of DG interconnection in passive networks should demonstrates the ANM method and the enabling solutions
be replaced with a statistical planning approach in active simultaneously. The project will develop and demonstrate:
networks (Repo 2005). The increment of fault current level protection of distribution network including DG
due to new DG units may cause investments in networks if o application of communication based relays at
the rating of components is exceeded. The voltage control or distribution network
reactive power capability of DG units could also be utilized o fault location with the influence of DG
in network management. o co-ordinated protection planning application on
Network Information System (Mki 2007)
Requirements for the protection of distribution networks are voltage control of distribution network including DG
changing considerably (Mki 2007). Protection schemes o droop control of small-scale micro-turbine.
designed for unidirectional power flow may become o co-ordinated voltage control application on
ineffective. Unnecessary tripping as well as undetected faults SCADA/DMS which is controlling the setting values
or delayed relay operations may occur due to high DG of local voltage/reactive power controllers. (Kulmala
penetration. DG may also disturb the automatic re-closing. 2007)
The operation sequence of protection devices during a fault is
new-generation medium voltage STATCOM
thus important. Due to DG, the existing methods used in fault
o capable of filtering harmonics, eliminating flickers
location could also become inappropriate.
and compensating reactive power (Lauttamus 2008)
The current operational practice of distribution network o on top of above characteristics it can participate in
requires disconnection of DG units when a fault occurs. This mitigation of voltage dips and in controlling the
will keep the operational conditions simple and clear, safe voltage level of the distribution network.
and suitable for auto-reclosing. The purpose of DG unit
Figure 2 is visualizing the control levels of distribution
connection point protection (e.g. frequency and voltage
network and how the ANM method is affecting them. All
relays) is to eliminate the feeding of fault arc from a DG unit
hardware devices are basically working on local level. They
and to prevent unintended island operation. When the
get measurements from local measurement devices and
penetration level of DG increases the consequences of
operate based on this information. This is a de-centralized
immediate tripping of DG units may become adverse when
operation of protection and control like they are operated
short-circuit in transmission grid is seen by several DG units.
today. However the locations of protection and control
Even during a fault at distribution network unnecessary
devices are not limited to primary substation but they may
disconnection of DG units may occur due to unwanted trips
locate along the distribution network. Some control devices
of feeder or DG unit protection relays, loss of synchronism of
may also locate on a customer owned active device like DG
synchronous generators, sustained over-speed and over-
or STATCOM. Protection relays are working on the lowest
current of asynchronous generators or over-current and DC
level like in passive networks, but new feeder protection
over-voltage of power electronic converters. The current
schemes like directional over-current, distance and
operational practice clearly creates a contradiction between
differential protection, and new fault location applications are simulation environment for any combination of devices and
introduced. Automatic voltage regulation (AVR) of DG units, situations which are not possible to demonstrate in real
AVR of OLTC, STATCOM controller and power factor network without special arrangements. The two simulators
controller are added to automatic control system level. The are combined in order to exploit the capabilities of both
new issue in ANM is the utilization of automatic control simulators for interaction studies of power system and power
systems in distribution network operation greatly. electronic devices. Further, the simulation environment used
Demonstrations will show how the protection and voltage makes possible to develop control strategies for wind turbines
regulation in distribution networks can be improved through to qualify more demanding grid codes in the future.
advanced protection schemes and decentralised control of DG Minimization of the simulation time and the possibility to
units. distribute calculation power between two simulators can be
considered as additional benefits.

Fig. 3. Combined simulation and demonstration environment

Fig.2. Overview of the active distribution network in the of RTDS and dSPACE.
ADINE project.
The ANM method does not take into consideration practical
communication challenges e.g. communication media or The primary role of AMR (Automatic Meter Reading)
protocol for a small DG unit. The method itself is a systems has been to provide energy consumption data to the
conceptual description of management of active distribution utility, but the cost of retrofitting the existing energy
network. It is capable of applying any communication metering system may not be justified without added value
medium and protocol which is however restricted by functions. At present many utilities in Europe level are
capabilities of individual devices and software. Naturally the installing large-scale AMR projects. So far the focus of the
whole system would be more scalable, cheaper and simple if installations has been mainly on remote reading of energy
multi-functional and -vendor open interfaces (e.g. IEC measurements. Also some specific applications have been
61850) and data models (e.g. IEC 61970 and 61968) are developed, e.g. for load control. The comprehensive concept
applied at information exchange. of using AMR system and data in network and electricity
market management is still rare.
4.3 Real-time simulation and demonstration environment One requirement for creating additional value functions is the
open architecture in AMR systems to provide necessary
The ANM method is tested and demonstrated in a real-time integration possibilities. Standard integration ways e.g. OLE
simulation environment which combines the simulation for Process Control (OPC) or open connectivity via open
environments of RTDS (real-time digital simulator for standards makes it possible to develop new types of
electricity network) and dSPACE (real-time digital simulator intelligent system integrations.
for power electronics) as illustrated in Figure 3. It includes
also a library of simulation models of DG units and their Traditionally AMR and Distribution Management System
control algorithms. RTDS/dSPACE is not a traditional (DMS) have been separate systems without any integration
simulation environment rather it is test bed for secondary with each other as illustrated in Figure 4. The primary role of
devices of power system like protection relays and controllers AMR has been to provide energy consumption data to the
of DG units integrated into power system model. utility for billing and balance settlement purposes. AMR
system has also been used for load control in some
The electricity network can be modeled and simulated in installations. So far automatic monitoring and control center
transient level in RTDS side and developed applications in measures by the DMS have been used mostly for operating
dSPACE side via Simulink interface. The uniqueness of the 20 kV medium voltage networks. A fault in low voltage
simulation environment is the combined operation of these network is cleared automatically by blown fuse, but no
two simulation environments. It provides a general information about that is received to the control center. The
existence of a LV-network fault is usually indicated only by The integrated AMR, DMS and power quality monitoring
customer calls. systems offer information to be used in overall asset
management and network planning. At present advanced
Asset network data network calculation applications of network information
management customer data
Billing systems and DMS use hourly-load curves as load
Control Measurement
information. AMR system offers mass of measurement data
DMS data base
settlement to determine more detailed load models for different purposes
in network management and load prediction. Real-time AMR
SCADA AMR system data can be used in state-estimation, but for network planning
purposes load models are still needed. For network operation
purposes more accurate real-time state estimation of the
Substation whole network gives information on voltages, loads, losses,
and stressing of components, and also make it possible to
optimize e.g. network topology, voltage control, and load
control actions. In network planning more accurate load
models (e.g. more accurate division of customer groups,
Fig. 4. Traditional way of network management. regional models, etc.) for network calculations and
information on realization of power quality (i.e. interruptions,
The present AMR meters offer the platform (i.e the voltage dips, voltage levels) can be used to allocate measures
infrastructure and communication) to determine and develop and investments. More accurate information on hourly
new upper-level functions (see Figure 5). These will be used variation of losses is also valuable for the network company.
in developing the network asset management, market
enhancement as well as the customer service. First 5.1 An application for low voltage network fault indication
implementations of advanced AMR systems have already
changed the function of basic energy meter to be a smart
Low voltage network management may include functions, for
terminal unit and gateway to enable real time two-way
example, to indicate automatically if a fuse in the low voltage
communication between customers and utilities. In advanced
network has burnt or a conductor is broken, to locate the
meters alarms based by exceptional events i.e. network faults
fault, to provide accurate interruption data, to monitor
and voltage violations are enabled. Meters may also have
voltages at customer site in real-time and provide voltage
some protective functions adding the safety. The use and
level as power quality information for customer service.
integration of AMR in network operation can be seen as an
extension of SCADA and distribution automation to the low- In one Finnish distribution company (i.e. Koillis-Satakunnan
voltage level. As Figure 5 illustrates, AMR system can be Shk Oy) a development project were realised for
utilised in many functions of distribution company, e.g. to developing a comprehensive technology solution of new
support network operation (e.g automatic LV-fault indication, functions of AMR and related information systems for low
isolation and location, precise voltage and load data), voltage network monitoring and management (Jrventausta
network planning and asset management (e.g. exact load 2007). The aim was to combine new-generation energy
profiles for network calculations), power quality monitoring meters, data communication solutions and distribution
(e.g interruptions, voltage characteristics), customer service, management systems into an entity with an open architecture.
and load control in addition to traditional use in billing and The project consortium included different equipment and
load settlement. system vendors (i.e. ABB Oy, Aidon Oy, MX Electrix Oy,
PowerQ Oy), the research organisation (i.e Tampere
Customer service
Asset network data University of Technology) and the pilot distribution
management customer data
Billing company.
Control QMS Measurement Balance
DMS data base settlement The pilot company has an advanced DMS for real time
network analysis (i.e. load flow, fault currents), fault location,
SCADA AMR system switching planning etc. However, low voltage network
management has been totally in off-line mode since on-line
information has been available only from primary substations
and from some secondary substations along medium voltage
feeders as presented in Figure 4. The integration of AMR
makes it possible cost effectively to monitor low voltage
DMS = Distribution Management System
QMS = Quality Monitoring System
network and analyze fault situations since AMR
communication infrastructure can be used. Because network
monitoring in SCADA/DMS requires that events from meters
Fig. 5. Integrated information systems for comprehensive
are received in near real-time manner an effective way to
network management.
forward data from AMR is required. In the development
project OPC technology was selected for this purpose.
An advanced AMR meter works as an intelligent monitoring distribution network design and operation (e.g.
device and utilizes the communication infrastructure to investment plans and management of voltage drops and
provide spontaneous event or alarm information to control fluctuations, harmonics and other disturbances)
center with vital information on low voltage network faults outage statistics (e.g. needs of the Energy Market
and voltage levels. The meter includes algorithms to Authority).
inference the existence of a fault and type of the fault. In
certain cases, e.g. when neutral conductor is broken, the In recent years the department of Electrical Energy
advanced AMR meter may even isolate automatically the Engineering at Tampere University of Technology has
customer from the network. This requires a specific switching actively been developing a comprehensive power quality data
device which can be integrated into the advanced AMR management system together with distribution utilities and
meter. several different equipment and data system vendors.
(Koponen 2002, Mkinen 2003, Antila 2005)
Figure 6 illustrates a part of DMS screen in a case of broken
neutral conductor. Network coloring shows the results of The novel AMR technology makes it possible to integrate
inference to locate the broken line-section. basic power quality functions to AMR meter. In addition to
register interruptions with time-stamps the following
quantities for each of the three phases can also be able to
meter depending on the meter configuration: voltage and
current variations, active power, apparent power, total
reactive power, fundamental frequency reactive power,
voltage dips and swells, total distortion of the supply voltage,
some harmonic voltages, DC-voltage component, frequency
of the supply voltage, voltage unbalance between the three
phases. (Koponen 2002, Mkinen 2003)
The idea is to gather information from low voltage level and
integrate it, for example, to network data bases and different
planning and operation systems to increase knowledge with
much larger amount of information. The measurement data
Fig. 6. A case of broken neutral conductor (Krenlampi (i.e. even over several years) can be stored in the open
2007). relational Power Quality Database (PQDB) (see Figure 7).
The measurement data of the PQDB can be studied using the
When event data from meters are combined with topological Webbased application, PQNet system, in addition to DMS
network model in the DMS the original reason or faulted and network planning systems. The use of Web-based
component can be located. This gives enormous benefits for technology in PQ monitoring may be an internal or an
low voltage network management when, for example, blown outsourced service for distribution companies. Web-based PQ
fuses, broken conductors and voltage problems can be monitoring is an example of ASP (Application Service
presented to the operators at control center almost in real time Provider) functions. Power quality data can also be offered to
manner. the customers (e.g. industrial customers) with their energy
Corresponding comprehensive integrated information system consumption and billing data through the Web.
entity for network management based on installation of
advanced AMR meter to all 350 000 customers is also being
introduced in real operation Vattenfall Verkko Oy in Finland
(Kernen 2009).

5.2 Power quality monitoring

At the moment, voltage quality is usually monitored

temporarily at customer sites based on customer
reclamations, not comprehensively and continuously from the
entire distribution network. Power quality monitoring
including continuous voltage quality monitoring in larger
extent gives however important information for various Fig. 7. Realised power quality monitoring as part of
operations of distribution company. The development of integrated information systems
systematic procedures for power quality data management
supports in general: Factors that increase the need and possibilities of monitoring
customer services (e.g. quality reports, clarifying power quality also on low voltage level are e.g.:
customer requests, planning of compensation of reactive need for better customer service
power, instructions for the use of various equipment) reasonable priced meters
telecommunication development
applications which can use the power quality data in There are some technology trends which makes it possible to
network planning and operation define the customer interface in a brand new way. Large scale
regulation requirements AMR implementations are underway or planned in many
European countries. In the European level vision of smart
6. INTERACTIVE CUSTOMER GATEWAY grids smart metering has been seen as an essential part of
smart grids, especially for interconnection of distributed
For developing distribution management and functionality of generation, demand response and active distribution
electricity market one essential objective is to make the management. Communication technology and computer
customer, or at least customer connection point, active for systems and their integration are under rapid development.
improving e.g. interconnection of distributed generation, Power electronics has typically been used in high voltage
efficient use of energy, market-based demand response, transmission networks, but at the distribution level only some
quality of supply, and management of active distribution applications have been seen. Traditionally customer interface
networks. Remote readable energy meter is being developed has been bills, customer complaints and on-site readable
to be an intelligent equipment (i.e. interactive customer energy meter (i.e. conventional kWh-meter). Now the
gateway) including in addition to traditional energy metering concept of intelligent customer gateway enables the customer
also different kind of advanced functions based on local or the automatic functions related to the customer to be active
intelligence and power electronic applications as a part of in network management and in electricity market based on
active distribution networks. The interactive customer on-site applications and two-way communication with upper-
gateway will be based on the use of modern power level applications.
electronics, advanced AMR technology and two-way
communication between date bases and applications of the 7. POWER ELECTRONICS IN ELECTRICITY
distribution system operator (DSO), transmission system DISTRIBUTION
operator (TSO), service providers and electricity energy
market players (e.g. aggregators), as illustrated in Figure 8. The technical and economic development of power
electronics has been fast and continuous. Development of
Other service unit prices for conventional components used in distribution
DSO TSO Aggregator
supplier provider
networks, e.g. transformers, has been highly ascending. Price
Quality database Energy database On-line
erosion for power electronics components has been during
voltage billing data mesurements past decade about -7 %/a. This makes it possible to harness
interruptions load models control signals
the power electronics solutions to serve even the customer
Communication interface.
INTERACTIVE Actions by Efficient use of In several earlier studies and reports opinions and visions has
AC/DC Control of
been illustrated of utilization of power electronics in quality
distribution voltage Power
network interruptions electronics AMR frequency
generation control of electricity supply. Power electronics is already a
Active actions part of modern electricity distribution, for instance, in
network connections of small-scale generation units. In larger
Fig. 8. The concept of Interactive Customer Gateway (INCA) scale power electronics are still unexploited in actual
customer interfaces.
For developing interactive customer gateway a joint research
project is being executed together with Tampere University Apply of power electronics enables the LVDC supply in
of Technology, Lappeenranta University of Technology, distribution system. The European Union (EU) low voltage
Technical Research Centre of Finland (VTT), and with directive (LVD 72/23/EEC) defines the boundaries for the
several industrial partners and electricity companies. low voltage levels used in public distribution systems. It
(Jrventausta 2008) covers equipment designed for use with a voltage rating
between 50-1000 VAC and between 75-1500 VDC. HVDC
The development of interactive customer gateway enables: systems are used e.g. in power grid connection of off-shore
more efficient and flexible network interface e.g. for DG wind power mills having some hundreds or thousands of
and plug-in hybrid cars megawatts rating. Same principle gives many advantages also
on-line market (price) oriented load and DG control in low voltage level when connecting small scale distributed
management generation. E.g. wind mills hand fuel cells having some tens
frequency based load control during local or system level of kilowatt rating can be connected by DC/DC converter into
load and generation unbalance situations DC-network having DC/AC connection with AC distribution
services for energy savings and efficient use of energy system. Use of high DC-voltages, e.g. 750 VDC increases
on-line management and control of customer voltages, dramatically the power transmission capacity of low voltage
including also elimination of short interruptions (i.e. networks and makes it possible to connect small scale
reclosings and voltage dips) generators without more expensive medium voltage
more reliable constructions in distribution networks and installations. The Figure 9 illustrates the way power
advanced management of active distribution networks electronics may be applied in distribution networks.
using data on interactive customer gateway
Power electronic based customer interface improves
significantly the quality of supply. The voltage will be
constant 230 V independent on load variations. Short 7.1 Frequency dependent charging
interruptions can be eliminated, too. Intelligent power
electronics opens new tools for demand side management. As a part of the above mentioned development of Interactive
E.g. the local control of voltage can be used to decrease the Customer Gateway (INCA) also frequency dependent
loads. There is also the possibility to connect some loads charging has been studied (Rautiainen 2009). In these studies
(e.g. ovens) directly into LVDC supply. the control method of the battery chargers is based on local
frequency measurements carried out in the vehicles grid
interfaces. Control based on local measurements is an
AC/AC effective way to utilize large number of distributed resources
20 kV

20/1 kV
1 kV AC which have to react to frequency disturbances in a very
dynamic manner. Frequency dependent battery charging can
AC/DC DC/AC be used to enhance power systems frequency regulation
20 kV

20/1 kV
max. 1,5 kV DC capacity (which operates in normal grid conditions), to
enhance power systems disturbance reserves (which operate
AC/DC DC/AC in abnormal grid conditions) or to enhance both of these. One
20 kV

max. 1,5 kV DC
way to make a charger frequency dependent is to control the
20/1 kV charging power directly in accordance with frequency. The
most simple way to realize this is to stop the charging
Fig. 9. Possible main principles for power electronics in (regarding those chargers which are charging) when grid
distribution systems frequency falls under a predetermined level, or in some cases
charging can begin (regarding those chargers which are not
Practical questions are related to customer level converters, to
charging and which are engaged with a non-full battery) if
the DG-converters and to the technically and economically
frequency rises to a high enough level. Power can also be
optimal voltage levels to be used in DC-supply. Electrical
controlled in a continuous manner as a function of frequency.
safety questions and overall fault protection of LVDC system
In Figure 10, this principle is illustrated in three different
are essential study topics, too.
appliances: frequency regulation, disturbance reserve and a
More detailed descriptions and results are given e.g. in combination of these two.
(Kaipia 2008) and (Lassila 2009b)


Transportation has a very important function in todays
society. Globally, the energy production of transportation
systems is almost completely oil dependent. The
transportation sector is also a significant consumer of energy Fig. 10. Controlling charging power in accordance of
and a significant source of greenhouse gases and other frequency in a) frequency regulation application b)
emissions. Todays climate and energy policies imply disturbance reserve application c) combination application
strongly towards diversification of transportation fuels, (figure is not in scale)
improving energy efficiency and reducing emissions. The use
of electrical energy in a broader manner by means of plug-in In frequency regulation application, when frequency varies at
hybrid electric vehicles (PHEV) and electric vehicles (EV) interval f1...f2, power drawn from the grid by the charger is
offers a great potential to fulfill these challenging varied continuously at interval Pc1...Pc2. At nominal
requirements. frequency fn a charger draws power Pcn. In frequency
regulation application chargers regulate their power demands
This creates a challenge to the existing electricity distribution in frequency interval which is accepted as normal variation.
infrastructure. Depending on the electric car charging For example, in Nordel (i.e. Nordic interconnected
methods, the power demand may increase dramatically or transmission grid) interval f1...f2 could be 49.9...50.1 Hz.
stay almost at the present level. The existing electricity Varying charging power causes some uncertainty of the
distribution infrastructure sets some limitations on large-scale charging time of the batteries, and interval Pc1...Pc2 should
adoption of electric cars. Without any intelligence included in be limited so that charging of the vehicle is not disturbed too
the system, charging of electric cars may increase the peak much. The frequency regulation application could support the
power much higher compared with the present load level power balance related to some renewable energy resources
even though annual energy needed by the car is still low (i.e. which are of intermittent nature, for example wind power and
some thousand kWh). (Lassila 2009a) solar power. The disturbance reserve application is illustrated
Plug-in vehicles (PHEV and EV) could also be used to in Figure 10 b). The figure only presents the operation in low
produce different kinds of ancillary services to control and frequency area, because over frequency disturbances are not
manage electric power systems as a whole. usually a problem in large-scale power systems (Andersson
1992). In this application the charger does not react to small
frequency declines. When frequency falls low enough,
charger starts to reduce its power and power can be reduced
to zero. For example, in Nordel the operation frequency method the temperature settings of electric space heater
interval f1...f2 of this application could be 49.5...49.9 Hz. In thermostat is frequency dependent in accordance with local
this application chargers are set ready to participate to frequency measurement. Figure 11 illustrates the control
management of large power unbalances of the grid, and are method of the loads used in the study. Figure 11 a) depicts
thus usually fairly rarely used. The combination application, normal thermostat action. Heater is switched on whenever
which is presented in Figure 10 c) combines the two other temperature falls under certain level, and off-switching
applications presented earlier. During small frequency occurs when temperature rises high enough. Thereby
deviations (f2...f3) the charger changes its power within temperature varies around desired temperature Tdes, which is
interval Pc1...Pc2, and if frequency falls low enough (<f2), set by user of the load through a manual thermostat
power can be cut to zero. More detailed descriptions and adjustment. Figure 11 b) illustrates DDC method in a
simulation results can be found in (Rautiainen 2009). disturbance reserve appliance. The set point value of
temperature is now a function of grid frequency. Load reacts
Frequency dependent charging of plug-in vehicles offers an only at frequencies under f1 and maximum temperature
effective way to enhance a power systems frequency deviation from normal operation is set to Tmax.
stability. Distributed controllable loads offer a resource that
can very rapidly react to frequency disturbances, and the
harm (charging energy which is not received) to the vehicle
users caused by the charging control can be negligible.
However, wide utilization of the frequency dependent
charging requires economical incentives for the vehicle users
or legislative actions. The amount and availability of
controllable load depends strongly of many things including
plug-in vehicle penetration, driving habits, energy
consumption of the vehicles, charging opportunities
(including battery replacement service) and charging habits.
Plug-in vehicles are not yet widely in the markets, and Fig. 11. a) Conventional thermostat action b) Control method
thereby embedding new functions to the vehicles is of frequency dependent electric space heating load used in
technically fairly easy and could be executed with low costs. these studies (Figure is not in scale.)
In the future plug-in vehicles might communicate with More detailed descriptions and simulation results can be
different parties via wireless communication networks, and found in (Rautiainen 2009).
using this gateway the transmission system operator could
update the parameters of the frequency dependent controller. 9. CONCLUSIONS
Frequency dependency function could also be implemented
with vehicle-to-grid (V2G) concept. The use of V2G offers a This paper discusses on aspects of smart grids in general and
wider range of operation (for example nominal power) than gives some examples on smart grid features studied and
frequency dependent charging, although it is more complex. developed in Finnish research projects. The paper presents
some smart grid features at distribution level dealing with
8. FREQUENCY BASED LOAD CONTROL interconnection of distributed generation and active
distribution management, using automated meter reading
As a part of the above mentioned development of Interactive (AMR) systems in network management and power quality
Customer Gateway (INCA) also frequency dependent space monitoring, application of power electronics in electricity
heating has been studied to manage frequency disturbances in distribution, plug-in vehicles as part of smart grids, and
power systems. Setting values of space heater thermostats are frequency based load control. Remote readable energy meter
made dependent of locally measured network frequency. is being developed to be an intelligent equipment (i.e.
Studies are carried out through time domain simulations, and interactive customer gateway) including in addition to
also by laboratory testing by implementing a new function to traditional energy metering also different kind of advanced
the advanced AMR meter described in the chapter 5. Studies functions based on local intelligence and power electronic
imply that the use of frequency dependent loads in frequency applications as a part of active distribution networks. The
disturbance management is an efficient tool for managing interactive customer gateway will be based on the use of
power unbalances. This kind of load control methods modern power electronics, advanced AMR technology and
consequences and harm to the users of the space heaters can two-way communication between date bases and applications
be negligible, but the significance of this controllable load to of the distribution system operator (DSO), transmission
the power system can be very high. It is very important to system operator (TSO), service providers and electricity
coordinate the operation of frequency dependent load energy market players (e.g. aggregators).
carefully with other control actions taking into account the
cold load pick-up phenomenon related to the use of this type

In this study, dynamic demand control (DDC) method, ADINE (2007). Web-page of ADINE project, Available:
presented in (Schweppe) and further explored in (Short
2007), is applied to electric space heating loads. In the
Anderson, P. M., Mirheydar, M. (1992). An Adaptive Rautiainen, A., et al. (2009a). Using frequency dependent
Method for Setting Underfrequency Load Shedding electric space heating loads to manage frequency
Relays. IEEE Transactions on Power Systems, Vol. 7, disturbances in power systems. Proceedings of
Number 2, pp. 647655. PowerTech 2009, Romania
Antila, S., et al. (2005) Web- and event-based power quality Rautiainen, A., et. al. (2009b). Using Frequency Dependent
monitoring systems proved by application service Charging of Plugin Vehicles to Enhance Power Systems
provisioning model. International Energy Journal, Frequency Stability. European conference Smart Grids
Volume 6, Number 1, pp. 119-126 and mobility, Wrzburg, Germany
Chuang, A., (2008). EPRI Smart Grid R&D Overview. Repo, S., et al. (2005). Statistical short-team network
Presentation at HP Sustainability Innovation Workshop planning of distribution system and distributed
Jrventausta, P., et al. (2007). Using advanced AMR system generation, Proceedings of 15th Power Systems
in low voltage distribution network management. Computation Conference, Lige, Belgium
Proceedings of the 19th International Conference and Repo, S., et al. (2008). ADINE EU demonstration project
Exhibition on Electricity Distribution (CIRED 2007, of active distribution management. Proceedings of
Vienna, Austria CIRED Seminar 2008: SmartGrids for Distribution,
Jrventausta, P., et al. (2008). Interactive customer interface Frankfurt, Germany
for advanced distribution management and electricity Schweppe, F. C. Frequency adaptive, power-energy re-
market. Proceedings of CIRED Seminar 2008: scheduler, U.S. Patent 4317049
SmartGrids for Distribution, Frankfurt, Germany Short, J. A., Infield, D. G., Freris, L. L. (2007). Stabilization
Kaipia, T., et al. (2008). Power Electronics in Smartgrids of Grid Frequency through Dynamic Demand Control,
Impact on Power System Reliability Proceedings of IEEE Transactions on Power Systems, vol. 22, 3, pp.
CIRED Seminar 2008: SmartGrids for Distribution, 12841293
Frankfurt, Germany
Kernen, L., Usefulness of AMR data in the network
operation. Master of Science Thesis, Tampere University
of Technology, 62 p.
Koponen, P., et al. (2002). Supply voltage dip monitoring
with a sparse sampling kWh-meter. Proceedings of the
6th IASTED International Multi-Conference on Power
and Energy Systems (PES 2002), Marina Del Rey, CA,
Kulmala, A., et al. (2007). Active voltage level management
of distribution networks with distributed generation
using on load tap changing transformers. Proceedings of
IEEE Power Tech 2007, Lausanne, Switzerland
Krenlampi, M. (2007). Outage Management and Quality
Monitoring Utilizing AMR System. Transmission &
Distribution Europe 2007, Prague
Lauttamus, P., et al. (2008). Simulated electric arc furnace
voltage flicker mitigation with 3-level current-controlled
STATCOM, Proceedings of 23th Annual IEEE Applied
Power Electronics Conference and Exposition, pp. 1697-
1703, Austin, Texas, USA
Lassila, J., et al. (2009a). Electric cars challenge or
opportunity for the electricity distribution infrastructure?
European conference Smart Grids and mobility,
Wrzburg, Germany
Lassila, J., et al. (2009b). Potential and strategic role of
power electronics in electricity. Proceedings of 20th
International Conference and Exhibition on Electricity
Distribution (CIRED 2009), Praque
Mki, K. (2007). Novel methods for assessing the protection
impacts of distributed generation in distribution network
planning. Doctoral dissertation, Tampere University of
Mkinen, A., et al. (2003). Comprehensive development of
power quality management and evaluation of voltage dip
problems. Proceeginds of the 17th International
Conference on Electricity Distribution (CIRED 2003),
Barcelona, Spain