cired 2015 full paper: distributed and coordinated demand response for the supply of frequency...

CIRED 2015

gaspa

ABSTRACT

The availability of frequencyessential for any Utility to secure the power system in both interconnected and microgrid contexts. This paper presents a concept of Resources frequencywith both Frequency Containment Reserves (FCR) and Under Frequency Load Shedding (UFLS) requirements, are provided through a structure of Virtual Power Plant (VPP). performed on amotors and resistive loads control tdepending project

INTRODUCTION

European UtilitiesElectrical Renewable Energy Resources (RESspreadinggeneration unitenergy policies. Utilitreserves suppliers. system security robustnessduring 2003 Italian blackoutin

It appears sobecome a necessity controlled reserves over the network at any voltageThese reserves Distributed Generation (DG) units or by control through ability to distribute these over the network diffusion of ITechnologies loadsbasis This paper presents so a patented ICTdesigned for the

CIRED 2015

DISTRIBUTED AND COOR

Gaspard LEBEL Raphaël CAIRE Nouredine HADJSAID Univ. Grenoble Alpes, G2Elab CNRS, F-38000 Grenoble, France [email protected]

ABSTRACT

The availability of frequencyessential for any Utility to secure the power system in both interconnected and microgrid contexts. This paper presents a concept of Resources (DER)frequency-controlled reserveswith both Frequency Containment Reserves (FCR) and Under Frequency Load Shedding (UFLS) requirements, are provided through a structure of Virtual Power Plant (VPP). Physical demonstrationperformed on amotors and resistive loads control tdepending at the European scale within Dream FP7 project.

INTRODUCTION

European UtilitiesElectrical Renewable Energy Resources (RESspreading. This occurs generation unitenergy policies. Utilit ies of part of their historireserves suppliers. system security robustness face to large frequency excursionduring 2003 Italian blackoutin both interconne

It appears so become a necessity controlled reserves over the network at any voltageThese reserves Distributed Generation (DG) units or by control through ability to distribute these over the network diffusion of ITechnologies loads. The control of loads on a minute or even second basis used definitely not be doable in the past. This paper presents so a patented ICTdesigned for the

23

DISTRIBUTED AND COOR

Gaspard LEBEL Raphaël CAIRE

Nouredine HADJSAIDUniv. Grenoble Alpes, G2Elab

CNRS, G2Elab 38000 Grenoble, France

[email protected]

ABSTRACT

The availability of frequencyessential for any Utility to secure the power system in both interconnected and microgrid contexts. This paper presents a concept of coordinated

(DER), load modulation controlled reserves

with both Frequency Containment Reserves (FCR) and Under Frequency Load Shedding (UFLS) requirements, are provided through a structure of Virtual Power Plant

Physical demonstrationperformed on an off-grid systemmotors and resistive loads control t

at the European scale within Dream FP7

INTRODUCTION

European Utilities are currently facing Electrical Renewable Energy Resources (RES

. This occurs in a parallel context of bulk generation units decommissioning driven by climate and energy policies. Such trend is willing to deprive these

of part of their historireserves suppliers. It is then system security since RES

face to large frequency excursionduring 2003 Italian blackout

both interconnected and microgrid

the opportunity become a necessity – to distribute controlled reserves over the network at any voltageThese reserves could be supplied Distributed Generation (DG) units or by control through Demand Response ability to distribute these over the network is technicallydiffusion of Information anTechnologies (ICT) close to any kind of dispatchable

The control of loads on a minute or even second used definitely not be doable in the past.

This paper presents so a patented ICTdesigned for the supply of

23rd International Conference on Electricity Distribution

DISTRIBUTED AND COORFREQUENCY CONTAINMEN

Nouredine HADJSAID Univ. Grenoble Alpes, G2Elab

38000 Grenoble, France

[email protected]

The availability of frequency-controlled reserves is essential for any Utility to secure the power system in both interconnected and microgrid contexts. This paper

coordinated Distributed Energy oad modulation willing to supply

controlled reserves. These reserves, compliant with both Frequency Containment Reserves (FCR) and Under Frequency Load Shedding (UFLS) requirements, are provided through a structure of Virtual Power Plant

Physical demonstrations havegrid system. HVAC variable speed

motors and resistive loads control tat the European scale within Dream FP7

are currently facing Electrical Renewable Energy Resources (RES

in a parallel context of bulk decommissioning driven by climate and

uch trend is willing to deprive these of part of their historic frequency

then willing to challenge the RES-E generators

face to large frequency excursionduring 2003 Italian blackout [1]. Such paradigm is topical

and microgrid contexts

opportunity – that could perhaps even to distribute

controlled reserves over the network at any voltagesupplied by either co

Distributed Generation (DG) units or by Demand Response (DR)

ability to distribute these frequency-controlled is technically enabled by the large nformation and

close to any kind of dispatchable The control of loads on a minute or even second

used definitely not be doable in the past.

This paper presents so a patented ICT-based application, supply of frequency-controlled reserves

International Conference on Electricity Distribution

DISTRIBUTED AND COOR DINATED DEMAND RESPOFREQUENCY CONTAINMEN

Karel KUYPERSDep. of Electrical Engineering

Technische Universitet EindhovenEindhoven, the Nederland

[email protected]

controlled reserves is essential for any Utility to secure the power system in both interconnected and microgrid contexts. This paper

Distributed Energy willing to supply

These reserves, compliant with both Frequency Containment Reserves (FCR) and Under Frequency Load Shedding (UFLS) requirements, are provided through a structure of Virtual Power Plant

s have already been HVAC variable speed

motors and resistive loads control tests are now at the European scale within Dream FP7

are currently facing up to large Electrical Renewable Energy Resources (RES


uch trend is willing to deprive these c frequency-controlled

willing to challenge the E generators can show less

face to large frequency excursions as observed Such paradigm is topical

contexts.

that could perhaps even to distribute the frequency

controlled reserves over the network at any voltage level. by either controlling

Distributed Generation (DG) units or by performing load (DR) solutions. Thcontrolled reserves

enabled by the large Communication

close to any kind of dispatchable The control of loads on a minute or even second

used definitely not be doable in the past.

based application, controlled reserves


DINATED DEMAND RESPOFREQUENCY CONTAINMEN

Karel KUYPERS Dep. of Electrical Engineering

Technische Universitet EindhovenEindhoven, the Nederland

[email protected]

controlled reserves is essential for any Utility to secure the power system in both interconnected and microgrid contexts. This paper

Distributed Energy willing to supply

These reserves, compliant with both Frequency Containment Reserves (FCR) and Under Frequency Load Shedding (UFLS) requirements, are provided through a structure of Virtual Power Plant

already been HVAC variable speed

are now at the European scale within Dream FP7

p to large Electrical Renewable Energy Resources (RES-E)


uch trend is willing to deprive these controlled

willing to challenge the can show less

as observed Such paradigm is topical

that could perhaps even frequency-

level. ntrolling

performing load The

reserves enabled by the large

ommunication close to any kind of dispatchable

The control of loads on a minute or even second

based application, controlled reserves

to the Dreamdistributed loadtargeted by the solution FCR rangesupportcould thenrelays face to major frequency excursionsavoid the at any time for the system stabilitycomply as well with down regulation the shedding

The first part of the paper principle of operationpresentation of the proposedof the innovation by the Utilities and the DER involved in the VPP designed

CONCEPT

The fast frequency regulation TU/e Energy Management Services’ devicesCEMSs depending on the system frequency.these remotely CEMScomputing the frequency thresholds respected by each relay. defined in order system scale a targeted power

1 Commission


DINATED DEMAND RESPO NSE FOR THE SUPPLY OFREQUENCY CONTAINMEN T RESERVE (FCR)

Dep. of Electrical Engineering

Technische Universitet Eindhoven Eindhoven, the Nederland

[email protected] stéphane.bediou@schneider

to the power systemDream1 FP7 projectdistributed loadtargeted by the solution FCR or the Under Frequency Load Shedding (UFLS) range (Fig. 1). By doing so, the solution support the FCR in period of lack of bulk reserveould then pre-

relays face to major frequency excursionsavoid to affect vital loads andthe distributed gat any time for the system stabilitycomply as well with down regulation the shedding of Distributed Generation (DG).

Figure 1: Suitable operation frequency range of smoother UFLS (data applied in ENTSO

The first part of the paper principle of operationpresentation of the proposed. A thirdof the innovation by the Utilities and the DER involved in the VPP designed

CONCEPT

The fast frequency regulation TU/e and Schneider Electric leadsEnergy Management Services’ devicesCEMSs control the power supply of dispatchable loaddepending on the system frequency.these CEMSs remotely programmableCEMSs, a coordination platform is in charge of computing the frequency thresholds respected by each relay. defined in order system scale a targeted power

Dream is a p

Commission under

48Hz

UFLS

NSE FOR THE SUPPLY OT RESERVE (FCR)

Stéphane BEDIOUAlain GLATIGNYEnergy Division,

Schneider Electric IndustriesF-38000 Grenoble, France

stéphane.bediou@schneider

power system. The solution developed within FP7 project is based on a coordinated shedding of

distributed loads. The frequency range of operation targeted by the solution would encroach upon

Under Frequency Load Shedding (UFLS) . By doing so, the solution

the FCR in period of lack of bulk reserveempt partly the blind tripping of UFLS

relays face to major frequency excursionsaffect vital loads and

generators whose availability is at any time for the system stabilitycomply as well with down regulation

of Distributed Generation (DG).

: Suitable operation frequency range of smoother UFLS (data applied in ENTSO-E’s Regional Group Continental Europe)

The first part of the paper presentsprinciple of operation. A secondpresentation of the advances provided by

third part challenges of the innovation by the Utilities and the DER involved in the VPP designed

The fast frequency regulation and Schneider Electric leads

Energy Management Services’ devicescontrol the power supply of dispatchable load

depending on the system frequency. embed frequency

programmable thresholds. , a coordination platform is in charge of

computing the frequency shedding and reconnection thresholds respected by each relay. defined in order to reproduce by aggregation at the system scale a targeted power

Dream is a project funded

under FP7 grant

49.2Hz

49Hz

UFLS

Suitablerange of

Lyon, 15-

NSE FOR THE SUPPLY OT RESERVE (FCR)

Stéphane BEDIOUAlain GLATIGNYEnergy Division,

Schneider Electric Industries38000 Grenoble, France

stéphane.bediou@schneider

. The solution developed within based on a coordinated shedding of

The frequency range of operation would encroach upon

Under Frequency Load Shedding (UFLS) . By doing so, the solution enables

the FCR in period of lack of bulk reserveempt partly the blind tripping of UFLS

relays face to major frequency excursionsaffect vital loads and save from disconnection

whose availability is at any time for the system stability. The solution would comply as well with down regulation process

of Distributed Generation (DG).

: Suitable operation frequency range of smoother UFLS E’s Regional Group Continental Europe)

presents the conceptsecond part is

advances provided by part challenges finally the acceptability

of the innovation by the Utilities and the DER involved in the VPP designed.

The fast frequency regulation proposed by Grenoble INPand Schneider Electric leads on a pool of

Energy Management Services’ devices (CEMS)control the power supply of dispatchable load

depending on the system frequency. To do so, efrequency-based relaysthresholds. At the top of the

, a coordination platform is in charge of shedding and reconnection

thresholds respected by each relay. The thresholds are produce by aggregation at the

system scale a targeted power-frequency characteristic.

funded by agreement 609359

50Hz

49.8Hz

49.2Hz

FCR

49.98Hz

range of operation

-18 June 2015

Paper 1037

1

NSE FOR THE SUPPLY OF

Stéphane BEDIOU Alain GLATIGNY Energy Division,

Schneider Electric Industries 38000 Grenoble, France

sté[email protected]


The frequency range of operation would encroach upon either the

Under Frequency Load Shedding (UFLS) enables first to

the FCR in period of lack of bulk reserve. It empt partly the blind tripping of UFLS

. Thus it could save from disconnection

whose availability is essential The solution would

process thanks to of Distributed Generation (DG).

: Suitable operation frequency range of smoother UFLS

E’s Regional Group Continental Europe)

the concept and its part is devoted to

advances provided by the methodthe acceptability

of the innovation by the Utilities and the DER owners

proposed by Grenoble INPon a pool of Customer

(CEMS). These control the power supply of dispatchable load

To do so, each of based relays with At the top of the

, a coordination platform is in charge of shedding and reconnection

The thresholds are produce by aggregation at the

frequency characteristic.

the European609359

49.98Hz

operation

18 June 2015

Paper 1037

1/5


The frequency range of operation the

Under Frequency Load Shedding (UFLS) to

. It empt partly the blind tripping of UFLS

could save from disconnection

essential The solution would

thanks to

and its devoted to a

method the acceptability

owners

proposed by Grenoble INP, Customer

These control the power supply of dispatchable loads,

ach of with

At the top of the , a coordination platform is in charge of

shedding and reconnection The thresholds are

produce by aggregation at the frequency characteristic.

European

CIRED 2015

This coordinationthe behavior of bulk generation byThe able with either the FUnder Frequency Load Shedding schemes (UFLS)

AThe system can be deployed secondary substation, a primary substation or larger scale, thanks to a clouddependent architectures, the coordination platformisTerminal Unit (RTU).finally replicable autonomously from one substation to others or from one cloud The CEMS itfrequency sensor, a communication channel and a microcontroller. These componentindependent the one from another and linked by a local ICT bus or be an allcomponent can be directlyin a “regulated world” or being an proprietary device located down the metCEMS can embed as many triplets of {power meter, relay, comparator} as numbers of dispatchable loads willing to be driven independantly from one another. Fig. secondary substation, where the coordination platform is embed by the substationpresents then examplepresentsworlddeployed on an industrial sitesystemfunctions of each device are The selection of the architecture would be mainly driven by the wholesale and ancillary services marketframework:

and devices location

CEMS functionT

CIRED 2015

This coordinationthe behavior of bulk generation by any Virtual Power The present structureable to provide with either the FUnder Frequency Load Shedding schemes (UFLS)

ArchitectureThe system can be deployed secondary substation, a primary substation or larger scale, not correlated withthanks to a clouddependent architectures, the coordination platformis a softwareTerminal Unit (RTU).finally replicable autonomously from one substation to others or from one cloud The CEMS itfrequency sensor, a communication channel and a microcontroller. These componentindependent the one from another and linked by a local ICT bus or be an allcomponent can be directlyin a “regulated world” or being an proprietary device located down the metCEMS can embed as many triplets of {power meter, relay, comparator} as numbers of dispatchable loads willing to be driven independantly from one another. Fig. 2 presentssecondary substation, where the coordination platform is embed by the substationpresents then examplepresents an allworld. Fig. 3(b)deployed on an industrial sitesystem, with variable speed motorsfunctions of each device are The selection of the architecture would be mainly driven by the wholesale and ancillary services marketframework: • Regulated• Deregulated

and by the communication channels available devices location• Power Line Communication (PLC) available to link

the Utility power • ADSL, 3G only, etc.

CEMS functionThe CEMS carries out two main functions1. Inform periodically or on event the

platform of the power

23

This coordination enables the behavior of bulk generation

Virtual Power Plant (VPP) structure. present structure constitutes an autonomous system to provide frequency-controlled reserves, compliant

with either the Frequency CUnder Frequency Load Shedding schemes (UFLS)

rchitecture The system can be deployed secondary substation, a primary substation or

not correlated withthanks to a cloud-oriented architecture.dependent architectures, the coordination platform

a software, is embededTerminal Unit (RTU). Being an finally replicable autonomously from one substation to others or from one cloud-portfolio to others.

The CEMS itself is composed of a power meter, a frequency sensor, a communication channel and a microcontroller. These componentindependent the one from another and linked by a local ICT bus or be an all-in-one component. The allcomponent can be directly embed in a utility power meterin a “regulated world” or being an proprietary device located down the meter in a “deregulated world”. CEMS can embed as many triplets of {power meter, relay, comparator} as numbers of dispatchable loads willing to be driven independantly from one another.

presents an architecture designed at the scale of the secondary substation, where the coordination platform is embed by the substation’spresents then examples of CEMS architectures

all-in-one CEMS, adapted to the domestic (b) presents a split architecture,

deployed on an industrial site, with variable speed motors

functions of each device are

The selection of the architecture would be mainly driven by the wholesale and ancillary services market

egulated eregulated

the communication channels available devices location:

Power Line Communication (PLC) available to link Utility power meters and the substation RTU

3G only, etc.

CEMS functions he CEMS carries out two main functions

Inform periodically or on event the latform of the power willing to be shed by the relays


enables finally to reproduce virtually the behavior of bulk generation units, as

Plant (VPP) structure. constitutes an autonomous system

controlled reserves, compliant Containment Reserve (FCR) or

Under Frequency Load Shedding schemes (UFLS)

The system can be deployed locally at the scale secondary substation, a primary substation or

not correlated with the network topologyoriented architecture.

dependent architectures, the coordination platformed in the substation Remote

Being an autonomous system, finally replicable autonomously from one substation to

portfolio to others.

composed of a power meter, a frequency sensor, a communication channel and a microcontroller. These components can be either independent the one from another and linked by a local

one component. The allembed in a utility power meter

in a “regulated world” or being an proprietary device er in a “deregulated world”.

CEMS can embed as many triplets of {power meter, relay, comparator} as numbers of dispatchable loads willing to be driven independantly from one another.

an architecture designed at the scale of the secondary substation, where the coordination platform is

’s RTU. Fig. 3(a)of CEMS architectures

one CEMS, adapted to the domestic presents a split architecture,

deployed on an industrial site. It presents, with variable speed motors. In

functions of each device are highlighted in green


the communication channels available

Power Line Communication (PLC) available to link meters and the substation RTU

he CEMS carries out two main functionsInform periodically or on event the

willing to be shed by the relays


to reproduce virtually as usually targeted

Plant (VPP) structure. constitutes an autonomous system

controlled reserves, compliant Reserve (FCR) or

Under Frequency Load Shedding schemes (UFLS).

at the scale of a secondary substation, a primary substation or even at a

the network topologyoriented architecture. In topology

dependent architectures, the coordination platform which the substation Remote

autonomous system, it finally replicable autonomously from one substation to

portfolio to others.

composed of a power meter, a frequency sensor, a communication channel and a

s can be either independent the one from another and linked by a local

one component. The all-in-embed in a utility power meter

in a “regulated world” or being an proprietary device er in a “deregulated world”. The

CEMS can embed as many triplets of {power meter, relay, comparator} as numbers of dispatchable loads willing to be driven independantly from one another.


Fig. 3(a) and 3(b)of CEMS architectures. Fig. 3

one CEMS, adapted to the domestic presents a split architecture, willing to be

It presents an HVAC In Fig. 3(b), the

highlighted in green.


the communication channels available at the


he CEMS carries out two main functions: Inform periodically or on event the coordination

willing to be shed by the relays


to reproduce virtually usually targeted

constitutes an autonomous system controlled reserves, compliant

Reserve (FCR) or

of a even at a

the network topology n topology-

which the substation Remote

it is finally replicable autonomously from one substation to

composed of a power meter, a frequency sensor, a communication channel and a

s can be either independent the one from another and linked by a local

-one embed in a utility power meter

in a “regulated world” or being an proprietary device The

CEMS can embed as many triplets of {power meter, relay, comparator} as numbers of dispatchable loads


3(b) 3(a)

one CEMS, adapted to the domestic willing to be

an HVAC , the


at the


oordination willing to be shed by the relays

2.

To do soembeds to and forward this value to the coordination platform, thanks to its communication channel. The is then by comparing the system frequency measured locally with a frequency threshold coordination platform

CoThe cfunctions1.

2.

3.

Coordination Agent

PO

FF

(f)

Secondary

Substation


it controls. 2. Process autonomously in real

regulation, by driving the shedding or restoration of the loads he controls.

To do so, the CEMS leadembeds to measureand forward this value to the coordination platform, thanks to its communication channel. The

then in charge by comparing the system frequency measured locally with a frequency threshold coordination platform

Fig

Fig

oordination platform functionThe coordination platform functions: 1. Gather from all of the CEMS the close to real

data of the power amount 2. Gather the end

term of both marginal activation cost and frequency range of contribution

3. Set the load shedding order of each portfolio, based on this acceptance

Peer-

Primary

Support

Coordination Agent

Bottom

PO

FF

(f)

Load 1

…Load i…Load 3Load 2

UFLS

Range of operation

Market

platform

Clearing

results

Fa

Secondary

Substation

Process autonomously in realregulation, by driving the shedding or restoration of the loads he controls.

, the CEMS leads first on the measure the power consumption of the loads

and forward this value to the coordination platform, thanks to its communication channel. The

in charge in real-time of driving tby comparing the system frequency measured locally with a frequency threshold received coordination platform.

Fig 2. Large scale architecture

Fig 3. CEMS architectures

ordination platform functionoordination platform

ather from all of the CEMS the close to realdata of the power amount

ather the end-users load shedding acceptanceof both marginal activation cost and frequency

range of contribution. the load shedding order of each

portfolio, based on this acceptance

UFLS relays

Primary

Substation

-to-peer

Bottom-up

FCR

operationFb

Lyon, 15-

Process autonomously in real-time the frequency regulation, by driving the shedding or restoration of

s first on the the power consumption of the loads

and forward this value to the coordination platform, thanks to its communication channel. The microcontroller

of driving the relay position, by comparing the system frequency measured locally

received meanwhile

. Large scale architecture

. CEMS architectures

ordination platform function s oordination platform carries out

ather from all of the CEMS the close to realdata of the power amount available for shedding

users load shedding acceptanceof both marginal activation cost and frequency

the load shedding order of each portfolio, based on this acceptance criteria

Bulk

Business as FCR supplier

f

-18 June 2015

Paper 1037

2

time the frequency regulation, by driving the shedding or restoration of

s first on the controller he the power consumption of the loads

and forward this value to the coordination platform, microcontroller

he relay position, by comparing the system frequency measured locally

meanwhile from the

. Large scale architecture

carries out five major

ather from all of the CEMS the close to real-time available for shedding.

users load shedding acceptance, in of both marginal activation cost and frequency

the load shedding order of each load of the criteria.

Bulk Generation

Business as usualFCR supplier

Primary

Support

Coordination Agent

18 June 2015

Paper 1037

2/5

time the frequency regulation, by driving the shedding or restoration of

controller he the power consumption of the loads

and forward this value to the coordination platform, microcontroller

he relay position, by comparing the system frequency measured locally

from the

major

time

, in of both marginal activation cost and frequency

load of the

Coordination Agent

CIRED 2015

Coordination Platform Fig. two functions that major detailed

Fig

INNOVATION ADVANCES

Robustness The robustness of the innovation proposed comes from:

It can be noticed that the realload shedding and reconnection the CEMs. periodic or event processes.distribution of the real

CIRED 2015

4. Calculate the frequency threshold of 5. Broadcast the frequency

Coordination Platform Fig. 4 presents a simplified flowtwo functions that major functions of the coordination pdetailed.

Figure 4: Flowchart and structure of the messages exchanged between a

INNOVATION ADVANCES

Robustness The robustness of the innovation proposed comes from:• The architecture of system monitoring

coordination platform gathers the power consumption of each load directly monitored by each CEMS. The frequency monitoring is performed

• The coordination of the coordination platform decides by itselftime of the frequency behavior of all of the loads of its portfolio.

• The distribution of the decision processCEMS decidesshedding and reconnectionon-site system frequency monitoring, comply with the frequency behaviordecided upstream by the coordination platform.

• The replicability of the solution{coordination platform + coordinated CEMSs} is replicable enables to constitute pools of reasonnable size whose failure does not induce a danger for the system stability.

It can be noticed that the realload shedding and reconnection the CEMs. The coordination platform performs only periodic or event processes.distribution of the real

Coordination

Platform

Define load

shedding priority

Build Droop Curve

of Fig. 5: Assign

Frequency

Threshold fth

Load

Activation

criteria

Algo.

Fig.6

23

the frequency threshold of Broadcast the frequency

Coordination Platform presents a simplified flow

two functions that belongs functions of the coordination p

chart and structure of the messages exchanged between a coordination platform and a CEMS

INNOVATION ADVANCES

The robustness of the innovation proposed comes from:The architecture of system monitoringcoordination platform gathers the power consumption of each load directly monitored by each CEMS. The frequency monitoring is performed only by the CEMSsThe coordination of the coordination platform decides by itselftime of the frequency behavior of all of the loads of its portfolio. The distribution of the decision processCEMS decides finally autonomously shedding and reconnection

ite system frequency monitoring, comply with the frequency behaviordecided upstream by the coordination platform.

replicability of the solution{coordination platform + coordinated CEMSs} is replicable autonomously the one from another.enables to constitute

of reasonnable size whose failure does not induce a danger for the system stability.

It can be noticed that the realload shedding and reconnection

The coordination platform performs only periodic or event processes.distribution of the real-time processes

Coordination

Platform

Ask for a frequency

Threshold fthi(t),

Provide Load

Consumption Pi(t)

Define load -

shedding priority

Build Droop Curve

Assign

Frequency

thi(t)

Assign frequency

Threshold fthi(t),

Periodic Process

Msg

Msg


the frequency threshold of each load.Broadcast the frequency threshold to each C

Coordination Platform – CEMSs Interactions presents a simplified flowchart that includes the

belongs to the CEMS and functions of the coordination platform


INNOVATION ADVANCES

The robustness of the innovation proposed comes from:The architecture of system monitoringcoordination platform gathers the power consumption of each load directly monitored by each CEMS. The frequency monitoring is

only by the CEMSs. The coordination of the solution behaviorcoordination platform decides by itselftime of the frequency behavior of all of the loads of

The distribution of the decision processfinally autonomously

shedding and reconnection of the loadsite system frequency monitoring,

comply with the frequency behaviordecided upstream by the coordination platform.

replicability of the solution{coordination platform + coordinated CEMSs} is

onomously the one from another.enables to constitute at the system scale, several


It can be noticed that the real-time processes load shedding and reconnection – are performed o

The coordination platform performs only periodic or event processes. As detailed abo

time processes among the agents

CEMS i

i=

Monitor f

t-

Update fth

Continuous Real Time process

F(t

Yes


each load. threshold to each CEMS.

Interactions chart that includes the

to the CEMS and the five latform previously


The robustness of the innovation proposed comes from:The architecture of system monitoring: the coordination platform gathers the power consumption of each load directly monitored by each CEMS. The frequency monitoring is then

solution behavior: the coordination platform decides by itself before realtime of the frequency behavior of all of the loads of

The distribution of the decision process: Each finally autonomously of

of the loads, thanks to an ite system frequency monitoring, in order to

comply with the frequency behavior previouslydecided upstream by the coordination platform.

replicability of the solution: Each set of {coordination platform + coordinated CEMSs} is

onomously the one from another.at the system scale, several


time processes – mainly the are performed only by

The coordination platform performs only As detailed above,

among the agents

t=t+dt

CEMS i

=1...N

Monitor f(t)

dt<1s

t0>∆t

∆t~15min

Continuous Real Time process

t) < fthi

Open Relay

Close Relay

Yes

No


chart that includes the

the five previously

chart and structure of the messages exchanged between a

The robustness of the innovation proposed comes from: : the

coordination platform gathers the power consumption of each load directly monitored by

then

: the ore real-

time of the frequency behavior of all of the loads of

: Each of the

, thanks to an in order to previously

: Each set of {coordination platform + coordinated CEMSs} is

onomously the one from another. It at the system scale, several

of reasonnable size whose failure does not

mainly the nly by

The coordination platform performs only ve, this

among the agents

is a key factor of robustness of the solution. Neither the communication delay nor even a minutescommunication links can stdistributed frequencyCEMSs interpret the shedding and reconnection order only from an on

IntelligenceThe intelligence of the combined algorithmdefinition and frequency thresholds algorithm isto clear a market composed of several purchase and sale bids of frequencyeach bid cost, which depends on of power, the purchasing bids are defined by powerfrequency characteristic, while the selling bcommonly a power amountin all For clearing easiness, the whole purchcharacteristics are powerFa]The sale of each DER bid to match the demand, i.e. the assignment of a frequency thresholdperformed perfaj], with fastarting from the range the closest to the normal frequency.For each range defined in the whole interval order, up to reach the demand of the interval. price can be either a bid price or a spot price. algorithm is able to handle both capacity and energy pricsheddingconsidered as input for the clearing. The calculation of the frequency threshold of each DER assigned for operation within equa

For simplification issue, the equation (1) assumes a powereach range depending on theadapted to comply as well with noneven withinof the equation (1). The key idea plotted in

Open Relay

Close Relay


is a key factor of robustness of the solution. Neither the communication delay nor even a minutescommunication links can stdistributed frequencyCEMSs interpret the shedding and reconnection order only from an on-

Intelligence The intelligence of the combined algorithmdefinition and frequency thresholds algorithm is run by the coordination platformto clear a market composed of several purchase and sale bids of frequencyach bid is defined by a function of marginal shedding

cost, which depends on of power, the purchasing bids are defined by powerfrequency characteristic, while the selling bcommonly a power amountin all-or-nothing

For clearing easiness, the whole purchcharacteristics are power-frequency characteristicFa], regardless their limit prices. The sale of each DER bid to match the demand, i.e. the assignment of a frequency thresholdperformed per J

], with fa1 = Fstarting from the range the closest to the normal frequency. For each range ∆

defined in the whole interval order, up to reach the demand of the interval. price can be either a bid price or a spot price. algorithm is able to handle both capacity and energy price. In case of energy price,shedding depending on the frequency threshold is considered as input for the clearing.

The calculation of the frequency threshold of each DER assigned for operation within equation (1).

(1)

with �� , �∆��

For simplification issue, the equation (1) assumes a power-frequency characteristic defined as linear within each range ∆f j

depending on theadapted to comply as well with noneven within ∆f j. of the equation (1). The key idea plotted in

is a key factor of robustness of the solution. Neither the communication delay nor even a minutescommunication links can stdistributed frequency-controlled reserves, since the CEMSs interpret the shedding and reconnection order

-site system frequency monitoring

The intelligence of the innovation comes then from combined algorithm (Fig. 6definition and frequency thresholds

run by the coordination platformto clear a market composed of several purchase and sale bids of frequency-controlled reserves.

is defined by a function of marginal shedding cost, which depends on the frequency of power, the purchasing bids are defined by powerfrequency characteristic, while the selling bcommonly a power amount per

nothing for simplicity of operation.

For clearing easiness, the whole purchcharacteristics are first aggregated under a resultant

frequency characteristic, regardless their limit prices.

The sale of each DER bid to match the demand, i.e. the assignment of a frequency threshold

J adjacent sub-= Fa and fbJ =Fb

starting from the range the closest to the normal

For each range ∆f j, the DER defined in the whole interval ∆

order, up to reach the demand of the interval. price can be either a bid price or a spot price. algorithm is able to handle both capacity and energy

e. In case of energy price,depending on the frequency threshold is

considered as input for the clearing.

The calculation of the frequency threshold of each DER assigned for operation within

�� ; �� , � �� ∑ !"#$"�1�� , � ��&�

For simplification issue, the equation (1) assumes a frequency characteristic defined as linear within

. The slope can however be different depending on the range ∆f j. This equation could be easily adapted to comply as well with non

. Fig. 5 presents schematically the effect of the equation (1). The key idea plotted in

Lyon, 15-

is a key factor of robustness of the solution. Neither the communication delay nor even a minutescommunication links can stop the supply of these

controlled reserves, since the CEMSs interpret the shedding and reconnection order

site system frequency monitoring

innovation comes then from . 6) of load-shedding order

definition and frequency thresholds computationrun by the coordination platform

to clear a market composed of several purchase and sale controlled reserves. In term of price,

is defined by a function of marginal shedding frequency thres

of power, the purchasing bids are defined by powerfrequency characteristic, while the selling b

per DER which is activated for simplicity of operation.

For clearing easiness, the whole purchaggregated under a resultant

frequency characteristic, defined in a range [Fb ; , regardless their limit prices.

The sale of each DER bid to match the demand, i.e. the assignment of a frequency threshold to each DE

-frequency ranges =Fb. The ranges

starting from the range the closest to the normal

he DER whose price functions are defined in the whole interval ∆f j, are selected at the merit order, up to reach the demand of the interval. price can be either a bid price or a spot price. algorithm is able to handle both capacity and energy

e. In case of energy price, a probability of loaddepending on the frequency threshold is

considered as input for the clearing.

The calculation of the frequency threshold of each DER assigned for operation within an interval ∆

��' ( ∑)*+,∑#$*+,��&��-�./-�� .��0��&�� 0 ��0��-�-0


The slope can however be different This equation could be easily

adapted to comply as well with non-linear characteristic presents schematically the effect

of the equation (1). The key idea plotted in

-18 June 2015

Paper 1037

3

is a key factor of robustness of the solution. Neither the communication delay nor even a minutes-long loss of

op the supply of these controlled reserves, since the

CEMSs interpret the shedding and reconnection order site system frequency monitoring.

innovation comes then from shedding order

computation. This run by the coordination platform. It enables

to clear a market composed of several purchase and sale In term of price,

is defined by a function of marginal shedding threshold. In term

of power, the purchasing bids are defined by powerfrequency characteristic, while the selling bids are

which is activated for simplicity of operation.

For clearing easiness, the whole purchasing aggregated under a resultant

, defined in a range [Fb ;

The sale of each DER bid to match the demand, i.e. the to each DER, is then

frequency ranges ∆f j = [fbj

. The ranges are scanned starting from the range the closest to the normal system

whose price functions are selected at the merit

order, up to reach the demand of the interval. The trading price can be either a bid price or a spot price. The algorithm is able to handle both capacity and energy

a probability of loaddepending on the frequency threshold is

The calculation of the frequency threshold of each DER interval ∆f j follows the

1*,1*,

/-� -�∆�0��0��-�-0



linear characteristic presents schematically the effect

of the equation (1). The key idea plotted in Fig. 5 is to

18 June 2015

Paper 1037

3/5

is a key factor of robustness of the solution. Neither the long loss of

op the supply of these controlled reserves, since the

CEMSs interpret the shedding and reconnection order

innovation comes then from a shedding order

. This enables

to clear a market composed of several purchase and sale In term of price,

is defined by a function of marginal shedding . In term

of power, the purchasing bids are defined by power-ids are

which is activated

asing aggregated under a resultant

, defined in a range [Fb ;

The sale of each DER bid to match the demand, i.e. the R, is then

= [fbj ; are scanned

system

whose price functions are selected at the merit

The trading The

algorithm is able to handle both capacity and energy a probability of load-

depending on the frequency threshold is

The calculation of the frequency threshold of each DER follows the



linear characteristic presents schematically the effect

to

CIRED 2015

extend each piled block of dispatchable power amfrequency axis up to cross the targeted powercharacteristic, presented as

The DER offers not purchased within sale in the range can face the competition from DER that used not to be defined in Since the resultant demand is usually a linear function while the resultant offer is composed of piled blocks, the algorithm considers tolerance lboth curves stays close enough at the clearing output. The aggregatto system scale demand. The algorithm integrates finally a lprocess to cancel the trading price higher than an upper limit predefined by the purchaseraccept the purchase of exactlfinal bid partially cut

CIRED 2015

extend each piled block of dispatchable power amfrequency axis up to cross the targeted powercharacteristic, presented as

Figure

The DER offers not purchased within sale in the range can face the competition from DER that used not to be defined in ∆f j. Since the resultant demand is usually a linear function while the resultant offer is composed of piled blocks, the algorithm considers tolerance lboth curves stays close enough at the clearing output. The aggregationto make this discontinuity of offers negligible face to the system scale demand. The algorithm integrates finally a lprocess to cancel the trading price higher than an upper limit predefined by the purchaser. According to the same logic, a purchaser can accept the purchase of exactly equal to his initial purchase bid, or final bid partially cut

Load-shedding

order

23

extend each piled block of dispatchable power amfrequency axis up to cross the targeted powercharacteristic, presented as linear in the present example.

Figure 5: Targeted step

The DER offers not purchased within sale in the range ∆f j+1, if they are still defined, where they can face the competition from DER that used not to be

.

Since the resultant demand is usually a linear function while the resultant offer is composed of piled blocks, the algorithm considers tolerance lboth curves stays close enough at the clearing output.

ion of small size offersmake this discontinuity of offers negligible face to the

system scale demand. The algorithm integrates finally a lprocess to cancel the potential trading price higher than an upper limit predefined by the

. According to the same logic, a purchaser can accept the purchase of power

y equal to his initial purchase bid, or final bid partially cut during

Figure

POFF (f)

faj

Load 1

…Load i…Load 3Load 2

i=1…Kj

Pofftotjfth


extend each piled block of dispatchable power amfrequency axis up to cross the targeted power

linear in the present example.

: Targeted step-by-step droop curve

The DER offers not purchased within ∆fif they are still defined, where they

can face the competition from DER that used not to be

Since the resultant demand is usually a linear function while the resultant offer is composed of piled blocks, the algorithm considers tolerance limits, to guarantee that both curves stays close enough at the clearing output.

of small size offers is however supposed make this discontinuity of offers negligible face to the

The algorithm integrates finally a loop of bids validation potential purchase of reserves at a

trading price higher than an upper limit predefined by the . According to the same logic, a purchaser can

power-frequency characteristic y equal to his initial purchase bid, or

during the market clearing

gure 6: Frequency

� ∑ !23$245��

fbj

thj

fth3fth2

fth1


extend each piled block of dispatchable power among the frequency axis up to cross the targeted power-frequency


step droop curve

∆f j are put back for if they are still defined, where they


Since the resultant demand is usually a linear function while the resultant offer is composed of piled blocks, the

imits, to guarantee that both curves stays close enough at the clearing output.

is however supposed make this discontinuity of offers negligible face to the

oop of bids validation purchase of reserves at a


frequency characteristic y equal to his initial purchase bid, or accept as well

clearing.

Frequency market c

Slopeof the

power-frequency

characteristic

targeted

fj


ong the frequency


are put back for if they are still defined, where they


Since the resultant demand is usually a linear function while the resultant offer is composed of piled blocks, the

imits, to guarantee that both curves stays close enough at the clearing output.

is however supposed make this discontinuity of offers negligible face to the

oop of bids validation purchase of reserves at a


frequency characteristic accept as well

PERSPECTIVES

As much of Demand Response solutions, the success of the present concept upstream by the Utilities and downstream by the owners.

Utility acceptanceFirstly, the solution will have to convince the network operators of itscontribution in realdisturbance. The first stesolution design which is compliant with FCR requirements, defined in Loadnetwork stability studiesthe benefit on stabilityinstance islanded poweruniqueon the model of Kuching Island power system [reversibility of the solution proposed enables to smooth the opposite imbalance that occurslarge system without the availability of enough reversible reserve [ Reala 4kVA offdroop curve supplied, compared with the curve targeted and then the time response of the five CEMS designed just for these tests [

market clearing and threshold


PERSPECTIVES

As much of Demand Response solutions, the success of the present concept upstream by the Utilities and downstream by the owners.

Utility acceptanceFirstly, the solution will have to convince the network

erators of its contribution in realdisturbance.

The first step toward system stability solution design which is compliant with FCR requirements, defined in Load-Frequency Control and Reserves [network stability studiesthe benefit on stabilityinstance the interest of the solution

landed powerunique interconnection. on the model of Kuching Island power system [reversibility of the solution proposed enables to smooth the opposite imbalance that occurslarge tripping of UFLS relayssystem without the availability of enough reversible reserve [4].

Real-time physical a 4kVA off-grid system to check first the accuracy of the

oop curve supplied, compared with the curve targeted and then the time response of the five CEMS designed just for these tests [

threshold assignment algorithm

PERSPECTIVES

As much of Demand Response solutions, the success of the present concept is conditioned by its adoption upstream by the Utilities and downstream by the

Utility acceptance Firstly, the solution will have to convince the network

benefit on system stability, its effective contribution in real-time and its robustness face to any

toward system stability solution design which is compliant with FCR requirements, defined in the ENTSO

Frequency Control and Reserves [network stability studies have then been performed to test the benefit on stability. The first studies

the interest of the solution system, following the

interconnection. These tests have been performed on the model of Kuching Island power system [reversibility of the solution proposed enables to smooth the opposite imbalance that occurs

tripping of UFLS relayssystem without the availability of enough reversible

physical tests have grid system to check first the accuracy of the

oop curve supplied, compared with the curve targeted and then the time response of the five CEMS designed just for these tests [5].

assignment algorithm

Lyon, 15-

As much of Demand Response solutions, the success of conditioned by its adoption

upstream by the Utilities and downstream by the

Firstly, the solution will have to convince the network system stability, its effective

time and its robustness face to any

toward system stability is solution design which is compliant with FCR

the ENTSO-E Network Code on Frequency Control and Reserves [

have then been performed to test first studies demonstra

the interest of the solution to securefollowing the tripping of its

These tests have been performed on the model of Kuching Island power system [reversibility of the solution proposed enables to smooth the opposite imbalance that occurs in K

tripping of UFLS relays, and which is fatal for the system without the availability of enough reversible

s have been performed grid system to check first the accuracy of the

oop curve supplied, compared with the curve targeted and then the time response of the five CEMS designed

assignment algorithm

-18 June 2015

Paper 1037

4


upstream by the Utilities and downstream by the DER



is ensured by a solution design which is compliant with FCR

E Network Code on Frequency Control and Reserves [2]. Dynamic

have then been performed to test demonstrate for ecure an almost tripping of its

These tests have been performed on the model of Kuching Island power system [3]. The reversibility of the solution proposed enables to smooth

in Kuching after a , and which is fatal for the

system without the availability of enough reversible

been performed as well on grid system to check first the accuracy of the


18 June 2015

Paper 1037

4/5


DER



by a solution design which is compliant with FCR

E Network Code on Dynamic

have then been performed to test for

an almost tripping of its

These tests have been performed he

reversibility of the solution proposed enables to smooth after a

, and which is fatal for the system without the availability of enough reversible

as well on grid system to check first the accuracy of the


CIRED 2015

Finally, already been challenged robustnessdeployment of a cloudinterconnected Electric Malpensa Airportdemonstration

DER owner acceptance On the DER side, a limited disturbance of business as usual DER owner activity will constitute a key factor of success. identi

The targetingdomestic side, existing energy boxes like Wiser [would become easily compliant with the present solution, by controlling the tripping of boiler or electric heaters. On the tertiarycontrollers like logic controller from Schneider Electric for HVAC and pumping solutionsdown flexibility required. The M171 CEMS script installation.compliant as well. The lead to each CEMS at the expenplatform: each of them requests upstream by themselves for a frequency threshold update, that requires only the knowledge by the CEMS of building network’s proxi, and no port opening to become accessible from outside. Thefinally of the threshold assigned to each from the nominal frequency it is, the less shed49.9Hz ENTSOnumber of events fall to 7.8 at 49.85Hz and 0.7 at 49.9Hz(data from may 2005frequency threshold assignment algorithm limitenables to consider DER owner contribution wishes, that

CIRED 2015

Finally, the robustness face to any already been challenged robustness would then deployment of a cloudinterconnected Electric headquarter,Malpensa Airportdemonstration

DER owner acceptance On the DER side, a limited disturbance of business as usual DER owner activity will constitute a key factor of success. The three main kinds of disturbances have been identified: 1. The complexity of t

existing facility2. The cyber

connect the DER to a communication channel3. The number and the length of DER shedding events.

The complexity of the CEMS integrationtargeting specific loads, domestic side, existing energy boxes like Wiser [would become easily compliant with the present solution, by controlling the tripping of boiler or electric heaters. On the tertiarycontrollers like logic controller from Schneider Electric for HVAC and pumping solutionsdown flexibility required. The M171 CEMS script installation. Some other domestic and tertiary could be compliant as well. The cyber-security issueslead to each CEMS at the expenplatform: each of them requests upstream by themselves for a frequency threshold update, that requires only the knowledge by the CEMS of building network’s proxi, and no port opening to become accessible from outside. The question finally of the threshold assigned to each from the nominal frequency it is, the less shedding occurs49.9Hz is on average ENTSO-E Northern Europe synchronous area, the number of events fall to 7.8 at 49.85Hz and 0.7 at 49.9Hz(data from may 2005frequency threshold assignment algorithm limit a frequency range of operation enables to consider DER owner contribution wishes, that

23

robustness face to any already been challenged in the previous paragraph

would then be challenged in 2015 by the deployment of a cloud-oriented demonstrator, controlling interconnected DER. These DER are

headquarter, Grenoble INP buildingsMalpensa Airport and ICCS demonstration field tests.

DER owner acceptance On the DER side, a limited disturbance of business as usual DER owner activity will constitute a key factor of

The three main kinds of disturbances have been

complexity of the CEMS existing facility.

cyber-security issues inherent to the need to connect the DER to a communication channelThe number and the length of DER shedding events.

complexity of the CEMS integrationspecific loads, as

domestic side, existing energy boxes like Wiser [would become easily compliant with the present solution, by controlling the tripping of boiler or electric heaters. On the tertiary side, the association of variable speed controllers like the Altivar logic controller from Schneider Electric for HVAC and pumping solutions is willingdown flexibility required. The M171 CEMS script and connects

Some other domestic and tertiary could be compliant as well.

security issues arelead to each CEMS at the expenplatform: each of them requests upstream by themselves for a frequency threshold update, that requires only the knowledge by the CEMS of building network’s proxi, and no port opening to become accessible from outside.

question of shedding eventfinally of the threshold assigned to each from the nominal frequency it is, the less

ding occurs. While the frequency threshold of on average reached up to 339 times per day,

E Northern Europe synchronous area, the number of events fall to 7.8 at 49.85Hz and 0.7 at 49.9Hz(data from may 2005 to april 2006, frequency threshold assignment algorithm

frequency range of operation enables to consider DER owner contribution wishes, that


robustness face to any ICT in the previous paragraphbe challenged in 2015 by the

oriented demonstrator, controlling . These DER are located in

Grenoble INP buildingsICCS facilities (Greece

DER owner acceptance On the DER side, a limited disturbance of business as usual DER owner activity will constitute a key factor of


he CEMS integration

security issues inherent to the need to connect the DER to a communication channelThe number and the length of DER shedding events.

complexity of the CEMS integrationas described in

domestic side, existing energy boxes like Wiser [would become easily compliant with the present solution, by controlling the tripping of boiler or electric heaters.

the association of variable speed [7] with an M171 [

logic controller from Schneider Electric for HVAC and is willing to provide easily

down flexibility required. The M171 embedcts to existing

Some other domestic and tertiary could be

are partly solved lead to each CEMS at the expense of the coordination platform: each of them requests upstream by themselves for a frequency threshold update, that requires only the knowledge by the CEMS of building network’s proxi, and no port opening to become accessible from outside.

shedding event occurrencefinally of the threshold assigned to each from the nominal frequency it is, the less

While the frequency threshold of reached up to 339 times per day,

E Northern Europe synchronous area, the number of events fall to 7.8 at 49.85Hz and 0.7 at 49.9Hz

april 2006, [9]). Thefrequency threshold assignment algorithm

frequency range of operation enables to consider DER owner contribution wishes, that


ICT disturbance in the previous paragraph. Such be challenged in 2015 by the

oriented demonstrator, controlling located in Schneider

Grenoble INP buildings, Milano Greece) as Dream

On the DER side, a limited disturbance of business as usual DER owner activity will constitute a key factor of


integration on an

security issues inherent to the need to connect the DER to a communication channel. The number and the length of DER shedding events.

complexity of the CEMS integration is addressed described in Fig. 3. On the

domestic side, existing energy boxes like Wiser [would become easily compliant with the present solution, by controlling the tripping of boiler or electric heaters.

the association of variable speed M171 [8], the last

logic controller from Schneider Electric for HVAC and easily the up and

embeds directly existing sensors of the


solved by giving the se of the coordination

platform: each of them requests upstream by themselves for a frequency threshold update, that requires only the knowledge by the CEMS of building network’s proxi, and no port opening to become accessible from outside.

occurrence depends finally of the threshold assigned to each DER: the further from the nominal frequency it is, the less frequency

While the frequency threshold of reached up to 339 times per day,

E Northern Europe synchronous area, the number of events fall to 7.8 at 49.85Hz and 0.7 at 49.9Hz

. The ability of the frequency threshold assignment algorithm (Fig. 6)

frequency range of operation for each DERenables to consider DER owner contribution wishes, that


has . Such

be challenged in 2015 by the oriented demonstrator, controlling

Schneider Milano Dream1

On the DER side, a limited disturbance of business as usual DER owner activity will constitute a key factor of


on an

security issues inherent to the need to

The number and the length of DER shedding events.

is addressed by . On the

domestic side, existing energy boxes like Wiser [6] would become easily compliant with the present solution, by controlling the tripping of boiler or electric heaters.

the association of variable speed ], the last

logic controller from Schneider Electric for HVAC and the up and

the of the


by giving the se of the coordination

platform: each of them requests upstream by themselves for a frequency threshold update, that requires only the knowledge by the CEMS of building network’s proxi, and no port opening to become accessible from outside.

depends DER: the further

frequency While the frequency threshold of

reached up to 339 times per day, in E Northern Europe synchronous area, the

number of events fall to 7.8 at 49.85Hz and 0.7 at 49.9Hz ability of the

) to each DER

enables to consider DER owner contribution wishes, that

is willing to reduce its adoption relectance. events last few seconds only, that is not willing to disturb the DER processes (at 49.with an average duration of 1.5s [

CONCLUSION

The present concept has been developed with the objective to propose one solution to the appearing issue of lack of frequencylocationsconvince Utilities through business models for the whole power system chainwill to 2016.

AcknowledgmentsThe development of this concept has been sponsored by Schneider Electric, Energy BU, the European Commission within the framework of Dreamn° REFERENCES [1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]


is willing to reduce its adoption relectance. events last few seconds only, that is not willing to disturb the DER processes (at 49.with an average duration of 1.5s [

CONCLUSION

The present concept has been developed with the objective to propose one solution to the appearing issue of lack of frequencylocations, includingconvince Utilities through new demonstrationbusiness models for the whole power system chainwill be Schneider Electric and Grenoble INPto 2016.

AcknowledgmentsThe development of this concept has been sponsored by Schneider Electric, Energy BU, the European Commission within the framework of Dream FP7 European project n° 609359.

REFERENCES

ENTSO-E, 2004, “Committee on the 28 September 2003 Blackout in Italy

ENTSO-E, "Network Code on LoadReserves" ENTSOAISBL, Brussels, Belgium, Dec

Denis Lee Hau Aik,model incorporating load shedding: analytic modeling and applications," Power Systems, IEEE Transactions on , vol.21, no.2, pp.709,717, May 2

G. Lebel, D. Wang, R. Caire, N. Hadjsaid, SGlatigny, "Distributed Frequency-controlled reserves supply", Eindhoven IEEE

G. Lebel, K. Kuypers, R. Caire, N. Hadjsaid, S. Bediou, and A. Glatigny, "Domestic Demand Response for Primary Control Support", CIGRE International Colloquium on Lightning and Power System, Lyon 2014. Student Poster Session.

Schneider Electric, 2014, "for the home, Bring the essencElectric, Paris, France

Schneider Electric, 2011, "Altivar 212 Variable Speed Drive 3-phase asynchronous motors from 0.75 to 75 kWSchneider Electric, Paris, France

Schneider Electric, 2014, "Modicon Controller, Hardware User MFrance

Zhao Xu; Ostergaard, J.; Togeby, M.; Isleifsson, F.R., "Evaluating frequency quality of Nordic system using PMU data," Energy Society General Meeting Electrical Energy in the 21st Century, 2008 IEEEpp.1,5, 20-24 July 2008

is willing to reduce its adoption relectance. events last few seconds only, that is not willing to disturb the DER processes (at 49.8Hz, 95% last less than 27s, with an average duration of 1.5s [

CONCLUSION

The present concept has been developed with the objective to propose one solution to the appearing issue of lack of frequency-controlled reserves in some

, including microgridconvince Utilities and DER owners

demonstrationbusiness models for the whole power system chain

Schneider Electric and Grenoble INP

Acknowledgments The development of this concept has been sponsored by Schneider Electric, Energy BU, the European Commission within the framework of

FP7 European project

REFERENCES

, 2004, “FINAL REPORT of the Investigation Committee on the 28 September 2003 Blackout in Italy

E, "Network Code on LoadReserves" ENTSO-E Definitions and Acronyms", ENTSOAISBL, Brussels, Belgium, Dec

Lee Hau Aik, A generalmodel incorporating load shedding: analytic modeling and applications," Power Systems, IEEE Transactions on , vol.21, no.2, pp.709,717, May 2006. G. Lebel, D. Wang, R. Caire, N. Hadjsaid, SGlatigny, "Distributed and Coordinated Demand Response for

controlled reserves supply", Eindhoven IEEE, in press (abstractG. Lebel, K. Kuypers, R. Caire, N. Hadjsaid, S. Bediou, and A. Glatigny, "Domestic Demand Response for Primary Control Support", CIGRE International Colloquium on Lightning and Power System, Lyon 2014. Student Poster Session. Schneider Electric, 2014, "Wiser: energy management solutions for the home, Bring the essencElectric, Paris, France Schneider Electric, 2011, "Altivar 212 Variable Speed Drive

phase asynchronous motors from 0.75 to 75 kWSchneider Electric, Paris, FranceSchneider Electric, 2014, "Modicon Controller, Hardware User M

Zhao Xu; Ostergaard, J.; Togeby, M.; Isleifsson, F.R., "Evaluating frequency quality of Nordic system using PMU data,"

Society General Meeting Electrical Energy in the 21st Century, 2008 IEEE

24 July 2008

Lyon, 15-

is willing to reduce its adoption relectance. events last few seconds only, that is not willing to disturb

8Hz, 95% last less than 27s, with an average duration of 1.5s [9]).

The present concept has been developed with the objective to propose one solution to the appearing issue

controlled reserves in some microgrid. Next step is now to

and DER owners of its robustness, demonstrations and to define suitable

business models for the whole power system chainSchneider Electric and Grenoble INP

The development of this concept has been sponsored by Schneider Electric, Energy BU, Utility Innovation and the European Commission within the framework of

FP7 European project [2], under grant agreement

FINAL REPORT of the Investigation Committee on the 28 September 2003 Blackout in Italy

E, "Network Code on Load-Frequency Control and E Definitions and Acronyms", ENTSO

AISBL, Brussels, Belgium, Dec 2012. A general-order system frequency response

model incorporating load shedding: analytic modeling and applications," Power Systems, IEEE Transactions on , vol.21,

G. Lebel, D. Wang, R. Caire, N. Hadjsaid, Sand Coordinated Demand Response for

controlled reserves supply", PowerTech 2015 IEEE abstract waiting for approval).


Wiser: energy management solutions for the home, Bring the essence of energy efficiency

Schneider Electric, 2011, "Altivar 212 Variable Speed Drive phase asynchronous motors from 0.75 to 75 kW

Schneider Electric, Paris, France Schneider Electric, 2014, "Modicon M171 Performance Logic Controller, Hardware User Manual", Schneider Electric, Paris,

Zhao Xu; Ostergaard, J.; Togeby, M.; Isleifsson, F.R., "Evaluating frequency quality of Nordic system using PMU data,"

Society General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century, 2008 IEEE

-18 June 2015

Paper 1037

5

is willing to reduce its adoption relectance. Moreover the events last few seconds only, that is not willing to disturb

8Hz, 95% last less than 27s,


controlled reserves in some Next step is now to

of its robustness, and to define suitable

business models for the whole power system chain. That Schneider Electric and Grenoble INP priority up

The development of this concept has been sponsored by Utility Innovation and

the European Commission within the framework of , under grant agreement

FINAL REPORT of the Investigation Committee on the 28 September 2003 Blackout in Italy”

Frequency Control and E Definitions and Acronyms", ENTSO-

order system frequency response model incorporating load shedding: analytic modeling and applications," Power Systems, IEEE Transactions on , vol.21,

G. Lebel, D. Wang, R. Caire, N. Hadjsaid, S. Bediou and A. and Coordinated Demand Response for

PowerTech 2015 IEEE waiting for approval).


Wiser: energy management solutions e of energy efficiency", Schneider

Schneider Electric, 2011, "Altivar 212 Variable Speed Drive for phase asynchronous motors from 0.75 to 75 kW, Catalogue

Performance Logic anual", Schneider Electric, Paris,

Zhao Xu; Ostergaard, J.; Togeby, M.; Isleifsson, F.R., "Evaluating frequency quality of Nordic system using PMU data," Power and

Conversion and Delivery of Electrical Energy in the 21st Century, 2008 IEEE , vol., no.,

18 June 2015

Paper 1037

5/5

Moreover the events last few seconds only, that is not willing to disturb

8Hz, 95% last less than 27s,


controlled reserves in some Next step is now to

of its robustness, and to define suitable

hat priority up

The development of this concept has been sponsored by Utility Innovation and

the European Commission within the framework of , under grant agreement

FINAL REPORT of the Investigation

Frequency Control and -E

order system frequency response model incorporating load shedding: analytic modeling and applications," Power Systems, IEEE Transactions on , vol.21,

. Bediou and A. and Coordinated Demand Response for

PowerTech 2015 IEEE

G. Lebel, K. Kuypers, R. Caire, N. Hadjsaid, S. Bediou, and A. Glatigny, "Domestic Demand Response for Primary Control Support", CIGRE International Colloquium on Lightning and

Wiser: energy management solutions ", Schneider

for , Catalogue",

Performance Logic anual", Schneider Electric, Paris,

Zhao Xu; Ostergaard, J.; Togeby, M.; Isleifsson, F.R., "Evaluating Power and

Conversion and Delivery of , vol., no.,