the smart transformer: impact on the electric grid and

Chair of Power ElectronicsChristian-Albrechts-Universität zu KielKaiserstraße 224143 Kiel

The Smart Transformer: impact on the electric grid and

technology challengesMarco Liserre

Chair of Power Electronics | Marco Liserre| [email protected] slide 1

Outline

From the Solid-State-Transformer (SST) to the Smart Transformer

The Smart Transformer in the electric grid: identify the LV-grid, control the load/generation, offer services to MV-grid

The technological challenges of the DC/DC converter

The Smart Transformer: a grid-tailored Solid-State-Transformer


From the Solid-State-Transformer (SST) to the Smart Transformer


Concept and Definition of SST

Definition

• by Mr. McMurray, 1968 : Electronic Transformer is a device based on solid state switches which behaves in the same manner as a conventional power transformer.

• by Mr. Brooker, 1980 : Solid State Transformer is a apparatus for providing the voltage transformation functions of a conventional electrical power transformer with waveform conditioning capability.

• Currently: Power electronic based solution to replace the standard LF transformer, with the features:– galvanic isolation between the input and the output of the converter. – active control of power flow in both directions– compensation to disturbances in the power grid, such as variations of input voltage, short-term sag or

swell. – provide ports or interfaces to connect distributed power generators or energy storage device

• Smart Transformer: Solid State Transformer with control functionalities and communication.


What is the „Smart“ Transformer ?

The Smart Transformer is:

a “power electronics based” transformer

a power system management node

a link to different ac or dc infrastructures

a possible storage-integration technology

a link to other energy sources (gas, heat, hydrogen)

a support for the EV infrastructure

The Smart Transformer relieves the demandon single renewable sources and increases

the capacity of electrical lines


The Smart Transformer

A system level optimization !


Smart Transformer services for the electric grid

• Voltage support (steady state and LVRT)

• Reactive power compensation at HV/MV substation

• Power quality improvements• Islanding control (high DG in LV)

• Integration of EV-charging stations • Integration of storage for dispatching• Reverse Power Flow limitation

Smart Transformer

• Impedance identification• Load identification• Reverse Power Flow limitation• ST overload control• Soft-load reduction• Damping of harmonics and resonances• LV-side power quality


Control of the Smart Transformer

M. Liserre, G. Buticchi, M. Andresen, G. De Carne, L. F. Costa and Z. X. Zou, "The SmartTransformer: Impact on the Electric Grid and Technology Challenges," in IEEE IndustrialElectronics Magazine, vol. 10, no. 2, pp. 46-58, Summer 2016.


The Smart Transformer in the electric grid: 1. identify the LV-grid, 2. control the load/generation, 3.

offer services to MV-grid


On Line Load Identification

𝑃𝑃 = 𝑃𝑃0𝑉𝑉𝑉𝑉0

𝐾𝐾𝑝𝑝1 + 𝐾𝐾𝑓𝑓𝑓𝑓

𝑓𝑓 − 𝑓𝑓0𝑓𝑓0

𝑄𝑄 = 𝑄𝑄0𝑉𝑉𝑉𝑉0

𝐾𝐾𝑞𝑞1 + 𝐾𝐾𝑓𝑓𝑓𝑓

𝑓𝑓 − 𝑓𝑓0𝑓𝑓0

The load can be represented with an exponential model for the voltage and with a linear dependency from the frequency

V Independent of initial voltage and does not require initializationV Only one parameter is needed for active and one for reactive power.V The exponent is equal to load sensitivity to voltage

(1𝑎𝑎)

(1𝑏𝑏)

G. De Carne; M. Liserre; C. Vournas, "On-line load sensitivity identification in LV distribution grids," in IEEE Transactions on Power Systems , vol.PP, no.99, pp.1-1


On Line Load Identification: Influence of DG on the identification

Integrating eq. (9) in eq. (4) we obtain:

𝐾𝐾𝑓𝑓 =⁄∆𝑃𝑃 𝑃𝑃0⁄∆𝑉𝑉 𝑉𝑉0

= 𝐾𝐾𝑓𝑓,𝐿𝐿𝑃𝑃𝐿𝐿

𝑃𝑃𝐿𝐿 − 𝑃𝑃𝐺𝐺(10)

The net load reacts in different way depending on the presence of DG.

Example:

𝐾𝐾𝑓𝑓 = 11

1 − 0= 1

𝑃𝑃𝐿𝐿 = 1,𝑃𝑃𝐺𝐺 = 0 → 𝑃𝑃0 = 1𝐾𝐾𝑓𝑓,𝐿𝐿 = 1

𝑃𝑃𝐿𝐿 = 1.5,𝑃𝑃𝐺𝐺 = 0.5 → 𝑃𝑃0 = 1𝐾𝐾𝑓𝑓,𝐿𝐿 = 1

𝐾𝐾𝑓𝑓 = 11.5

1.5 − 0.5= 1.5


1


1.5Linear

responseMore than linear

response


Load parameter identification with respect to Voltage and Frequency

Load identification


Soft Load Reduction

High load consumption can affect the system stability. The generators may not follow the load demand during grid contingencies.

In case of perturbations (e.g., faults) or critical conditions (e.g., devices overload), the load shedding represents an effective, although costly, solution.

The Smart Transformer can instead reduce the load consumption performing a “soft-load reduction”. Controlling the voltage amplitude in LV grid, the load power consumption can be shaped.

G. De Carne, G. Buticchi, M. Liserre, C. Vournas, “Load Control using SensitivityIdentification by means of Smart Transformer” IEEE Transactions on Smart Grid.


Soft-load reduction

∆𝑃𝑃𝐴𝐴∗ =𝑃𝑃𝐴𝐴𝑉𝑉𝐴𝐴𝐾𝐾𝑓𝑓𝐴𝐴 𝑉𝑉 − 𝑉𝑉𝐴𝐴

𝑃𝑃𝐴𝐴∗ = 𝑃𝑃𝐴𝐴𝑉𝑉𝑉𝑉𝐴𝐴

𝐾𝐾𝑝𝑝𝑝𝑝

∆𝑃𝑃𝐴𝐴∗ + ∆𝑃𝑃𝐵𝐵∗ + ∆𝑃𝑃𝐶𝐶∗ = ∆𝑃𝑃

𝑉𝑉𝑉𝑉0

= 1 +∆𝑃𝑃+ 𝑃𝑃𝐴𝐴𝐾𝐾𝑓𝑓𝐴𝐴 + 𝑃𝑃𝐵𝐵𝐾𝐾𝑓𝑓𝐵𝐵 + 𝑃𝑃𝐶𝐶𝐾𝐾𝑓𝑓𝐶𝐶

𝑃𝑃𝐴𝐴𝑉𝑉𝐴𝐴𝐾𝐾𝑓𝑓𝐴𝐴 + 𝑃𝑃𝐵𝐵

𝑉𝑉𝐵𝐵𝐾𝐾𝑓𝑓𝐵𝐵 + 𝑃𝑃𝐶𝐶

𝑉𝑉𝐶𝐶𝐾𝐾𝑓𝑓𝐶𝐶

For each phase a,b,c

Differ.

The power variation in each phase gives:

Including (2) in (3) the general equation (4) is obtained

(1)

(2)

(3)

(4)


Soft-load reduction

𝑉𝑉𝑉𝑉0

= 1 +∆𝑃𝑃

𝑃𝑃𝐴𝐴𝐾𝐾𝑓𝑓𝐴𝐴 + 𝑃𝑃𝐵𝐵𝐾𝐾𝑓𝑓𝐵𝐵 + 𝑃𝑃𝐶𝐶𝐾𝐾𝑓𝑓𝐶𝐶(5)

Simplifying (4) in (5), the voltage variation to impose in order to get the desired power variation ΔP is obtained:

Example in Figure: the reductionpower request is 5%. The load hasvoltage sensitivity coefficients varyingbetween 0.5 and 0.9 pu (plot above).

The Soft Reduction algorithm decidesto decrease the voltage of 0.08 pu(central plot).

The load shed during the consideredtime window is about 5% (plot below).


Increasing DG hosting capacity

Typical DG penetration limits in MV feeders Voltage rise during light load Compensation of sudden loss of RES power

If at least some MV feeder loads are supplied through STs ST MV converter can apply voltage control Either locally or with remote measurement

ST can also provide emergency P control Acting on the LV connected load or DG

Gao, X., G. De Carne, M. Liserre, C. Vournas. "Increasing Integration of Wind Power in MediumVoltage Grid by voltage support of Smart Transformer." EWEA 2016.


Feeder with Wind Gen

Typical Distribution feeder with Distributed Generation

Case studies for voltage regulation and DG hosting capacity


MV feeder Test results

Without ST Consider

allowed overvoltage ΔV up to 2,5%

Max penetration limit 7.5 MW


MV feeder Test results

With ST ΔV limit +2,5% Max penetration

limit 12 MW Increase of

4.5MW (60%)


Voltage control in distribution systems is normally performed by on-load tap changer transformer and/or controllable DG. We consider a smart transformer where battery-based storage capacity is added on the DC bus (ST + storage) to extend the class of ancillary service it is possible to provide.

A control strategy for a ST + storage to:1. dispatch the operation of the

underneath distribution system;2. control the voltage of the LV and MV

grids on a best effort basis by exploiting smart meters and remote terminal units measurements and/or state estimation processes.

Dispatching by means of Smart Transformer-based storage

X. Gao, F. Sossan, K. Christakou, M. Paolone, M. Liserre Concurrent Voltage Control and Dispatch of Active Distribution Networks by means of Smart Transformer and Storage, IEEE Transactions on Industrial Electronics


Dispatching the operation of the LV network

The power flow at the MV/LV interface follows a dispatch plan (average power flow at 5 minutes resolution) established the day before the operation.

[1] Sossan, E. Namor, R. Cherkaoui and M. Paolone, "Achieving the Dispatchability of Distribution Feeders Through ProsumersData Driven Forecasting and Model Predictive Control of Electrochemical Storage," in IEEE Transactions on Sustainable Energy, 2016.

When the control strategy is actuated (-- line), the power flow at the MV interface is as the dispatch plan (shaded profile), while it would be stochastic otherwise (- line).

Dispatch error from 350 to < 0.1 kWh/day.

A two-stage procedure based on [1]:1. Day-ahead procedure: The dispatch plan is

computed. 2. Real-time operation: The battery charge/discharge is controlled to compensate the mismatch between the dispatch plan and realization.


Conclusions regarding integration of storagethrough ST

A control strategy for a smart transformer with integrated storage to stack the following ancillary services: dispatch the operation of the underneath distribution system; voltage control of the MV network; voltage control of the LV network.

Simulations on the 34-bus IEEE test feeder and the CIGRE reference network for LV systems. Dispatched operation is attained with an energy error < 0.1 kWh per day, the average voltage deviation

from the reference is reduced from 4.0% to 2.5% on the MV side, and 16.0% to 8.5% on the LV. Noncomplex architecture and IT infrastructure. All the control is localized at substation level, only

smart meters measurements are required from remote units.


The technological challenges of the DC/DC converter


Challenges of the DC-DC Stage

• High voltage Isolation• High Input voltage • High output current• Galvanic Isolation in Medium/High frequency• Power flow control – dc link control• Dc breacker feature (short circuit current

proctection)

DC-DC Stage: The most challenge stage

IsolationEfficiencyCost

Deserves more attention


DC-DC Stage: Implementation Concept

• Low voltage/current rating semiconductors• Scalability in voltage/power• Fault tolerance capability• Reduced dV/dt and dI/dt

• Fewer number of components• High Voltage WBG devices• Simple control/communication system

Non-Modular Vs Modular


Challenges of the DC-DC Stage

• High Voltage Isolation• Bidirectional power flow• Galvanic Isolation in Medium/High frequency• Power flow control – dc link control• Dc breacker feature (short circuit current

proctection)

DC-DC Stage: Building Block Converter

Efficiency


Review on high efficiency dc-dc converter

Relevant converters: Phase-shift Full-Bridge Series-Resonant Converter

Dual-Active-Bridge Multiple-Active-Bridge


Target: Efficiency

Reliability

Accurate losses modeling

Automatic design - (optimum parameter selection)

Wideband gap devices

Fault tolerant topology

Lifetime devices considerations

Series-Resonant Converter


• Wideband-gap devices plays an important role• Design: correct parameters selection

Max Eff = 98.61%Eff (@Pmax) = 98.1%

Overview of basic dc-dc topologiessuitable to be used as a building block ofthe ST dc-dc stage

CAU Kiel dc-dc converter

Influence on efficiency:

Series-Resonant Converter

L. F. Costa, G. Buticchi, M. Liserre, Highly Efficient and Reliable SiC-based DC-DC Converter forSmart Transformer, in IEEE Transactions on Industrial Electronics


• Extension of the DAB with 2 additonal ports

• Operation is similar to DAB

• Phase shift modulation for power transfer

• Power transfer between all ports possible:

Quadrupole Active Bridge


Zero voltage switching range of the DAB[Alonso, 2010].

Zero voltage switching range of the QAB and reactive currents.

• ZVS range of DAB and QAB are similar under symmetrical loading

• Reactive currents are also similar under symmetrical loading

Quadrupole Active Bridge


• Wideband-gap devices plays an important role• Design: correct parameters selection

Max Eff = 97.5%

Overview of basic dc-dc topologiessuitable to be used as a building block ofthe ST dc-dc stage

CAU Kiel dc-dc converter

Influence on efficiency:

Quadruple Active Bridge

(SiC)

Highest efficiency of a MAB converter


DC/DC for the Smart Transformer

Dual/Quad Active Bridge Serie Resonant Converter

controlability

Voltage and current sensors

simplicity

LVDC link controlDAB VSI


Dual/Quab Active Bridge Serie Resonant Converter

• CHB controlsthe MVDC linkand, consequenltythe LVDC link

• DAB controls the LVDC link

DC/DC for the Smart Transformer


DAB or QAB ?Nº of Units Unit power level

(kW)Nº of CHB

cellsMV dc-

link (kV)IGBT voltage

rating (kV)Mean

current (A)

IGBT current rating (A) Cost

(U$)QAB DAB QAB DAB3 9 333.33 111.11 9 3.4 6.5

17.6

150 124026 18 166.67 55.56 18 1.7 3.3 75 194049 27 111.11 37.04 27 1.13 1.7 50 648012 36 83.33 27.78 36 0.85 1.2 50 576015 45 66.67 22.22 45 0.68 1.2 50 7200


DAB or QAB ?


Scaled Prototype:Architecture

Three-phase converterEach phase contains:QAB converter3-Cell CHB converter

Peak power: 100kWRated power of a phase unit: 30kW


The Smart Transformer: a grid-tailored Solid-State-Transformer


How to rate the ST ?

L. Ferreira Costa, G. De Carne, G. Buticchi and M. Liserre, "The Smart Transformer: A solid-statetransformer tailored to provide ancillary services to the distribution grid," in IEEE Power Electronics Magazine, vol. 4, no. 2, pp. 56-67, June 2017.


How to rate the ST ?


Summary

• Difference between SST and ST is in functionalities

• The virtuous flow: identify, regulate, use the capacity for services

• SRC is for efficiency but for ST DAB/QAB are needed to decouple MV and LV

• A grid-tailored design means ST is not rated equally in all its stages


Join the PhD Course

21-23 Feb 2018

Half time in Lab !

the smart transformer: impact on the electric grid and

Documents