ARCHITECTURE AND PRINCIPLES OF SMART GRIDS

FOR DISTRIBUTED POWER GENERATION

AND DEMAND SIDE MANAGEMENT

Yonghua Cheng

VITO - Flemish Institute for Technological Research, Boeretang 200, B-2400 Mol, Belgium

EnergyVille, Dennenstraat 7, B-3600 Waterschei, Genk, Belgium

Keywords: Smart Grids, Agent Based Control, Distributed Power Generation, Demand Side Management, Power

Converters, Super Capacitors, Auction Based Market and Tariff Based Market.

Abstract: The main goals of smart grids are to provide an interface for fair transaction of electricity and to optimize

the power flow in electric power networks with less required extra energy storages, particularly in case of

integration of renewable energy sources (e.g. photovoltaic and wind) and the plug-in Hybrid Electric

Vehicles. Currently, the control principles in smart grids are mainly market-oriented (e.g. agent-based

control and event based control), which do not really take into account the constrains of electric power

networks. Moreover, the response time of coordination and control via ICT infrastructure might be

significant (few seconds to several minutes). Therefore, the architecture and control principles of smart

grids have been enhanced and presented. Particularly the concept of virtual agent has been introduced,

which interacts the business model of smart grids (e.g. agent based control) to optimize the power flow.

Additionally the time shift-able sources/loads of office buildings (e.g. plug-in hybrid electric vehicles) are

treated as another means of grid control. The evaluation results verify the architecture and control principles

which are presented in this paper.

1 INTRODUCTION

Nowadays the scale of integration of RES

(renewable energy sources) as distributed power

generation (Carrasco et al., 2006; Roman et al.,

2006; Romero-Cadaval et al., 2009; Timbus et al.,

2009; Cheng and Lataire, 2005) and plug-in HEVs

(hybrid electric vehicles) as mobile loads [6] into the

existing electric power networks is gradually

increasing to target the goal of 20, 20 and 20. Due to

the nature of intermittent power production of RES

(e.g. photovoltaic and wind energy) and the un-

predictable but maybe high power demand from

Plug-in HEVs, there will be notable consequences in

the electric power distribution networks (Cheng,

2011; Ueda et al., 2008; Morren and de Haan, 2009;

Bletterie and Pfajfar, 2007; Souto Perez et al., 2007).

For instance, over voltage, voltage flicker, poor

reliability and other power quality problems.

Recent years, the technologies of smart grids are

quickly developed to coordinate and control power

systems via ICT infrastructures (e.g. power

shedding, peak shaving) (Cheng, 2011a; Cheng,

2010; Gungoret al., 2010; Cheng, 2011b). In order to

achieve the local power balance, the time shift-able

power sources and loads are driven by micro-

economics. For example, some refrigerators,

washing machines and water boilers will be delayed

switching-on, and Micro-CHPs will start to generate

electricity, if the price of electricity is high. The

main advantages of smart grids (coordination and

control power system via ICT infrastructure) are to

create a platform for the implementation of financial

incentive and for efficiently using the existing

resources in electric power networks (e.g. energy

storages). However, the power balance based on the

ICT infrastructures is in fact to balance the average

power within each time interval (e.g. 5 min, 10min,

15 min), and it can have a long response time. Until

now, the implementation of business model does not

really take into account the constrains in the electric

power networks. Moreover, the power variation of

renewable energy sources (e.g. PV) can be almost

100% of the peak power within only few seconds in

the electric power distribution networks.

In order to limit the impact of the integration of

Cheng Y..

ARCHITECTURE AND PRINCIPLES OF SMART GRIDS FOR DISTRIBUTED POWER GENERATION AND DEMAND SIDE MANAGEMENT.

DOI: 10.5220/0003949700050013

In Proceedings of the 1st International Conference on Smart Grids and Green IT Systems (SMARTGREENS-2012), pages 5-13

ISBN: 978-989-8565-09-9

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

renewable energy sources and electric vehicles into

the existing electric power networks, power

difference must be limited in the different time

scales (Cheng, 2011b). The possibly time shift-able

sources and loads of office buildings and dwellings

can be used as the means of grid control. In addition,

supercapacitors can also be applied as the peak

power unit (Cheng, 2011b) and (Cheng, 2009), to

ensure the stability and reliability during the

transient state (around 1min) of the demand side

power management (which is based on ICT

infrastructure). Further islanding mode of microgrids

and seamless switching between islanding mode

and grid-connected mode are also expected (Cheng,

2009; Guerrero et al.2011, Balaguer et al., 2011). In

this case, the corresponded control principles are

very important and have been developed(Cheng,

2009; Guerrero et al.2011, Balaguer et al., 2011;

Cheng, 2004; Guerrero et al., 2008; Vasquez, 2009;

Iwanski and Koczara, 2008; Guerrero et al., Pai et

al., 2010; Zhong et al., 2011; Yuen et al., 2011;

Zhou and Francois, 2011) ).

In this paper, first the architecture of smart grids

is presented. Then the control principles of smart

grids in particular, the agent-based control and

event-based control are introduced. In addition, the

principle of virtual agent for power flow

optimisation and control in smart grids is also

explained. After that, the characteristics of smart

grids in function of electricity price are assessed.

Further the system dynamics perspective (due to the

long response time of coordination and control via

ICT infrastructure) and improvement by using the

controllable sources and loads in office buildings

(e.g. super capacitor storage and plug-in electric

vehicles) are explored. Finally, the architecture and

principle of smart grids are evaluated. The

evaluation results prove that the architecture and

principles can be applied to optimize the power flow

in the electric power distribution networks.

2 ARCHITECTURE OF SMART

GRIDS

Smart grid is the intelligent electric power network,

which can achieve the goals of fair transaction of

electricity and at the same time reducing the peak

power and maximally supplying local loads from the

distributed power generators (e.g. renewable energy

sources), to efficiently use the electric power

networks. In order to verify the benefits of smart

grids, the architecture of smart grid can be presented

as in Fig.1, where the market based operation in

distribution networks is established. In smart grids,

there must be certain percent individual power

sources and loads being time shift-able and

controllable. These sources and loads can be

represented by their agents to their preferable ARP

(access responsible party) and act in the auction-

based market; or they simply respond in the tariff-

based market. DSO (distribution system operator)

will play the role of SR (settlement responsible) in

the distribution network besides smart metering and

providing the measurements to TSO (transmission

Figure 1: Architecture of a smart grid.

SMARTGREENS2012-1stInternationalConferenceonSmartGridsandGreenITSystems

system operator) for the existing market in

transmission networks. In Fig. 1, the possible

sources and loads of office buildings can be little big

renewable energy systems and electric vehicle

charging station as in Fig.2. The DC/AC inverter can

be thought as the time shift-able AC sources/loads,

while there are also time un-shiftable AC

sources/loads in the office building.

Figure 2: The possible sources and loads of office building.

elec-

cooker

Frigo

elec-

boiler

washing

machine

micro-

CHP

energy

storage

time shiftabletime un-shiftable

Figure 3: The possible sources and loads of dwelling.

In Fig. 1, the possible sources and loads of

dwellings can be photovoltaic (PV), small wind

turbine (WT), micro-CHP, lamps, TV, electric-

cooker, refrigerator, electric-boiler, washing

machine and energy storage as in Fig.3. Some of

them are time shift-able (right part), while others are

not (left part).

This emulated smart grid (as in Fig. 1) can be in

synchronous (K1=on and K2=off) or in

asynchronous connection (K1=off and K2=on) with

the main grid. The high power rating PV or/and

wind turbine (PV/WT park) can also be integrated.

The power flow from these renewable energy

systems is the disturbance in the grid control.

Therefore, the control principles of smart grids are

very important in the different time scales for

distributed power generation and demand side

management.

3 PRINCIPLE OF SMART GRIDS

If there are more generators and loads are time shift-

able, then the electric power networks become more

controllable. This is being achieved by government

incentive now and by market based operation model

in near future. The time shift-able sources and loads

of office buildings and dwellings can be in agent-

based control or in event-based control. In order to

simplify the presentation in this paper, the

controllable sources &loads of office buildings are

assumed to be in agent-based control for auction-

based market operation; and the controllable sources

&loads of dwellings are assumed to be in event-

based control only for tariff-based market operation.

For the purpose of settlement responsible and

optimisation of power flow in smart grids, a virtual

agent is also presented from DSO as explained in

sub-section3.3.

3.1 Agent-based Control in Auction

Based Market

The goal of agent-based control in auction based

market is to achieve the equilibrium between the

demanded quantity and the supplied quantity in

price-quantity pair in microeconomics way. The

principle of agent-based control in smart grids is as

in follows. Each agent of the controllable source and

load sends a quantity vs. price pair to ARP as in Fig.

1. For example, the quantity vs. price of agent1, the

quantity vs. price of agent2 and the quantity vs. price

of agent3 are shown in Fig. 4, Fig. 5 and Fig. 6

respectively.

-25

-20

-15

-10

-5

0 5 10 15 20

Price

Q u an ti ty 1

Figure 4: Quantity vs. Price pair of agent1.

-25

-20

-15

-10

-5

0 5 10 15 20

Price

Quantity2

Figure 5: Quantity vs. Price pair of agent2.

ARCHITECTUREANDPRINCIPLESOFSMARTGRIDSFORDISTRIBUTEDPOWERGENERATIONAND

DEMANDSIDEMANAGEMENT

-25

-20

-15

-10

-5

0 5 10 15 20

Price

Quantity3

Figure 6: Quantity vs. Price pair of agent3.

-100

-80

-60

-40

-20

100

02468101214161820

Price

Quantity0

Figure 7: Quantity vs. Price pair of agent0.

-20

100

120

140

160

0 2 4 6 8 10 12 14 16 18 20

Price

Q u an ti ty

demand

P roduc tion

Figure 8: Power demand and power production balance at

price=8.5.

The sent quantity vs. price pair of each agent

represents its capacity of power production and

power consumption at each time. Here, agent1 has a

zero quantity around 7 cent; agent2 has a zero

quantity around 8 cent; and agent3 has a zero

quantity around 9 cent. The maximum capacities of

power production and power consumption of these

three agents are saturated at 20 kw in this paper (but

not necessary to be the same and symmetrical in the

real case). The presented virtual agent0 in Fig. 1

sends a quantity vs. price as in Fig. 7 for the

optimization of power flow in the distribution power

network. (This will be explained in sub-section 3.3)

According to the quantity vs. price of each agent, the

server in auction based market finds the total

demand quantity vs. price of the agents and the total

production quantity vs. price of the agents. As a

result, the crossing point of the power demand vs.

price and the power production vs. the price can be

found in Fig. 8. After that the price is sent via ARP

to each agent. According to the price of electricity

and the quantity vs. price pair of each agent, an

agreement of power production or consumption of

each agent with ARP is established for next time

interval (e.g. 15min). As a result, the average power

per time interval (e.g. 15min) between the time shift-

able sources and the time shift-able loads can be

balanced in auction based market. ARP is also

responsible for the integration of renewable energy

sources. But the power production of renewable

energy sources and the power consumption of other

loads are not always predictable. Even they can be

highly fluctuated. Indeed, the goal of ARP is to get

maximum profit. Therefore, ARP will optimize the

power exchange within micro-grid as well as with

the main grid. In this case, besides the auction based

market, the tariff based market is also required to

limit the difference between the predicted power and

the measured power.

3.2 Event-based Control in Tariff

Based Market

Not all of the time shift-able sources and loads

necessarily send the quantity vs. price pair to ARP.

Some of them only need to respond on some events.

For instance, the change of electricity price or/and

the change of frequency. In this case, the

controllable sources and loads are in event-based

control. For examples, the water boiler, washing

machine refrigerator can be the event-based

controllable loads; whilst micro-CHP can be the

event-based controllable source. In general, the

demanded power will de decreased and the supplied

power will be increased if the price of electricity is

increased; or vice versa. So event-based control can

be used for reconciliation of power imbalance.

Consequently a tariff based market is created by

ARP. On the participants’ side, each moment of

switching on/off the time shift-able devices/systems

can be based on the change of the electricity price

from tariff based market or/and the change of the

frequency with consideration of the original

functionalities of these systems. In the first instance,

according to the electricity price some household

devices or systems can be switched on or off at the

right time for the financial profits. In the second

instance, according to the electricity frequency some

distributed power generators and some time shift-

table loads can also be controlled to generate or

consume a suitable power for the power system

stability(in this instance, the time scale will be in ms

SMARTGREENS2012-1stInternationalConferenceonSmartGridsandGreenITSystems

and it is based on the local measurement). One

should be aware that the average power between the

time shift-able sources and the time shift-able loads

in event based control is not necessarily equal.

3.3 Virtual Agent for Optimization of

Power Flow in Electric Power

Distribution Networks

The main interest of market operator and

participators is their maximum profit. As a

consequence, an offset of power flow is generated

due to the tariff based market, but optimized power

flow in electric power distribution network is still

not ensured. Therefore, DSO needs to take the role

of settlement responsible and take care of the

constrains in electric power distribution networks, to

ensure power quality and to minimize the losses.

In this paper, we assume that a virtual agent0 as

in Fig. 1 is presented from the distribution system

operator (DSO) to the ARPs. As the strategy of the

distribution system operator is to efficiently use the

distribution networks and to have some financial

profit, the optimization of power flow in the electric

power distribution network will be done with respect

to the power flow in the electric power distribution

networks and the real-time price of electricity in

transmission networks.

Nowadays the price of electricity in the electric

power transmission networks is transparent. For

example, Belpex is the Belgian Power Exchange for

anonymous, cleared trading in day-ahead electricity,

providing the market with a transparent reference

price. In this case, DSO will interact with the market

based operation model as in Fig. 1. According to the

measurements from the smart meter(s), DSO will

also send the quantity vs. price pairs via the virtual

agent to ARPs. As a result, there will be an

additional offset of power flow. Properly changing

the quantity vs. price pairs via the virtual agent,

DSO will optimize the power flow in smart grids.

However, one should be aware that shifting the

quantity vs. price pair in the market based operation

model is not free action. Unlikely ARP will still

have the maximum profit if DSO shift the power

flow. Maybe DSO needs to compensate the financial

losses to ARP. So there is a trade-off between the

benefit of power optimization and the cost of

shifting the quantity vs. price pair in market based

operation model. This will be a main concern of

DSO if DSO will take the role of settlement

responsible in the electric power distribution

networks.

During the execution phase of market based

operation, the demand side management will limit

the difference between the expected power and

really produced/consumed power of the distributed

power generators or loads, to ensure proper

operation in auction based market as well as

reconciliation of power imbalance in tariff based

market. In this market based operation model of

smart grids, both the market operators and

participants have the opportunities to gain their

profits and face competitions.

4 CHARACTERISTICS OF

SMART GRIDS IN FUNCTION

OF ELECTRICITY PRICE

Due to the time shift-able sources and loads under

agent-based control or event-based control, the

characteristics of smart grids will be changed. The

difference of average power (in the time scale about

several seconds to few minute) between the

generated power and the demanded power will

gradually be optimised towards to be as the

expected.

In this section the characteristics of smart grids as

the function of electricity price will be explored.

This can be done by the interaction between the

virtual agent and the server of smart grids. First the

quantity vs. price pair of the virtual agent0 is shifted

from left to right, and then it is shifted from right to

left. In fact this procedure will also assess the

controllability and reachability of the smart grids

under certain amount time shift-able loads and with

certain amount time shift-able sources. When the

quantity vs. price pair of virtual agent0 is maximally

shifted to right (increasing price) as in Fig. 9, the

maximal power production of the distributed power

generators (in agent-based control) is reached. As a

result, the power balance reaches the saturation point

at price equal to 11 cents as in Fig. 10.

-40

-20

100

0 2 4 6 8 101214161820

Price

Quantity0

Figure 9: Quantity vs. Price pair of agent0 maximally

shifting to right (increasing price).

ARCHITECTUREANDPRINCIPLESOFSMARTGRIDSFORDISTRIBUTEDPOWERGENERATIONAND

DEMANDSIDEMANAGEMENT

-20

100

120

140

160

0 2 4 6 8 10 12 14 16 18 20

Price

Q u an ti ty

demand P roduc tion

Figure 10: Power demand and production balance

saturated at price=11 (reach maximal power production).

-90

-80

-70

-60

-50

-40

-30

-20

-10

0 2 4 6 8 101214161820

Price

Quantity0

Figure 11: Quantity vs. Price pair of agent0 maximally

shifting to left (decreasing price).

-20

100

120

140

160

0 2 4 6 8 10 12 14 16 18 20

Price

Q u an tity

dema nd P roduc tion

Figure 12: Power demand and power production balance

saturated at price=5 (reach maximal power demand).

When the quantity vs. price pair of virtual agent0

is maximally shifted to left (decreasing price) as in

Fig. 11, the maximal power demand of the loads (in

agent-based control) is reached. As a result, the

power balance reaches the saturation point at price

equal to 5 cents as in Fig. 12.

In both cases, the maximal power production or

power demand in agent-based control is 60 kW,

which corresponds to the maximal capacity of power

production or consumption of agent1, agent2 and

agent3 as shown in their quantity vs. price pairs in

Fig. 4, Fig. 5 and Fig. 6. In this way, the

controllability and reachability for the optimization

of power flow in smart grids are determined.

5 SYSTEM DYNAMICS

PERSPECTIVE AND

IMPROVEMENT WITH SUPER

CAPACITOR STORAGE

Indeed, the curve of demanded power in the function

of price and the curve of supplied power in the

function of price are always shifted due to the

change of the real situation in smart grids. This can

logically be classified as the case of demand curve

shifting as in Fig. 13 and the case of supply curve

shifting as in Fig. 14, though they can be happened

in the same time.

Price

Qantity

Supply1

Demand1

Q1Q2

Demand2

Figure 13: Demand curve shifting.

Price

Qantity

Supply1

Supply2

Demand

Q1 Q2

Figure 14: Supply curve shifting.

Assuming the supplied power in the function of

price is unchanged, and if the load characteristics are

changed, then the demanded power in the function

of price will be shifted. For example, the total loads

in the smart grids can be decreased from the price

and quantity at the time t1 (P1, Q1) to the price and

quantity at the time t2 (P2, Q2) as in Fig. 13.

SMARTGREENS2012-1stInternationalConferenceonSmartGridsandGreenITSystems

Assuming the demanded power in the function of

price is unchanged, and if the total capacity of power

generators is changed, then the supplied power in the

function of price will be shifted. For example, the

total capacity of the distributed power generators in

the smart grids can be increased from (P1, Q1) to be

(P2, Q2) as in Fig. 14.

As it takes time to achieve a new equilibrium (P2,

Q2) from the old equilibrium (P1, Q1), there is a

transient state of demand side power management

via ICT infrastructure. In this case the demand side

management has to be in cooperation with the

proper operations of intelligent power converters

with energy storage system, to limit the high power

difference within short time duration, particularly if

there are large scale integration of renewable energy

sources and big un-predictable loads. Only if there is

sufficient inertia in the electric power networks,

stability, reliability and power quality can be

ensured. In this case, the possible controllable

sources and loads of office building as in Fig.2 can

be treated as means of grid control, which generate

the required damping in the electric power networks,

to allow the demand side management functioning

well via the communication network(s).

In the presented architecture of smart grid as

shown in Fig. 1, the super capacitor storage is

installed with the possibly controllable DC sources

and loads of office building as in Fig.2. Accordingly

the control principles of the converters have been

developed (Cheng, 2011b), (Cheng, 2009). The

DC/AC inverter as in Fig.2 is the distributed power

generator in smart grid. The distributed power

generators will supply the power for the local loads,

and they ensure the power quality and grid stability

in smart grids. During the transient state of demand

side management (via ICT infrastructure), power

balance in the smart grid will be achieved by the

DC/AC inverter with the possibly controllable

sources and loads of office buildings as in Fig.2.

According to the real situation in the smart grids, the

DC/AC inverter can be operated in power

despatching mode or/and in load following mode. In

the power despatching mode of the DC/AC inverter,

the peak power from renewable energy sources will

be filtered by super capacitors, then to charge the

batteries of EVs; only the moving average power

will be injected into the AC grid (via the DC/AC

inverter). In the load following mode of the DC/AC

inverter, the high power difference (on the AC side)

will immediately be compensated by the super

capacitor storage. In addition, reducing the charging

power of the on-board batteries of EVs can have

some additional power to follow the power demand

on the AC side. In this case, power balance can be

achieved in long time scale as well as in short time

scale in smart grids.

6 EVALUATION RESULTS

In order to verify the architecture and principles of

smart grids (which are presented in this paper), the

change of power vs. real-time price of electricity is

evaluated. The power flows from agent1, agent2 and

agent3 are P1, P2 and P3 respectively. In order to

simplify the presentation in this paper, only the

influence of agent-based sources and loads is shown

here. The power injection from renewable energy

sources P_res minus the fluctuated loads P_loads is

Figure 15: Evaluation results of the principle of power management in smart grids.

ARCHITECTUREANDPRINCIPLESOFSMARTGRIDSFORDISTRIBUTEDPOWERGENERATIONAND

DEMANDSIDEMANAGEMENT

used as the disturbance (P_res-P_loads) in the smart

grids as in Fig.1.

The evaluation results are presented in Fig. 15.

One can see that the maximal power of the

renewable energy sources minus other loads is

130kW, but the maximal power of agent0 (through

transformer) is only 70kW. This evaluation results

prove that the architecture and principles (in

particular the virtual agent as presented in this

paper) can optimize (flatten) the average power flow

in the electric power distribution networks.

7 CONCLUSIONS

Market based operation in smart grids will ensure

fair transaction of electricity and enable more time

shift-able sources and loads involving in the

reconciliation of power imbalance. Particularly,

agent-based control and event-based control in smart

grids will change the characteristics of electric

power networks.

However, the power balance only between the

sources and the loads in agent-based control is not

sufficient to guarantee an appropriate power flow in

the electric power distribution networks. By

introducing a virtual agent, the principle of power

management in smart grids has been enhanced and

presented in this paper. The evaluation results show

that the average power flow in case of the

integration of renewable energy sources and

fluctuated loads can effectively be optimized by

applying our method of power management.

As there is a significant response time in demand

side management (via the ICT infrastructures), short

duration energy storage might be required, to ensure

the power quality in the smart grids, if the power

fluctuation due to the renewable energy sources (e.g.

PVs) or/and other fast varied loads is significant. In

this case, the possibly time shift-able and

controllable sources and loads of office building

(e.g. plug-in electric vehicles with super capacitors)

can be treated as another means of grid control. As a

result, power balance in smart grids can be achieved

in the different time scales with less additional

energy storages.

REFERENCES

J. M. Carrasco, L.G. Franquelo, J.T. Bialasiewicz, E.

Galvan, R.C. PortilloGuisado, M.A.M. Prats, J.I.

Leon, N. Moreno-Alfonso, "Power-Electronic Systems

for the Grid Integration of Renewable Energy Sources:

A Survey," IEEE Trans. on Industrial Electronics, vol.

53, no. 4, pp. 1002- 1016, August 2006.

Roman, R. Alonso, P. Ibanez, S. Elorduizapatarietxe, D.

Goitia, "Intelligent PV Module for Grid-Connected PV

Systems," IEEE Trans. on Industrial Electronics, vol.

53, no. 4, pp. 1066- 1073, August 2006.

E. Romero-Cadaval, M. I. Milanes-Montero, E. Gonzalez-

Romera, F. Barrero-Gonzalez, "Power Injection

System for Grid-Connected Photovoltaic Generation

Systems Based on Two Collaborative Voltage Source

Invert," IEEE Trans. on Industrial Electronics, vol. 56,

no. 11, pp. 4389-4398, Nov 2009.

A. Timbus, M. Larsson, C. Yuen, "Active Management of

Distributed Energy Resources Using Standardized

Communications and Modern Information

Technolog," IEEE Trans. on Industrial Electronics,

vol. 56, no. 10, pp. 4029-4037, Oct 2009.

Y. Cheng, Ph. Lataire, “New concepts for distributed

power generation and power quality for large scale

integration of renewable energy sources,” EPE2005, in

Dresden, Germany.

A. Y. Saber, G.K. Venayagamoorthy, "Plug-in Vehicles

and Renewable Energy Sources for Cost and Emission

Reductions ," IEEE Trans. on Industrial Electronics,

vol. 58, no. 4, pp. , April 2011.

Y. Cheng, “Methods for Mitigating the Effects of

Intermittent Energy Production of Photovoltaic

Sources,” the III International Conference on Power

Engineering, Energy and Electrical Drives (IEEE

POWERENG 2011), in Malaga, Spain, on 11 -13

May, 2011.

Y. Ueda, K. Kurokawa, T. Tanabe, K. Kitamura, H.

Sugihara, "Analysis Results of Output Power Loss

Due to the Grid Voltage Rise in Grid-Connected

Photovoltaic Power Generation Systems," IEEE Trans.

on IES, vol. 55, no. 7, pp. 2744-2751, July 2008.

J. Morren, S. W. H. de Haan, “Maximum penetration level

of distributed generation without violating voltage

limits,” CIRED 20th International Conference on

Electricity Distribution, June 2009.

Bletterie, T. Pfajfar, “Impact of Photovoltaic generation on

voltage variations -how stochastic is PV,” CIRED

19th International Conference on Electricity

Distribution, May 2007.

P. Souto Perez, J. Driesen, R. Belmans, “Characterization

of the Solar Power Impact in the Grid,” International

Conference on Clean Electrical Power, ICCEP '07.

2007.

Y. Cheng , “Power Management in Smart Grids for the

Integration of Renewable Energy Resources and

Fluctuated Loads,” ICCEP, 2011a, in Ischia, Italy.

Y. Cheng, “Fault-Tolerant Resonant Converters for Highly

Efficient and Reliable Power Conversion of Solar

Panels in Smart Grids,” the 14th International Power

Electronics and Motion Control Conference, (EPE-

PEMC2010), in Ohrid, Macedonia, on 6-8 September

2010.

V. C. Gungor, Bin Lu, G.P. Hancke, "Opportunities and

Challenges of Wireless Sensor Networks in Smart

SMARTGREENS2012-1stInternationalConferenceonSmartGridsandGreenITSystems

Grid ," IEEE Trans. on Industrial Electronics, vol. 57,

no. 10, Oct 2010.

Y. Cheng, “Super Capacitor Applications for Renewable

Energy Generation and Control in Smart Grids,” ISIE,

2011b, in Poland.

Y. Cheng, “Intelligent Power Electronic Systems for the

Grid Interaction with Large Scale Integration of RES,”

the II International Conference on Power Engineering,

Energy and Electrical Drives (POWERENG 2009), in

Lisbon, Portugal, on 18-20 March, 2009.

J. M. Guerrero, J.C. Vasquez, J. Matas, L.G. de Vicuna,

M. Castilla, "Hierarchical Control of Droop-

Controlled AC and DC Microgrids—A General

Approach Toward Standardization ," IEEE Trans. on

Industrial Electronics, vol. 58, no. 1, pp. , Jan 2011.

I. J. Balaguer, Qin Lei, Shuitao Yang, U. Supatti, Fang

Zheng Peng, "Control for Grid-Connected and

Intentional Islanding Operations of Distributed Power

Generation ," IEEE Trans. on Industrial Electronics,

vol. 58, no. 1, pp. , Jan 2011.

Y. Cheng, Ph. Lataire, “Advanced control methods for the

3-phase unified power quality conditioner,” IEEE

PESC2004, in Aachen, Germany.

J. M. Guerrero, L. Hang, J. Uceda, "Control of Distributed

Uninterruptible Power Supply Systems," IEEE Trans.

on Industrial Electronics, vol. 55, no. 8, pp. 2845-

2859, August 2008.

J. C. Vasquez, R. A. Mastromauro, J. M. Guerrero, M.

Liserre, "Voltage Support Provided by a Droop-

Controlled Multifunctional Inver," IEEE Trans. on

Industrial Electronics, vol. 56, no. 11, pp. 4510-4519,

Nov 2009.

G. Iwanski, W. Koczara, "DFIG-Based Power Generation

System With UPS Function for Variable-Speed

Applications," IEEE Trans. on Industrial Electronics,

vol. 55, no. 8, pp. 3047-3054, August 2008.

J. M. Guerrero, J. C. Vasquez, J. Matas, M. Castilla, L.

Garcia de Vicuna, "Control Strategy for Flexible

Microgrid Based on Parallel Line-Interactive UPS

Systems," IEEE Trans. on Industrial Electronics, vol.

56, no. 3, pp. 726-736, March 2009.

Fu-Sheng Pai, Jiun-Ming Lin, Shyh-Jier Huang, "Design

of an Inverter Array for Distributed Generations With

Flexible Capacity Operations ," IEEE Trans. on

Industrial Electronics, vol. 57, no. 12, pp. , Dec 2010.

Qing-Chang Zhong, G. Weiss, "Synchronverters: Inverters

That Mimic Synchronous Generators ," IEEE Trans.

on Industrial Electronics, vol. 58, no. 4, pp. , April

2011.

C. Yuen, A. Oudalov, A. Timbus, "The Provision of

Frequency Control Reserves From Multiple

Microgrids ," IEEE Trans. on Industrial Electronics,

vol. 58, no. 1, pp. , Jan 2011.

Tao Zhou, B. Francois, "Energy Management and Power

Control of a Hybrid Active Wind Generator for

Distributed Power Generation and Grid Integration ,"

IEEE Trans. on Industrial Electronics, vol. 58, no. 1,

pp. , Jan 2011.

Y. Cheng, “Principles of modelling and control energy

sources in hybrid propulsion systems,” International

Journal of Electric and Hybrid Vehicles, Published by

Inderscience Publishers Ltd, 2009, vol2, no.1.

Y. Cheng, “Assessments of Energy Capacity and Energy

Losses of Super Capacitors in Fast Charging-

Discharging Cycles,” IEEE Transactions on Energy

Conversion, March 2010, Vol. 25, No.1.

Y. Cheng, J. Van Mierlo, Ph. Lataire, “Methods of

Configuring and Managing Super Capacitor Energy

Storage as Peak Power Unit,” International Journal of

European Power Electronics and Motor Drive (EPE),

Vol 18, no.4.

Y. Cheng, J. Van Mierlo, Ph. Lataire, “Test Bench of

Hybrid Electric Vehicle with the Super Capacitor

based Energy Storage,” International Review of

Electrical Engineering (IREE), Published by Praise

Worthy Prize, May-June 2008 issue, Vol.3, No.3.

ARCHITECTUREANDPRINCIPLESOFSMARTGRIDSFORDISTRIBUTEDPOWERGENERATIONAND

DEMANDSIDEMANAGEMENT