Abstract—In common practice a Microgrid can operate while connected to a MV network. When a preplanned or unplanned event like holding or occurring fault occurs in the MV network it is possible to cause the islanding state in the Microgrid. In this Paper the operation of the MV network in the islanding mode and how to control the Microgrid by using the controlling structure are investigated. In this paper the conventional droop method has been described and two new controlling methods for controlling the Microgrid are also proposed which in the first method the AC power theory and in the second one the Cascade controller are being used for the controlling of the Microgrid. For comparing these three methods with the controlling method which is proposed in reference [14], these four methods have been simulated with PSCAD/EMTDC software. In the performed simulating the inverter has been used as the bridge between DGs and Microgrid and these controlling methods are used to control the output voltage and current of the inverter. In addition for switching the inverter the SPWM and hysteresis methods are being used.

Keywords-Connected mode, DG, Droop method, Islanding mode, Microgrid, SPWM, Hysteresis Switching.

I. INTRODUCTION Micro-scale distributed generators (DGs), or micro

sources, are being considered increasingly to provide electricity for the expanding energy demands in the network. The development of DGs operated in a Microgrid also helps to reduce greenhouse gas emissions and increase energy efficiency. Distributed generation encompasses a wide range of prime mover technologies, such as internal combustion engines, gas turbines, micro turbines, photovoltaic, fuel cells, and wind power. Most emerging technologies such as Microturbines, photovoltaic, fuel cells, and gas internal combustion engines with permanent magnet generators have an inverter to interface with the electrical distribution system. In the last decades, the interest on distributed generation has been increasing, essentially due to technical developments on generation systems that meet environmental and energy policy concerns. The interconnection of distributed generation has been predominately confined to MV and HV levels, but the development of micro generation technology, the decline of its costs and the public incentives to distributed generation lead to an increased installation of micro generation in LV networks. Although there is in general a lack of regulations to frame the operation of micro generators [1], in some countries, Portugal, there is already specific legislation about micro generation. This legislation is intended to encourage the investment on micro generation, namely by subsidizing the remuneration of the electricity produced by these generators and partially funding the investment. The simple integration of micro generation in LV networks, similar to the one that is being

used to integrate distributed generation on MV networks [1], may result in technical problems on LV and MV networks (excessive voltages, increase in fault levels, voltage unbalance, overloading, etc.), namely when the penetration of micro generation becomes high [2][3]. The new concept of Microgrid emerged as a way to ease this integration, but in fact corresponds to an entirely new way of understanding LV networks, with potential benefits far beyond the easy integration of micro generation [5]-[8].

The high penetration of DGs, along with different types of loads, always raises concern about coordinated control and power quality issues. In Microgrid, parallel DGs are controlled to deliver the desired active and reactive power to the system while local signals are used as feedback to control the converters. The power sharing among the DGs can be achieved by controlling two independent quantities frequency and fundamental voltage magnitude.

A Microgrid can operate in grid-connected mode or islanded mode and hence increase the reliability of energy supplies by disconnecting from the grid in the case of network faults.

If the grid has to go in islanding mode due to fault or low-voltage, the network controllers must keep the system frequency and voltage in a desirable area leads to sending the qualified power to consumers. The Microgrid should also recover the frequency and voltage in islanding mode rapidly. When Microgrid goes to islanding mode, its last situation should be considered: if it was importing energy then after entering to the islanding mode, it has to increase the generation level to compensate the lost power; therefore in this mode the Microgrid frequency will be decreased. On the other hand, if Microgrid was exporting energy and afterwards it has gone to islanding mode, it should decrease the generation level so the grid frequency will rise.

Now a considerable research has been undertaken on the control strategy of the Microgrid. Flexible and fast controls of real and reactive power are important requirements during transient and steady-state operation of a Microgrid system in both grid-connected and islanded modes [4].

Operation during network disturbances while maintaining power quality in the islanded mode of operation, a more sophisticated control strategy for Microgrid needs to be developed. In order to warranty both quality of supply and ensuring power management supervising critical and non-critical loads. Also, the basic issue on small grids is the control of the number of microsources. Microgrid concept allows larger distribution generation by placing many microsources behind singles interfaces to the utility grid. Some key power system concepts based on power vs. frequency droops

1

Microgrid Control In Islanding And Connected Mode

methodology, voltage control and can be applied to improve system stability, enhance active and reactive support. Flexible and fast controls of real and reactive power are important requirements during transient and steady-state operation of a Microgrid system in both grid-connected and islanded modes. A Microgrid when subjected to disturbances can experience angle instability and poor voltage quality due to the presence of DG units with slow response rotating machines and DG units with power electronic converters as the interface to the utility system. To ensure stable operation during network disturbances while maintaining power quality in the islanded mode of operation, a more sophisticated control strategy for Microgrid needs to be developed.

The work described in this paper regards the evaluation, through numerical simulation, of the inverter-fed MG behavior under islanded operation and connected operation for different load conditions and using different control strategies with limiter for power. The impact of considering the primary energy source dynamics and the inverters contributions to faults are issues addressed in this paper. PSCAD Simulink was employed in order to develop a simulation platform suitable for identifying MG control requirements and evaluating MG dynamic behavior under several operating conditions. Different Micro Source (MS) technologies were modeled and are considered to operate together in the simulation platform. The Micro sources are combination of fuel cell and microturbine is considered. This paper is organized as follows, in section II shows droop controlling method, In Section III, was described proposing the suggested controlling methods. Section IV shows case study system. In Section V the Microgrid simulation and results was shown. In Section VI, the conclusions of this paper are given.

II. DROOP CONTROLLING METHOD

In the Microgrid, the DG supply can be connected to a network by using an inverter. An inverter can be tuned by frequency and voltage control.

For controlling the inverters usually the drop method is used. The active and reactive power of inverter can be controlled by means of this method in order to control the frequency and to produce voltage in a specific level. This section gives the details of the final form that the control assumes during the implementation. Hardware realization of the control has been plagued by issues of noise propagation from the analog to the digital word inside the controller. The fundamental frequency selective filter is a first tool to handle some of the higher harmonics of the noise, but the calculated values of active and reactive power, as well as the voltage magnitude still suffered from oscillations determined by random spikes in the measured quantities.

The complete control of the micro source is shown in Fig. 1. The inputs are either measurements (like the voltages and currents) or set points (for voltage, power and the nominal grid frequency). The outputs are the gate pulses that dictate when and for how long the power electronic devices are going to conduct. The inverter voltage and current along with the load voltage are measured. The voltage magnitude at the load bus and the active power injected are then calculated. When

controlling active power, there is a choice of regulating the power coming from the unit or the power flowing in the feeder where the source is connected. If the power in

The feeder is regulated then there is a need to bring in the measure of the current flowing in that branch.

The voltage is calculated from the stationary frame components of the filtered instantaneous voltages. The desired and measured values for the voltages are then passed to a dynamic block that implements the P-I response of the voltage control. The output is the desired voltage magnitude to be implemented by the gate pulse generator block. This version includes low pass filters on the P, Q, and V calculated quantities to reduce the propagation of the effects of noise. These filters have also a secondary desirable effect: they attenuate the magnitude of the 120Hz ripple that exists during unbalanced operation. This allows the control to survive conditions of load unbalance without having to take any corrective action.

This final version of the control focuses on the configuration that regulates the power injected by the unit. This is not a loss of generality, since to implement the load tracking configuration one needs just two modifications: the first is to introduce the measure of the line current to calculate feeder flow, still retaining the inverter current measure to calculate the reactive power injected by the unit. The second modification is to invert the sign of the droop coefficient of the power- frequency characteristic to take into consideration that now it is the line power that is regulated and not the power injected by the unit.

In this method frequency changes is affected by changing in active power. Fig. 2 shows the Droop of power versus frequency changes.

According to Fig. 2 it can be deduced that whenever frequency decreases in a network, controllers make the

2

Figure1. Micro source Control

Figure 2. Power versus Frequency

supplies to increase their production and it means the production level changes from p0 to p1.

This state will happen to a Microgrid whenever a fault occurs and it changes a Microgrid status into islanding state. The Microgrid reconnects to principle network meanwhile the frequency increases then controllers make the DG to decrease the production level. Equation 1 represents this controlling structure:

(1)

Reactive power control versus voltage

This control mode, changes in voltage have an interaction with changes in reactive power. According to Fig. 3, this controlling mode is represents by Equation 2:

(2)

Where E is the amplitude of the inverter output voltage; is the frequency of the inverter; ω*and E* are the frequency and amplitude at no-load, respectively; and n is the proportional droop coefficient.

III. PROPOSING THE SUGGESTED CONTROLLING METHODS

In this chapter first of all the two proposed controlling methods for controlling the Microgrid will be introduced. Then by implementing these methods on a Microgrid by using the PSCAD/EMTDC software, the results obtained by using the proposed methods will be compared with the OFT controlling method.

A. Applying the instantaneous active power theory for controlling the Microgrid:

In this method AC powers AGAGI are used for supplying the essential amount of TAVAN of load. AC powers theory was proposed by AGAGI in 1984 and is still being used in the industry. According to this theory AC powers can be calculated by using the voltage and current amount. In

equation 3 the voltage and current are in α and β coordinate which result in the amounts of p and q.

(3)

This method compensates the reactive power by active filter. So that if by using equation 4 the current harmonic values which are effective in generating the reactive power could be obtained, it would be well possible to compensate the reactive power and to eliminate harmonically current by inserting the aforementioned obtained current to system by an inverter.

(4)

This theory can also be used for controlling the level of generating in the Microgrid. If the values of active and reactive power which should be generated by DG are set in the Equation 5, the DC source of the inverter is compelled to supply these values of p and q through inverter. Therefore the source current which is built by these power values could be given to the current controlling section to insert it to the system by switching. By inserting the appropriate current to system, the active and reactive power essential for the load would be supplied and distributed. The needed DC source for generating the power is the aforementioned DG which is connected to the system through the inverter.

(5)

Now we consider the two situation of islanding state and connected in the DG. In each situation if the source power values of for generating and distributing to the network are put in Equation 5, the inverter output could be changed according to system demand. In this situation with the change of the demand of the load (change of state from connected to islanding state or vice versa), the power value and consequently the source current value in Equation 5 will change and thus for doing this change in the inverter output, the level of DG generating connected to it under the appropriate controlling method, would be changed.

3

Figure 3. Voltage versus Reactive Power

After that for switching the inverter the hysteresis controlling method could be used which has advantages like god stability, simple and economically designing and implementing and quick dynamical response. For doing so the appropriate hysteresis band in the both sides of the source current value will be considered. If the output current of the inverter crashes hits the above band, the output voltage of the inverter will diminish and if the output current crosses the under band, the inverter's switches will be activated and the output voltage will increase. Thus the current will remain in the interested band and the interested wave shape will be constructed and inserted to the system.

By using the described controlling method, the active and reactive power generated by DG in Microgrid can be controlled and the supply and demand in the power network with Microgrid in two situations of connected and islanding state will be adjusted.

B. Using the cascade controller

One of the major advantages of using the PI controller is using two PI controller simultaneously which led to a better closed-loop dynamical operation and these controllers are called cascade controller. In cascade controller two controllers are connected in such a way which one controller produces the other controller's input. The major controller in the outward loop controls the primary physical variable which in this case is frequency. The secondary controller in the inward loop reads the outward loop which is power and so controls the powers which is changing. Since the changes in the network depends on the costumers and is changing constantly, with this controlling system these changes enters the inward loop and there will be compensated and these changes will then less affects the outward loop. It could mathematically be proved that the working frequency of the controller will increase and time constant will decrease dramatically.

The advantages of using the cascade controller are:

Better controlling of the primary variable

Noise decrease of the primary variable

Increase of the retrieve speed after noise occurring

Increase of the natural frequency of the system

Dramatically decrease of the time constant of the system

Improving of the dynamical operation of the system

IV. CASE STUDY

A. The structure of the Microgrid

In the Fig. 5 the single line diagram of the tested Microgrid is shown. The Microgrid is consisted of radial distribution system which is connected to a 1/1 KV system. The distribution system is consisted of feeder 5, 2 DG unit and 3 load of 100 KVA which are modeled as RLC. The DG units are consisted of one Fuel cell unit and one micro turbine unit which are connected to the system through a DC-DC converter and an inverter.

The advantages of using the Fuel cell and micro turbine are high reliability and low emission. The dynamical structure of the fuel cell and micro turbine is described as follows.

1) DC-DC inverter: Fuel cell has a slow response and generates little voltage. Thus it cannot be connected to the load directly and output voltage before transforming DC to AC should be improved. In addition the DC output of fuel cell in is not controlled and an inverter for transforming the uncontrolled DC voltage to controlled DC voltage in an appropriate level is needed. Thus a buck-boost inverter, something between SOFC and inverter is needed. This convertor can have more or less output voltage according to Duty cycle (D). If D is less than 0.5, the inverter acts like a buck inverter, otherwise it will act like a boost inverter.In Fig. 6 the simulated model of Buck-Boost converter in the PSCAD/EMTDC is shown.

4

Figure 4. Control Structure of Cascade

Figure 5. Case Stusy

2) Inverter: The convertors of the direct current to AC current (DC-AC) are known as inverter. The task of an inverter is to transform a direct input voltage or current to an AC Symmetric output with the desired amplitude and frequency. In inverters the switching techniques are being used so that by giving on and off orders to switches, the desired output wave shapes will be generated.There are two kinds of switching techniques which can be seen in Fig. 7. The first kind which is known as open loop and are using for voltage controlling is consisted of methods like SPWM and SVM. The second one which is known as closed loop is being used for current controlling which in this method a feedback is taken from the output current for decreasing the current error. This kind contains hysteresis band control, The linear PI, Approximating regulators and et cetera. The current controlling method has some advantages over the voltage controlling method which are:

Instantly control of the current and high accuracy Protecting the peak current Very good dynamical response Compensating the effects of load characteristic

changing

Here two kinds of switching are being used. In the first section which the Droop controlling method was used, the SPWM and

in the second section, the hysteresis band controlling have been used.

a) SPWM Switching

SPWM switching is used for the inverter which is applied here. In this kind of inverter the desired output wave is obtained by comparing the desired source wave with the high frequency triangular wave which is shown schematically in Fig. 8, considering the voltage is bigger or lesser than the carrier wave, one of the positive or negative voltages would be used.

5

Figure 8. SPWM Switching

Figure 6. Simulation model of Buck-Boost Converter

Figure 7. Inverter Switching Technique

b) Hysteresis switching

Hysteresis switching model in PSCAD/EMTDC was shown in Fig. 9.

g1

g2

g3

g4

g5

g6

D +

F

-

Ia1

D +

F

-

D +

F

-

Ic1

IaIb

Ic

*1000

Ib1*

1000

*1000

B. The controlling method used in reference [14]

In this method SPWM is used for inverter's switching. The switching signals are obtained by using two control PI loops which are used for controlling the voltage and frequency values. The phase voltages and currents are moved to d-q space. Id-ref is the reference current value of the d axis and Iq-ref is the reference current value of the q axis. These currents are being compared to the real currents and the difference value is being sent to the two PI controllers. The d axis current is used for controlling the inverter current value and the q axis current is used for controlling the frequency. The outputs of the controllers are moved from the p-q space to abc space and are sent to PWM.

V. SIMULATION AND ANALYSIS

In this section the operation of DG sources in Microgrid by using proposed controlling methods in this Paper will be investigated and the obtained results are compared to obtained results in reference [14]. Here three tests have been carried out on the Microgrid which is respectively as follows:

1) The distinct operation of each DG unit : in which the DG1 and DG2 are alone responsible for supplying the load of L1 and L3 in the steady state.

2) The connected state : in this situation some of the required power in Microgridis supplied by the major network.

3) Islanding Mode: In this situation the Microgridis separated from major network and goes to islanding mode.

A. The results from the distinct operation of each DG unit

In this situation each DG unit is responsible for supplying a defined load of 100 KVA. The results of this simulation are shown in Figs 11-14. As it could be seen in each of these figures, the micro turbine has a transient state in the first moment which is because of absorbing the reactive power. The voltage profile of each unit is shown in the B part of each figure which is showing the similar response of both units for approaching the desired voltage level. As it could be seen applying the AC power theory for controlling the Microgrid have the best response and had the ability to stabilize the transient state of the micro turbine quickly. Also it can be deduced from the diagrams that using the PID controller and the AC power theory control method has the better voltage profile compared to droop method.

6

Figure 10. Controlling Method used in refrence [14]

Figure 9. Hysteresis Switching

Figure 11. distinct operation of each DG unit with Cascade Controlling Method

b) Voltage Profile Of Unit 1 & 2

a) Active & Reactive power for DG1 & DG2

B. The results from the operation of each DG unit while connecting to the major network

In this situation Microgridis connected to the major network and is exchanging energy with it. In this situation DG1 and DG2 are responsible for supplying the required power for L1, L2 and L3 loads which 73% of the required power in the connected situation is supplied by the generating units of the Microgridand 23% of it is supplied by the major network. And the controller could keep stabling the voltage within 98% of the nominal voltage.

7

a) Active & Reactive power for DG1 & DG2

b) Voltage Profile Of Unit 1 & 2

Figure 12. distinct operation of each DG unit with instantaneous active power theory Controlling Method

a) Active & Reactive power for DG1 & DG2

b) Voltage Profile Of Unit 1 & 2

Figure 13. distinct operation of each DG unit with Droop Controlling Method

a) Active & Reactive power for SOFC

c) Voltage Profile Of Unit 1 & 2

Figure 14. distinct operation of each DG unit with Controlling Method used in reference [14]

b) Active & Reactive power for Micro turbine

8

a) Active power for DG1 & DG2

c) Voltage Profile of Unit 1 & 2

b) Reactive power for DG1 & DG2

a) Active power for DG1 & DG2

b) Reactive power for DG1 & DG2 c) Voltage Profile of Unit 1 & 2Figure 16. Performance of DG units in connected mode

with instantaneous active power theory Controlling Method

Figure 15. Performance of DG units in connected mode with Cascade Controlling Method

a) Active power for DG1 & DG2

b) Reactive power for DG1 & DG2

C.

D. Investigating the operation of Microgrid in the islanding mode

In the fourth second the Microgrid in the state of B2 will cut and the Microgrid goes to islanding mode. In this situation the controllers command increase generation to the generating units in Microgrid to supply the demand. The diagram of generated power changes are shown in Figs. 19-22. In the islanding mode, the reactive power of DG units diminishes because the reactive power produced by the network is cut.

9

c) Voltage Profile of Unit 1 & 2

Figure 17. Performance of DG units in connected mode with Droop Controlling Method

a) Active power for DG1 & DG2

b) Reactive power for DG1 & DG2

c) Voltage Profile Of Unit 1 & 2

Figure 18. Performance of DG units in connected mode with Controlling Method used in reference [14]

a) Active power for DG1 & DG2

b) Reactive power for DG1 & DG2

Figure 19. Performance of DG units in islanding mode with Cascade Controlling Method

a) Active power for DG1 & DG2

0.0 1.0 2.0 3.0 4.0 5.0

-0.100

-0.050

0.000

0.050

0.100

0.150

0.200

y

P(SOFC) p(MTG) p(MAIN)

0.0 1.0 2.0 3.0 4.0 5.0

-0.030

-0.020

-0.010

0.000

0.010

0.020

0.030

y

q(SOFC) q(MTG) q(MAIN)

VI. CONCLUSIONSIn this paper we have investigate and simulated some

approaches for controlling Microgrid as follows:

1) A full description of the Droop controlling method and using it for controlling the voltage and frequency of the inverter as a bridge between DG and Microgrid.

2) Applying the AC power theory as a new approach to control the active and reactive power of Microgrid.

3) Applying the cascade controlling method for improving the responding speed of the system to load changes

4) Simulating the Droop controlling method and two new controlling methods and comparing them with each other and observing the superiority of these two controlling methods in their speed to responding to the changes in the load and reducing the time constant of the system and improving the dynamical operation of the system compared to Droop controlling method.

References

[1] R.H. Lasseter, "Microgrids and distributed generation", Journal of Energy Engineering, vol. 133, pp. 144-149, March, 2007.

[2] R.H. Lasseter, "Certs Microgrid", in System of Systems Engineering, 2007. SoSE '07. IEEE International Conference, 2007, pp. 1-5.

[3] D.C. Lopes, J., Peças, A., Tomé Saraiva, J., Hatziargyriou, N., and Jenkins, N. “Management of Microgrids.” Proc., JIEE Conf, 2003.

[4] F. Katiraei, M. R. Iravani, P. W. Lehn, “Micro-Grid Autonomous Operation During and Subsequent to Islanding Process,” IEEE Trans. Power Delivery, vol. 20, no. 1, pp.248-257, Jan. 2005.

[5] F. Katiraei, and M.R. Iravani, “Micro-Grid autonomous operation during and subsequent to islanding process,” IEEE Trans. Power Delivery, vol. 20, no. 1, Jan. 2005, pp.248-257.

[6] F. Katiraei, M.R. Iravani, “Power management strategies for a Microgrid with multiple distributed generation units,” IEEE Trans. Power Systems, vol. 21, no. 4, pp.1821-1831, Nov. 2006.

[7] Y. A.-R. I.Mohamed and E. F. El-Saadany, “Adaptive decentralized droop controller to preserve power sharing stability of

10

b) Reactive power for DG1 & DG2Figure 20. Performance of DG units in islanding mode

with instantaneous active power theory Controlling Method

a) Active power for DG1 & DG2

b) Reactive power for DG1 & DG2

Figure 21. Performance of DG units in islanding mode with Droop Controlling Method

a) Active power for DG1 & DG2

b) Reactive power for DG1 & DG2Figure 22. Performance of DG units in islanding mode

with Controlling Method used in reference [14]

paralleled inverters in distributed generation Microgrids,” IEEE Trans. Power Electron., vol. 23, no. 6, pp. 2806–2816, Nov. 2008.

[8] M. N. Marwali, J. W. Jung, and A. Keyhani, “Control of distributedgeneration systems—Part II: Load sharing control,” IEEE Trans. Power Electron.,vol.19,no.6,pp.1551–1561,Nov.2004.

[9] J. A. P. Lopes, C. L. Moreira, and A. G. Madureira, “Defining control strategies for Microgrids islanded operation,” IEEE Trans. Power Syst., vol. 21, no. 2, pp. 916–924, May 2006.

[10] J. M. Guerrero, J. Matas, L. G. de Vicuña, M. Castilla, and J. Miret, “Decentralized control for parallel operation of distributed generation

[11] inverters using resistive output impedance,” IEEE Trans. Ind. Electron., vol. 54, no. 2, pp. 994–1004, Apr. 2007.

[12] N. Killingsworth and M. Krstic, “Auto-tuning of PID controllers via extremum seeking,” American Control Conference 2005, pp. 2251-2256.

[13] J.W.Kim,H. S. Choi, and B.H. Cho, “Anovel droopmethod for converter parallel operation,” IEEE Trans. Power. Electron., vol. 17, no. 1, pp. 25–32, Jan. 2002.

[14] A. Mohamed, M. Nizam, A.A. Salam, “Performance Evaluation of Fuel Cell and Microturbine as Distributed Generators in a Microgrid”, European Journal of Scientific Research, vol.30 no.4, pp.554-570, 2009.

11