Voltage-Frequency Control (v-f) of Islanded Microgrid Based on Battery and MPPT Control

With the development and increasing influence of the use of distributed generation resources, distribution networks have changed from passive networks to active networks. In these new networks, operation on an island will increase both the reliability of the network in the competitive market and the maximum utilization of distributed generation resources. But despite the various benefits, Microgrids with distributed generation resources also have disadvantages. One of the disadvantages is the increasing complexity of control systems, protection systems and operating systems of the global distribution network. In recent research, various methods have been proposed for controlling the Microgrids, especially voltage and frequency control. This paper presents a method for controlling a photovoltaic (PV) system with maximum power point tracking (MPPT) controller and battery storage to provide voltage-frequency (v-f) support in an islanded microgrid. It introduces a new algorithm for MPPT control that offers control strategies, effective coordinated between v-f control in inverter, MPPT control, and battery storage control. Finally, the proposed scheme is implemented on the IEEE 13-bus distribution feeder in islanded mode using MATLAB software, the results of which clearly demonstrate the efficiency of the control methods.


Introduction
With the increasing use of electricity worldwide, the electricity industry has faced issues such as the high cost of building new power plants and the development of transmission, distribution networks, environmental concerns, and climate change. In order to overcome these problems, increase customer service reliability and reduce congestion and loss in transmission and distribution lines, distributed and renewable energy sources are new and viable options introduced over the last two decades [1]; Renewable energy has led to their planning for greater use in the electricity industry and the increasing expansion of Microgrids in most countries worldwide [2]. Microgrids are small power networks made up of several renewable energy sources and local loads. Microgrids are normally connected to one of the distribution network buses, but in an emergency, if large disturbances occur, they are disconnected from the power network and feed some loads (critical loads). In this regard, the use of solar energy is of great importance, and PV systems have been widely used as a source of distributed generation in Microgrids [1].
Microgrids include distributed generation units, energy storage systems and electrical loads, operated as a standalone, controllable system capable of providing power and heat to local customers [2]. The Point of Common Coupling (PCC) is connected to the mains voltage through a power breaker circuit. In islanded mode, the Microgrid must be capable of controlling voltage and frequency independently of the grid and thus operate as a P-V (power-voltage) bus.
Similar to the traditional frequency control method in the synchronous generator [3], the Microgrid voltage and frequency control can also be performed using drop control methods [4][5][6][7][8]. The present study provides a faster response for voltage and frequency control than the secondary control discussed in [8]. A comparison between inverter control and synchronous generator control in an islanded microgrid is presented in detail [9]. Operating and controlling the inverter American Journal of Electrical and Computer Engineering 2020; 4(2): 35-48 36 connected to the distribution of renewable energy sources such as a PV system in a Microgrid is particularly challenging to maintain within acceptable voltage-frequency range. The voltage control method based on classical drop control for voltage drop has been studied [10].
In the field of voltage-frequency control using a PV system with MPPT control and battery storage in Microgrid, there is a lack of research. Frequency control using a PV system is studied in a Microgrid, although voltage control is not the purpose of the paper and does not include an energy storage system [11].
In summary, previous research has not performed well in view of the involvement of energy storage devices in the field of voltage control coupled with frequency tuning or the consideration of control modes in different scenarios. In the present study, taking into account these shortcomings as well as eliminating the calculation of the transition between the reference frame abc to dq0 and unlike, it attempts to provide better results than in the past.
The Continue of this paper is organized as described in Section 2, Solar Array Analytical Modeling, PV system configuration and battery modeling, outlines the relevant control algorithms in Section 3, Section 4 contains the Microgrid description understudy. The results of the simulation will be followed by a final conclusion in Section 5.

Describe the System Under Study
In this section, the various components of the system under study will be described in detail.

Solar PV Modeling
The generally accepted model for solar cells is the singlediode model [12]. The present paper uses a single-diode model for a solar cell in the Kyocera KC200GT solar array shown in Figure 1. The V-I characteristic of a solar array is obtained from Equation (1) according to Figure 1:
Where I PV and I o are represented photon and diode saturation currents, respectively. V therm is the thermal voltage of the array, R S and R Sh respectively of series and parallel resistors and α is the ideal coefficient which is usually chosen between 1 and 1.5 and is assumed to be 1 here. The characteristic V-I, KC200GT for different levels of radiation at 25°C cell temperature is shown in Figure 2. The consistency of the above graphs with the datasheet confirms the validity of the model used. The parameters of the solar panel are presented in Table 1. The PV panel understudy has 125 strands in 4 series. The maximum power point (MPP 2 ) for a panel at a radiation level of 1000 W/ m 2 and at 25°C (STC 3 ) is 200 W and therefore 100 kW for the total system. MPP will change with changes in radiation levels and cell temperature.

PV System Configuration
As shown in Figure 3, the system has a two-level configuration that uses a DC-DC converter to control MPPT. There is also battery storage for emergencies and maintaining voltage and frequency of the grid or providing critical loads. The battery is connected to the PV in order to inject or absorb the active power through a DC-DC bidirectional converter. When the battery absorbs power, the converter operates in a reduced state and when injected into the grid, in an incremental state.
The PV system is connected to the network through the Lc inductance. This inductance eliminates the distortions of the PV output current. The connection point is called PCC and its voltage is V c (t). The rest of the system is a 13-bus IEEE distribution feeder, which is modeled as an impedance station equivalent to R+jwLs. The PV source is connected to the DC inverter through the C dc capacitor. PV is the active power source and the capacitor is the reactive power source.
According to the instantaneous power definitions for a balanced three-phase system assuming no harmonics, if V t (t) and V C (t) respectively represent the PCC instantaneous voltage and the inverter output voltage, then the average active and reactive power of the PV system is P(t) and Q(t) will be calculated as follows [13]: Where α is the angle of phase V C (t) relative to the voltage of PCC. If the angle α is small, P(t) and Q(t) in Equation (2) and Equation (3) can be estimated with the first component of the Taylor series and arrive at the following equations:

Battery Modeling
Due to the variable nature and the uncertainty in solar power generation as well as the extreme load fluctuations, full-cycle lead-acid batteries are used. Battery charging limitations are also provided for 20% discharge and 80% recharge. The accurate details of the battery model are given in [14], which is an analytical type with two equations for the discharge and charge states. In this model, to model opencircuit voltage, a term for voltage and resistance polarization is considered that makes it much more accurate.
The battery charge and discharge relationships are expressed in Equation (6) and Equation (7) [15].
In Equation (7), the first and second brackets represent the polarization resistance and the polarization voltage, respectively. The size of the battery is selected so as to be able to compensate for the failure of the PV system to produce under conditions very low or zero irradiation and provide maximum power point. In this, the MPP in the STC is 100 kW and therefore the battery should be able to provide this power for up to 1 hour.
Because the battery is intended for short-term applications such as frequency control or critical load emergency response, it provides one-hour support for any other generator to handle frequency control or balance the power in the Microgrid, which will be sufficient.

Expression of Control Methods
In this section, the control methods considered for the photovoltaic system will be investigated in the present study.

MPPT Control Algorithm
Given the high cost and low efficiency of the PV system, it is desirable that the solar array is always operated at a given radiation level and temperature and at the highest possible power. This control is known as the MPPT. Operating the PV system at the maximum power point (MPP) can be used when connecting to the network to support the network by providing clean energy and avoiding the installation of costly energy storage systems. As such, MPP-sized loads will be supplied by solar power and the remainder of the electrical loads by the power network. The P-V curves of the KC200GT solar panel for different radiation levels and variable temperatures are shown in Figures 4 and 5, respectively. Figures 6 and 7 also show the relationship between the solar panel's maximum power point (MPP) and the radiation level and cell temperature, respectively. It is clear that MPP with radiation levels, directly and inversely with temperature cells, and in both cases the relationship is linear. So it is logical to use a table with linear estimation to find non-existent data.

Coordinated Control Voltage-Frequency (v-f) and MPPT and Battery
The integrated voltage-frequency (v-f) and MPPT and battery control diagrams are shown in Figures 8 and 9, respectively [15]. The method consists of a control loop for MPPT control, two different loops for controlling v-f on the inverter side, and another loop for managing battery power. The loop 1 in Figure 8 is the MPPT control on the PV array side and uses the reference maximum power point (P MPPref ), Obtained from the MPP radiation table, with the actual PV output power (P PV ). And send the corresponding error to the controller PI (PI 1 ) whose output is the DC / DC converter incremental cycle (δ*). So by changing this cycle, the solar array will always be exploited at the reference point. The relation of this control loop is given in equation (8)   Integrated v-f control diagram for PV with MPPT and battery [15]. Figure 9. Battery power control diagram [15].
In order to control the voltage on the AC side, another feedback controller, PI 2 , was used. As shown in Figure 8, in PCC loop 2 the voltage is measured and the effective value (rms) V t (t) is calculated, then this effective value is given by the reference voltage V t *(t) which can be determined by the network., Is compared and the corresponding error is given to the PI controller. The output voltage of the inverter V c *(t) is controlled so that it is in phase with the PCC voltage and its size is controlled to adjust the PCC voltage to the given level V t *(t). This control scheme can be expressed specifically in relation (9): Where, K p2 and K i2 are the proportional and integral gains of the controller, respectively. In Equation (9), the reason for adding 1 to the right of the equation is that when no power is injected into the PV system, the PV output voltage is exactly the same as the terminal voltage. Frequency control as shown in loop 3 is done by controlling the output power of the inverter side. The Microgrid reference frequency of 60 Hz is compared with the measured value and its error is given to the controller PI 3 , which controls the α 1 * phase shift to displace the voltage waveform time scale to inject sufficient active power to maintain the frequency in the nominal value. It produces 60 Hz. The equation of this controller is as follows: In loop 3 there is another controller, PI 4 . This controller balances the active power between the AC and DC side of the inverter. Its reference signal is obtained from the dynamic changes of the active power injection from the AC side of the inverter is determined by the output of PI 3 . The measured AC side power, P ACmeasured , is multiplied by a factor of 1.02 to assume the inverter efficiency of 98% so that the DC side power is 102% of the AC side power. The two powers are compared with each other and the corresponding error is given to PI 4 to obtain the α 2 * phase shift of the loop whose relationship is as follows: Based on Battery and MPPT Control According to Equation (12) to achieve the final phase change (α*) of the voltage waveform (V c1 *), respectively, the average phase change on the DC and AC side, α 1 * and α 2 *, is calculated, which is then the reference voltage signal V c * produces for inverter PWM switching scheme.
The purpose here is to consider the phase shift of the active power of both the DC and AC side, control of the DC side voltage and obtain the desired value. If α 1 * and α 2 * have values near to the controller gain, it can be assured that the DC and AC power are in balance. This is coupled to a voltage control loop that ensures that the DC side voltage is maintained by the AC side voltage at optimum value.
The controllers, as well as those described above, are integrated with the battery power control in Figure 9. To supply or absorb the active power and frequency control by the PV system, a battery is used in the PV system configuration. If the solar power is high and the power required to control the frequency is less than the P MPP , the battery will charge. If the available solar power is insufficient and the active power required to control the frequency is greater than the P MPP , the battery will supply the power required to maintain the microprocessor frequency of 60 Hz.
The battery charge/discharge control method as outlined above and also considering the battery charge state (SOC) constraints are shown in Figure 9. This form, the Battery Reference Power (P Battref ) is dynamically generated by the difference between the injector inverter active power (P inverter ) and the PV Generation Power (P PV ). The controller consists of a PI 5 controller that receives the error signal after the difference between the battery active power and the reference power. To generate the S* signal, the signal obtained from PI 5 is compared to a triangular waveform of uniform size, which is similar to Pulse Width Modulation (PWM) in inverter control. K p5 and K i5 are the proportional and integral gains of the controller, respectively, and the equation of this control scheme is obtained from: The next step is to distinguish between the battery charge and discharge mode, which results from comparing the P PV to the P inverter . If it is a P PV ≥ P inverter , the battery is in a charged state and therefore, to generate the switching signal in order to enable the DC-DC converter buck mode, the signal from the PWM and S* must pass a logical AND.
If it is P PV <P inverter , it just the opposite signal and S* passes a logic AND in order to enable DC-DC converter boost mode and switching signal generation. With this control logic, the converter is capable of operating in both directions and can effectively charge or discharge the battery whenever needed. Figure 10 shows the diagram of the feeder distribution of the 13 IEEE bus studied along with the location of the PV array. This distribution feeder consists of one substation, 13 bus, 11 lines and eight loads. The loads consist of constant impedance, constant current and constant power (ZIP) types, most of which are constant power. In the Microgrid connected mode, the station is regarded as the source at the 650 bus and 115 kV level. In island mode, a diesel generator is set in the same bus as the reference set by the central controller (CC) of the Microgrid, to provide the active power at constant value. Figure 10. The studied network [15].

Testing of the Control Algorithm (v-f) in Islanded Microgrid
To evaluate the performance of the v-f control algorithm, two different radiation levels with radiation levels of 1000 and 750 W/ m 2 are considered as Scenario 1 and Scenario 2, respectively. The gains of PI controllers are presented in Table 2. During switching from network connection to island mode, a controlled diesel generator will be required to produce a certain amount of constant power according to the central controller command. In Figures 11 to 16 the According to Figure 11 and Figure 17, during the simulation period, the diesel generator produces a constant value of 1.25 MW. These figures also show the reactive power output of the diesel generator in kVar. In the island mode, too, the diesel generator's active power output is not sufficient to meet the demand for Microgrid power.                Figure 12 and Figure 18 show the Microgrid frequency, which initially decreased to 57.8 Hz due to the imbalance between load and production. Frequency control of the PV system starts at 0.1 second, which quickly adjusts the frequency and returns it to 60 Hz in 0.2 seconds. Figure 13 and Figure 19 show the PCC voltage as per unit (pu). It can be seen that with the start of the control, the voltage is rapidly set to 1 per unit. Figure 14 and Figure 20 show the active and reactive power of the PV inverter to regulate the voltage and frequency of the Microgrid. Injection of inverter active power to adjust the frequency of 60 Hz in both scenarios is about 80 kW. When providing this amount of power required for the Microgrid, there was a difference in the PV power share and battery energy storage as shown in Figure 15 and Figure 21. In Scenario 1, the sun radiation is abundant and has the amount of 1000 W/m 2 and therefore the PV generates a maximum power of 100 kW which is higher than the value needed to maintain the frequency. That uses an extra 20 kW to charge the battery. A negative sign in battery power means charging and absorbing power.
In Scenario 2, due to reduced radiation levels, PV generates only about 75 kW in MPP, which is insufficient to maintain the Microgrid frequency of 60 Hz. Therefore, the power supply of about 5 kW is supplied by the battery as shown in Figure 21. A positive sign in battery power means the injection of active power into the grid. Figure 16 and Figure 22 illustrates the DC voltage for the two scenarios mentioned above. It can be seen that the voltage for the two scenarios is maintained at about 850 V and 650 V, respectively. Active power is generally balanced by the existence of controllers. This power balance coupled with AC side voltage control keeps the DC side voltage at a constant value which is the unique feature of the coordinated inverter control and the proposed MPPT.  Figure 23 shows the frequency of the Microgrid which initially increases abruptly and then gradually decreases and is fixed at 0.4 seconds at 60 Hz. Clearly, diesel generators take longer to recover than the combination of PV and Battery. Figure 24 is the Microgrid voltage stabilized in about 0.5 seconds at 1 per unit. Figure 25 shows the diesel generator output power and Figure 26 shows the active and reactive power of the PV inverter operated in constant PQ mode.

Contribution of Diesel Generator to v-f Control
It takes about 0.5 seconds for all the powers injected to stabilize and reach the desired level. It should be noted that the injection of PV inverter power is also affected by the diesel generator output power fluctuations in the initial seconds before reaching a steady-state.
Therefore, it is well proven that the battery and PV system without inertia of diesel generator, despite the inertia, is capable of fast and effective control of the voltage and frequency of the Microgrid.

Discussion
Today, fossil fuel restrictions and air pollution have created an incentive to use distributed generation resources and expand the technology. Power generation near the place of consumption, in addition to reducing system losses, can provide more flexibility to provide diverse services to Based on Battery and MPPT Control consumers [21][22]. One of the views proposed to effectively increase the contribution of these resources is the integration of these resources with the objectives of visibility, the proper connection between these resources and the electricity network, as well as the more efficient control of these resources. One of the ways to integrate distributed generation resources is a new concept called Microgrid. In the event of a disturbance in the network, the Microgrid is separated from the distribution network to isolate it from the disturbance in the network. This ability to create separate islands that have both production and consumption increases reliability over conventional networks and, on the other hand, provides better service, improved capacity and high security to subscribers.
Microgrid brings many benefits to consumers, including improved reliability by providing reliable power, by the island itself during power outages and improving power quality. In addition, the Microgrids reduce losses by shortening the production and consumption gap and managing blackouts, making the maintenance process very easy, as well as using renewable energy sources and renewable and clean energy and energy storage technologies, are many benefits to society. The Microgrid control system is designed to safely and efficiently control and support the system's activities in two modes: island work and work as part of the electricity distribution network. This control system can be based on a central control system or can be scattered on any island forming a Microgrid. When the island is disconnected from the main network, the control system must control the local voltage level and frequency of production, calculate the difference between the generation and consumption of active and reactive power on the island, and take the necessary measures to stabilize the grid and the microgrid. Opposite maintain faults, incidents, shutdowns, and accidents and keep subscribers in optimum condition.
But for some reason, the power distribution companies are not going to the Microgrid right now. Some of the reasons are as follows: 1. Issues and problems related to quality control such as the existence of unusual voltage levels or the frequency outside the tolerable range of equipment [19]. 2. Prevent the reconnection of the Microgrid to the main grid when there is a different voltage phase between the distribution network and the distributed generation resources in the Microgrid [16,24]. 3. Interference occurs when the island is connected to the distribution network [17]. 4. Preventing accidents and hazards to the agents of power distribution companies when operating on lines that must not be electric, that but have become electrified due to the existence of Microgrid [18]. 5. Due to safety and security issues in troubleshooting that may not be detected by the Microgrid systems [20]. 6. Prevent condemnation of electricity distribution companies in matters and circumstances beyond their control [23]. But increasingly, economic incentives, technology growth, and environmental issues are changing the configuration and current layout of the electricity network from production to distribution. This paper presents a coordinated control technique for maintaining the voltage and frequency of the network in the islanded mode.

Conclusion
In this paper, a voltage-frequency control solution for a PV system with battery storage in an island Microgrid is presented. In this method, the PV system is operated at maximum power point and the battery storage is used to inject or absorb power through a charge/discharge cycle. The results show effective coordination between the v-f inverter control, MPPT control and battery storage control.
The results of the PV system control transition from v-f to constant power mode and the transfer of diesel generator control from constant power to frequency control have been satisfying. This feature helps the controller to adapt to changes in radiation levels as well as battery storage. The above-mentioned v-f control is much faster than a diesel generator to restore voltage and frequency to nominal value with diesel generator. In addition, calculations within the abc reference framework have made it easier and faster. Therefore, in general, installing a PV system with future battery storage can be very efficient in maintaining the voltage and frequency of the Microgrid.