Yüksek Güç Faktörlü Şebeke Bağlı Bir PV Sistemin Modellenmesi ve Farklı Işınımlar Altında Kontrolü

Nowadays, energy consumption is rapidly increasing due to developing technology, whereas energy resources are rapidly exhausting. The use of non-renewable resources such as coil, natural gas and uranium e.t.c. is difficult and cost for the generation of electricity. At the same time, because of their carbon content, it causes environmental pollution and global warming. Thus, interest in the use of renewable energy sources for instance solar and wind is increasing. Solar energy is especially preferred due to endless and free solar energy. The energy obtained from the sun can vary depending on irradiation, panel temperature and pollution. There is nonlinear relationship between the voltage and current at the output of the solar panels and the power obtained from the sun changes. For these reasons Maximum Power Point Tracking (MPPT) algorithms are being developed to draw maximum energy from a solar panel. Among these algorithms, the most common and easiest application is Perturb and Observe (P&O) method. In this study, modelling and control of a high power factor grid connected solar system under different irraditions and their effects on the grid are investigated. The proposed PV system is constructed by combining a boost converter and a full bridge inverter. At the input of the system, there are many serial solar panes to provide the petitive power and voltage. The boost converter is operated with P&O algorithm to provide * Sorumlu Yazar: Yıldız Teknik Üniversitesi, Elektrik-Elektronik Fakültesi, Elektrik Mühendisliği, İstanbul, Türkiye, ORCID: 0000-0002-10309816, erdem.akboy.87@gmail.com European Journal of Science and Technology e-ISSN: 2148-2683 795 MPPT. At the same time, the inverter is operated with Average Current Mode Control (ACMC) to supply current to the grid with high Power Factor (PF). Here, both converters are controlled analogously. For this purpose, a grid connected PV system with 1 kW and 100 kHz switching frequency was established and simulated with PSIM program. In the developed system, 4 panels with 250W-1000 W/m irradition properties are connected in series. By operating the system under different irradiation and power values, the control of the system and the effects of the line were examined. In the results obtained, it was observed that under different conditions, the proposed system has a fast dynamic response time depending on the relevant control algorithms and always has high PF on the AC side.


Introduction
Nowadays, due to technological developments and the welfare level of the societies, energy consumption is increasing and energy resources are rapidly exhausting. Using resources such as coal, diesel, oil, natural gas and nuclear e.t.c. as energy sources is difficult and cost by the producers in electricity generation. In addition, due to the carbon content of the relevant resources, it causes problems such as global warming and environmental pollution. Thus, interest in renewable energy sources as solar and wind is increasing gradually. By the virtue of solar energy is an endless source, which makes it more attractive to be used in energy production. So, the attention on the solar energy systems at academic and industrial areas, has been increasing.
A solar power plant consists of several semiconductor PhotoVoltaic (PV) modules. The operation of PV modules is based on the principle that electrons gain energy when light is applied to them. Thus, basically in a PV system, solar energy is turned into electrical energy in PV modules, directly. This makes PV systems to be considired the most efficient and and well-accepted renewable system. Although a PV system has many advantages, there are many problems. The power which is generated by PV depends upon the solar irradition, panel tempature, panel pollution and load impedance. Also, the output characteristics of the PV module (V-I) is nonlineer, especially under partial shading conditions, which causes only one maximum power point at P-V curve. For these reasons, it is necessary to tracking the maximum power point in order to benefit from each PV system at the highest efficiency. For this purpose, several Maximum Power Point Tracking (MPPT) methods are built up. At these methods output can be fed with maximum power from solar input by converters. Thus, Cuk converter (Sahu et al, 2014), buck converter (Gosumbonggota, 2016) and boost (Chowdhury et al, 2017) converters are used to achieve MPPT. In terms of PV, boost converter is more attractive because constant current can be drawn from its input.
P&O method proposes to track MPP of the PV module by using fixed pertubation value. At this algorithm, the output characterictics (VPV-IPV) are measured and the output power (PPV) is calculated of the PV module. If the variation of the VPV and PPV at the same direction, the perturbation, or the next value of the VPV, is increased. Otherwise next value of VPV is decreased. This operation cycle continues to catch the MPP, periodically. So, it causes to oscillations around the MPP and it leads more losses. Also, response time of the algorithm can change due to the pertubition value. Large step size can yield fast response time but more oscillations. The good solution can be achieved by small step sizes but the tracking speed get slow (Liu and Lopez, 2004). Many new P&O methods have being developed in order to eliminate these disadvantages and achieve effective solution.
In study of (Liu and Lopez, 2004), it is aimed to accelerate the response time of the P&O system with smaller perturbation steps by peak current control method. Similarly, in the study (Jung et al, 2005) and (Sharma and Purobit, 2012), it is aimed to reduce oscillations with small perturbation steps against variable weather conditions by hysteresis current control. In (Abdourraziq et al, 2014), a new algorithm is presented with variable perturbation steps to track fast. Similar to (Abdourraziq et al, 2014), in (Jiandong et al, 2018), a new algorithm with variable perturbation steps is presented with grid tie PV inverter system. At this system, the frequency is low and Power Factor (PF) is not referred. Also, the behavior of the system under different power conditions is not studied.
In this study, a high PF grid connected PV system modelling and control under different irradiations and their effects on the grid are presented. The proposed PV system is constructed by combining the DC-DC boost converter and the full bridge inverter. At the entrance of the system, PV panels are connected in series to provide the petitive power and voltage. The boost converter is operated with the P&O algorithm to provide MPPT. At the same time, the inverter is operated with Average Current Mode (ACM) to supply sinusoidal current to the grid with high PF. Here, both circuits are controlled analogously with fast dynamic response time. For this purpose, a grid connected PV system with a power of 1 kW and a switching frequency of 100 kHz was installed and simulated with the PSIM program. In the developed system, 4 PV panels with 250 W-1000W / m 2 radiation properties are connected in series. By operating the system under different irradiation and power values, the control of the system and the effects of the line were examined. In the results obtained, it was observed that under different conditions, the proposed system has a fast dynamic response time depending on the relevant control algorithms and always has high PF on the AC side.

Basic Principle of PV
PV generation systems consist of PV cells. Besides, each PV cell consists of large area p-n junctions. When sunlight strikes on this semiconductor, the free electrons in the n junction gain energy. Through the electrons gaining energy, hole-electron pairs are formed throughout the structure. Thus, the electron-hole redundancy that occurs through the whole structure, allows current to flow through the p-n junction and generate electricity.

PV Module Analysis
An equivalent case of a PV cell is given in Fig.1. At this figure, IPC is the cell current, which is generated directly by the sun light, D is the diode equivalent to p-n junctions, RS is the serial and RSH is the parallel equivalent resistors, IPV and VPV are ouput current and voltage of the PV module, respectively.
The PV module should be implemented from the mathematical model in equations, which are derived from the equivalent circuit given in Fig. 1.
At these equations, S is the light intensity, SO is the light intensity under test conditions, especially under 1000 W/m 2 irradiation, Eg is band energy, Ta is room temperature, IO is reverse saturation current, Ct is temperature coefficient, Tref is temperature under standart test conditions, q is electron charge (1,6 10 -16 C), A is ideality factor, ISCO is short citrcuit current of each solar cell at Tref, Ks describes how the light intensity affects the cell temperature, ks is Boltzman constant (1,3806505 10 -23 J/K). So, the ouput characteristics of the PV such as I-V and P-V curves depend on the internal characteristics of the solar module and on outside influences such as the temperature and irradiation level as given in Fig. 2. (Christopher and Ramesh, 2013 ) a) b)

Boost Converter
A boost converter circuit scheme is given in Fig 3.This converter combines of Lb, Sb and Db. At this converter, Vin and Vout are input and output voltages, respectively. The relationship between input and output is given as below.  Figure 3. The circuit scheme of conventional boost converter In this circuit either Sb or Db is on state under ideal conditions. So, there are two operation modes. When the switch is on state, diode is off state and Lb is charged via input, the load is fed by Co. When the diode is on state, the switch is off state and input and Lb transfer their energies to the output together. At this converter, semiconductors are exposed to output voltage.
Most PV systems require a large capacitor at the output in parallel with panel to achive low voltage ripple. However, the large capacitor reduces the dynamic response of the MPPT systems according to varying athmospheric conditions (Liu and Lopes, 2004). In conventional boost converters inductance is at the input, the input current remains constant and ripples of the input voltage are low. This provides ease of the control, system reliability and minimum size capacitor.

Full Bridge Inverter
Inverters convert DC input to AC output. For this purpose, the DC voltage is applied to the output as positive in the first half and negative in the second half. Sum of the time duration of these periods gives the inverter operation frequency. So, at the grid connected systems, the frequency is equal to the line frequency. A conventional grid tied full bridge inverter scheme is given in Fig 4. In this figure, Vd is input voltage, S1 and S3 are positive and S2 and S4 are negative switches, iac and vac are line current and voltage, respectively. Lac is the filter inductance to smooth line current.  In prinsible, there are two operation modes for this inverter. S1 and S2 are on state in the first half of vac and iac is positive. S3 and S4 are on state in the second half of vac and iac is negative. There is dead time between positive and negative term semiconductor control signals to avoid short circuit of DC line. Also, switches are exposed to input voltage at this inverter scheme.

Perturb and Observe Method (P&O)
Perturb and Observe (P&O) method is the most known MPPT method. In prinsible, P&O algorithm takes output voltage and power datas of the PV periodically, and operates by perturbing the panel output voltage according to the variance in these datas. The flowchart and the truth table of implementation of P&O algorithm are given in Fig. 5 and Table 1, respectively. This algorithm increases or decreases the output voltage of PV panel according to previous perturbation cycle.  This algorithm has some drawbacks. After the algorithm reaches the maximum power point, oscillations occurs on the MPP continuously, according to constant or varying atmospheric conditions. These oscillations cause power loss and low efficiency. This problem can be achieved with a sollution by decreasing the perturb rate, but, the tracking response may get slower. So, the optimum solution can be achieved by selecting suitable perturbation step. Thus, at this algorithm, a modulation signal for the PWM with chosing about %10 of the switching frequency is provided. For this purpose, the error is processed by a PI controller between the ref. and real average amount of the PV output terms (Liu and Lopes, 2004).

Average Current Mode Control
Average Current Mode Control (ACMC) is generally used in Power Factor Correction (PFC) converters. Briefly, this control is intended to settle the actual sensed inductance current at a predetermined current value under constant frequency. So, it requires current amplifier circuit. Also, this control method provides the line current to be in Continuous Current Mode (CCM). The control scheme block diagram is given as follows. It can be seen from the figure that it requires two control loops as voltage and current. At grid tied inverter applications, the line voltage (vac) is sensed firstly to determine reference current value (iref) of the control. Then, the sensed voltage is rectified via diodes and rectified current reference (irefd) is obtained. Besides, the real inductance or line current is sensed with a suitable current sensor(ireal) and rectified (ireald). The current error (ierr) is obtained by passing the measured and determined current values through the current amplifier circuit. This error is compared with sawtooth or triangular waveform (vsawt) whose frequency is equal to switching frequency to obtain PWM signals. The current waveform of the line according to this control method is as follows.

Simulation Results
The block scheme of the proposed grid tied solar system is given in Fig. 8  The circuit sheme of the grid tied inverter is given in Fig. 9 with using PSIM simulation programme. The PV module parameters are given in Table 2 under 1000W/m 2 irradiance and 25 °C tempature. In this circuit four PV modules are connected serial to obtain 1 kW.   Figure 9. The circuit scheme of the proposed grid tied PV system in PSIM At this converter, variation in irradiation has been done for specific rate of iteration (C=0.3 V) and thus tracking of power is achieved in DC side and high PF is achieved with ACMC in the AC side. For a more detailed analysis, the system has been tested based on varying irradiations at specific time intervals. In the first test, the irradiation values are increased from 200 W/m 2 as follows. AC line current waveform is given in Figure 10. It can be seen from this figure, the system response for varying irraditions is better at higher values. In low irradition values system can not achive high PF, although MPPT can be achieved. PF above 600 W/m 2 is 0,99 and line current THD is %3. In the second test, the irradiation values are decreased from 1000 W/m 2 as follows and AC line current waveform is given according to varying irradition. It can be seen from this figure, the system response is better than increasing values. PF for all irradiation values is 0,99 and MPPT is achieved perfectly.

Figure 11. The AC line current under decreasing irradition values
In the third test, the irradiation values are inreased and decreased as follows and AC line current waveform is given according to varying irradition. It can be seen from this figure, the system response is better for both conditions. This is suitable for partial shading conditions. PF for all irradiation values is 0,99 and MPPT is achieved perfectly.

Conclusion
Solar energy is especially preferred due to endless and free solar energy. The energy obtained from the sun can vary depending on irradiation, panel temperature and pollution. In this study, a high PF grid connected PV system model and control of under different irradiations and their effects on the grid are presented. The proposed PV system is constructed by combining a boost converter and a full bridge inverter. The boost converter is controlled by P&O algortihm to provide MPPT and the inverter is controlled with ACMC to achieve high PF in the AC side. Both control algorithms are achieved analoguosly. To definethe performances of the control algortihms, three tests are applied according to varying irraditions. For this purpose, the comparative analysis for a grid connected solar system with 1 kW power and a switching frequency of 100 kHz was installed and simulated with the PSIM program. Here, fast dynamic response time under varying irraditions are achieved under three test conditions. Also, at this proposed converter, low THD values are obtained for line current. So, both control algorithms can be applied together to achieve MPPT and high PF.