Harmonic characteristics of power generation unit of centralized photovoltaic power station connected to Tibet power grid

Abstract

Centralized photovoltaic power station is an important part of building a new power system, whose power generation unit is the main equipment of the photovoltaic power station. Therefore, it is the basis of analyzing the harmonic characteristics of photovoltaic power generation to study the harmonic generation mechanism and influencing factors of power generation units and establish the harmonic equivalent model of photovoltaic power generation units. In this paper, firstly, the harmonic generation mechanism is analyzed from the aspects of sinusoidal pulse width modulation (SPWM) and dead time of inverter. Secondly, the correlation analysis of influencing factors of output power of photovoltaic power generation unit is carried out, and then the influence of output power on harmonics is analyzed. Thirdly, the harmonic characteristic model of photovoltaic power generation unit based on actual photovoltaic power station and influencing factors is constructed. Finally, the theoretical simulation of SPWM modulation and dead time is verified and analyzed, and the simulation results of harmonic characteristics are compared with the measured data. The results show that the harmonic characteristic of the centralized photovoltaic power station is mainly to generate high-frequency odd-order harmonics, which will generate extra high-frequency even-order harmonics when the irradiation is too high or too low, and cause interharmonics when the irradiance fluctuates on the DC side of the inverter.

Keywords

Photovoltaic power generation correlation analysis SPWM dead time harmonic characteristic

1. Introduction

Photovoltaic (PV) power generation is one of the main ways of new energy power generation under the new power system [1], but because the structure and operation mode of PV power station are different from the traditional power station, which consists of a large number of power electronic devices, and electronic devices generate a large number of harmonics due to their inherent characteristics, which cause serious harm to the electricity grid [2]. Therefore, it is of great significance to study the harmonic characteristics of grid-connected PV power stations in order to ensure the safe and stable operation of PV power generation.

The continuous increase of PV grid connected capacity makes the grid connected harmonic analysis of PV power generation system attract the general attention of domestic and foreign scholars. Modal analysis, harmonic linearization, time domain analysis and other mathematical methods are used to analyze system harmonics [3, 4, 5]. Chen et al. [3] built a three-phase single-stage PV grid-connected multi-inverter, and studied and summarized the resonance characteristics of the system from three aspects: the number of inverters, considering environmental factors and transmission distance, etc. by using modal analysis method. Li et al. [4] used harmonic linearization to establish positive and negative sequence impedance models of three-phase LCL grid-connected inverter, which was used to study the stability of grid-connected inverter without considering the influence of inverter harmonics. Godinez-Delgado et al. [5] combined companion-circuit analysis (CCA) and numerical differentiation (ND) in time domain analysis to predict harmonic injection at the common coupling point (PCC) between PV power generation system and AC power grid. Focuses are put on PV inverters for an improved analysis of system stability problems due to harmonic interference and control strategies used for harmonic suppression [6, 7, 8, 9, 10, 11, 12, 13, 14, 15]. Among them, Wu et al. [7] put forward the equivalent circuit model of controlled source of voltage source inverter, which can get the law of system resonance characteristics under different conditions by changing control parameters and power grid parameters. Jahan et al. [11] proposed an advanced control technology PI $+$ LLC controller based on proportional integral (PI) and lead-lag compensator (LLC) based on the control algorithm of voltage source inverter. Babu et al. [13] used the hybrid control scheme of the fuzzy logic proportional integrator derivative (FLPID) of the shunt active filter, the multiple complex coefficient filter (MCCF), and the multiple second-order generalized integrator frequency-locked loop (MSOGI-FLL) to improve the power quality of the PV grid-connected distribution system. The distributed capacitance of transmission lines and the influence of background harmonics are mainly considered, and the corresponding harmonic suppression control strategy is proposed [16, 17]. The harmonic effects and harmonic characteristics of distributed PV power generation are mainly analyzed [18, 19, 20, 21]. Based on the research background of PV system connected to traction power supply system, the problems and challenges faced by power quality such as harmonics are discussed, and the interaction of harmonics is modeled and simulated [22, 23, 24, 25]. The above-mentioned documents are domestic and foreign researches on harmonic problems caused by PV systems in recent three years, and lack of research on harmonic problems of centralized large-scale PV power plants.

In this paper, a simulation model of centralized PV power station power generation unit connected to Tibet power grid is built based on MATLAB/Simulink. Firstly, based on the case study of a PV power station with an installed capacity of 20 MW in Shannan, Tibet, the harmonic generation mechanism is analyzed from the aspects of inverter modulation mechanism and dead time, with the power generation unit as the core. Then, the correlation analysis is used to summarize the influencing factors of output power of PV inverter, and further analyze the influence of output power on harmonics. Next, the equivalent model of harmonic characteristics of PV power station based on actual inverter sinusoidal pulse width modulation (SPWM), dead zone and PV output power is constructed. Finally, the comparison between the model simulation and the measured data shows that the harmonic equivalent model established in this paper is consistent with the harmonic characteristics of the actual power station, which can provide reference for the harmonic characteristic analysis of new energy generation grid.

2. Generation units and harmonic generation mechanism in PV power stations

2.1 Structure of generation units in PV power stations

In this paper, a 20 MW PV power station in Tibet is taken as an example to introduce its power generation units structure, as shown in Fig. 1, which is mainly composed of PV panels, DC junction boxes, DC junction cabinets, inverters, box-type transformers, station loads and transmission lines.

Figure 1.

PV power generation unit structure.

In the power station, the scheme of block power generation and centralized grid connection is adopted, and the 20 MW grid-connected power generation system is divided into 20 subsystems, each of which is 1 MW connected with a 1250 kVA box transformer to form a subsystem-box transformer unit line. Wherein every 17 PV cell modules in the subsystem are connected in series as one path, and 6 paths are connected in parallel to one DC lightning protection junction box. Every 16 outgoing lines of DC lightning protection junction boxes are respectively connected to two DC junction cabinets, either is connected to a 500 kW inverter, and each two inverters is connected to a 1250 kVA box-type transformer. The unit line increases the line voltage from 0.27 kV to 35 kV through the subsystem inverter components, and every 10 unit wiring units are outgoing and sent to the 35 kV switch cabinet as one incoming line, which is incorporated into the power grid as one 35 kV overhead line. The 20 subsystems of the whole station are equipped with 20 box-type transformers to form a 20-circuit subsystem-box-type transformer unit line.

Grid-connected inverter, the power electronic device as the core of grid-connected and DC/AC conversion in PV power station, is the main source of output harmonics of power station, so its harmonic characteristic factors are more complicated than traditional power equipment such as transformers. In this paper, the output harmonic of power station is analyzed by studying inverter, and the mechanism of inverter output harmonic is analyzed from pulse width modulation (PWM) and dead-time effect.

2.2 Mechanism of PWM harmonic generation

Harmonics generated by PWM in power electronic devices such as inverters are determined by their working mechanism and cannot be avoided. According to the modulation mechanism, the harmonic voltages generated by the inverter are distributed in groups near the switching frequency, generally high-frequency harmonics. The harmonic frequency is related to the rated capacity of the inverter and the switching frequency. When the rated capacity of the inverter increases and the switching frequency decreases, the harmonic frequency decreases. The inverter of PV power station investigated in this paper is controlled by SPWM, so the harmonic generated by modulation mechanism is analyzed with SPWM strategy as the research object.

The inverter adopts bipolar SPWM. The following conclusions can be drawn by Fourier analysis of the output line voltage $U_{ab}$ of the inverter [26].

(1) Fundamental component is as Eq. (1):

$\displaystyle U_{ab}=\frac{\sqrt{3}}{2}mU_{dc}\sin\left({\omega_{r}t+\varphi+% \frac{\pi}{3}}\right)$ (1)

where,

$m=$ the modulation degree; $\varphi=$ the power-factor angle; $\omega_{r}=$ the modulated angular frequency; $U_{dc}=$ the DC side voltage.

(2) The amplitude of the harmonic component at ( $k\omega_{c}\pm n\omega_{r}$ ) is as Eq. (2):

$\displaystyle U_{\textit{invabn},m}=\frac{\sqrt{3}}{2}\frac{4U_{dc}}{n\pi}J_{k% }\left({\frac{n\pi m}{2}}\right)$ (2)

where,

$n=$ the harmonic order; $J_{k}=$ the Bessel function of the first kind; $k=$ the order; $\omega_{c}=$ the carrier angular frequency.

Equations (1) and (2) show that the amplitude of harmonics generated by SPWM modulation is related to DC side voltage and modulation index, while the frequency is related to modulation wave and carrier angular frequency.

2.3 Mechanism of dead-time harmonic generation

In order to avoid the short circuit caused by the direct connection between the upper and lower bridge arms of the inverter, the control signal of the switching device is usually used to delay for a period of time, because the action of the switching device itself has a certain delay time, resulting in a certain time difference between the actual switching signal and the ideal signal, which is called dead time [27]. Due to the dead time of the device, the actual output voltage of the inverter differs from the ideal output voltage by a pulse error voltage [28]. The average error voltage obtained by equal time voltage area method is as Eq. (3):

$\displaystyle\Delta U_{AN}=\left\{{\begin{array}[]{ll}f_{c}T_{d}U_{dc}&i_{a}>0% \\ -f_{c}T_{d}U_{dc}&i_{a}<0\\ \end{array}}\right.$ (3)

where,

$T_{d}=t_{d}+t_{on}-t_{\textit{off}}$ ; $t_{d}=$ the dead time; $t_{on}=$ the on-time of the power tube; $t_{\textit{off}}=$ the off-time of the power tube; $f_{c}=$ the carrier frequency.

During one switching period, the relationship between the average value of the actual output voltage $U_{AN}$ , the average value of the ideal output voltage $U_{AN0}$ , and the average value of the error voltage $\Delta U_{AN}$ is as Eq. (4):

$\displaystyle U_{AN}=U_{AN0}-\Delta U_{AN}$ (4)

The harmonic voltage obtained by Fourier decomposition of the error voltage formed by the dead time is as Eq. (5) [29]:

$\displaystyle U_{\textit{inva},h}=\frac{8f_{c}t_{d}U_{dc}}{n\pi}\sin[{n({% \omega_{r}t-\varphi})}]$ (5)

where,

$\varphi=$ the initial phase angle of the modulated wave; $n=$ 3, 5, 7, …; $f_{c}=$ the carrier frequency.

Equations (3)–(5) show that the harmonics generated by the dead zone are mainly related to DC side voltage, carrier frequency, dead zone time and harmonic times. When the number of harmonics increases, the harmonics generated in the dead zone will be reduced, so the influence of the lower harmonics will be mainly considered in the dead zone effect.

3. Influence factors of output power harmonics of PV power generation unit

3.1 Factors influencing output power of PV power generation unit

The output of PV power generation unit mainly depends on the active power output by inverter, whose power is influenced by many meteorological factors such as daily irradiance, ambient temperature, wind speed, humidity and air pressure. In order to determine the influence weight of each meteorological factor on the PV output power, the Pearson correlation coefficient method is adopted through correlation analysis to determine the main factors affecting the power generation capacity, and also to judge the correlation degree between each meteorological factor [30]. The so-called correlation coefficient method, that is, the correlation between two vectors x and y can be expressed by correlation coefficient, and the calculation equation of correlation coefficient is as Eq. (6):

$\displaystyle r_{xy}=\frac{\sum\nolimits_{i=1}^{n}{({x_{i}-\overline{x}})({y_{% i}-\overline{y}})}}{\sqrt{\sum\nolimits_{i=1}^{n}{({x_{i}-\overline{x}})^{2}% \sum\nolimits_{i=1}^{n}{({y_{i}-\overline{y}})^{2}}}}}$ (6)

where,

$n=$ the number of variables; $x_{i}=$ the independent variables (meteorological factors such as temperature, irradiance, air pressure, humidity and wind speed $x_{1}-x_{5}$ ); $y_{i}=$ the dependent variable (it is the output power $y$ here).

The value range of the correlation coefficient $r_{xy}$ is $|r_{xy}|\leqslant 1$ . The closer the absolute value of $r_{xy}$ is to 1, the stronger the correlation between the variables is. Generally, an absolute value above 0.8 indicates a particularly strong correlation between $x$ and $y$ , a strong correlation between 0.5 and 0.8, a weak correlation between 0.5 and 0.3, and no correlation between 0.3 and below. The plus or minus of the value indicates whether the correlation is positive or negative.

After the data of a PV power station in Tibet with a 15-minute test point on one day in August 2018 were processed, SPSS was used for correlation analysis. The results of correlation analysis are shown in Table 1.

Table 1

Correlation analysis results

Pearson correlation
Variates	$y$	$x_{1}$	$x_{2}$	$x_{3}$	$x_{4}$	$x_{5}$
$y$	1	–	–	–	–	–
$x_{1}$	0.662 ${}^{**}$	1	–	–	–	–
$x_{2}$	0.778 ${}^{**}$	$-$ 0.69 ${}^{**}$	1	–	–	–
$x_{3}$	$-$ 0.175	$-$ 0.78 ${}^{**}$	$-$ 0.214	1	–	–
$x_{4}$	$-$ 0.833 ${}^{**}$	$-$ 0.846 ${}^{**}$	$-$ 0.886 ${}^{**}$	0.358 ${}^{**}$	1	–
$x_{5}$	$-$ 0.108	$-$ 0.679 ${}^{**}$	$-$ 0.252	0.907 ${}^{**}$	0.325 ${}^{*}$	1

Note: ${}^{**}$ indicates a significant correlation at the level of 0.01 (double-tail-side); ${}^{*}$ indicates a significant correlation at the level of 0.05 (double-tail-side).

Under the condition of double-tail-side test, the correlation among variables is shown in Table 1. The output power ( $y$ ) had a strong positive correlation with ambient temperature ( $x_{1}$ ) and irradiance ( $x_{2}$ ), and a strong negative correlation with humidity ( $x_{4}$ ), but no correlation with air pressure ( $x_{3}$ ) and wind speed ( $x_{5}$ ). Furthermore, $x_{1}$ was significantly and negatively correlated with $x_{2}$ , $x_{3}$ , $x_{4}$ , $x_{5}$ at the level of 0.05; $x_{2}$ and $x_{4}$ were negatively correlated significantly at the level of 0.05; $x_{3}$ was significantly positively correlated with $x_{4}$ and $x_{5}$ at the level of 0.05; And $x_{4}$ and $x_{5}$ were significantly positively correlated at the level of 0.01.

The above analysis showed that the output power ( $y$ ) was mainly related to the environmental temperature ( $x_{1}$ ), irradiance ( $x_{2}$ ), and humidity ( $x_{4}$ ), and the correlation between the respective variables was strong.

3.2 Influence of output power of PV power generation unit on harmonics

According to the conclusion in Section 2.1, the relationship among PV output power, irradiance, temperature, humidity and harmonic voltage distortion rate of the grid point was analyzed with the help of measured data of PV power station.

As shown in Fig. 2, the output of PV power station under different irradiance and temperature is determined by irradiance, and temperature has little influence on PV output.

Figure 2.

Irradiance-temperature-PV output power diagram.

As shown in Fig. 3, the output of PV power station under different irradiance and humidity is determined by irradiance, and humidity has little effect on PV output.

Figure 3.

Irradiance-humidity-PV output power diagram.

Figure 4 is a diagram showing the relation between harmonic voltage distortion rate and PV output power, irradiance, temperature and humidity at the grid point of PV power plant, wherein the cone type represents increment, which means that harmonic voltage distortion rate increases with the increase of PV output power, irradiance, temperature and humidity.

As shown in Fig. 4, the irradiance is the main factor affecting the PV output power and the harmonic voltage distortion rate of the drawing point, followed by temperature and humidity. Therefore, the input variables of the harmonic equivalent model of PV power generation unit studied in this paper are mainly irradiance and temperature.

4. Harmonic equivalent model of PV power generation unit

PV modules and PV inverters are used as the core components in the PV power generation unit. In this section, the basic parameters and structure were used as the research basis. The modeling idea is shown in Fig. 5.

In the figure, based on the conclusion in Section 2, irradiance $S$ and temperature $T$ were used as external input variables of the PV array, i.e. model inputs. Firstly, a PV array model was established to simulate PV power generation. The inverter model was established considering dead-time effect and modulation link. Considering the actual situation of PV power station, a 1MVA power generation unit model was constructed by combining two sets of PV arrays, two inverter models and two split transformers. Finally, the line impedance of the PV power station and the load in the station were constructed to jointly build the harmonic equivalent model of the PV power station power generation unit. The specific parameters of PV module, PV inverter, double split transformer and PV power station are shown in Table 2.

Table 2
Table of PV power station parameters

Parameters	Values
PV module	The rated power of a single block is 300 W; open circuit voltage is 45.3 V; the short-circuit current is 8.79 A; the rated voltage is 36.5 V; the rated current is 8.22 A; the serial-parallel numbers are 17 and 196; the adjustment range of MPPT is 450–880 V.
Inverter	The rated power is 500 kW, which is modulated by SPWM. The carrier frequency is 1050 Hz; Dead time is 4 $\mu$ s; AC rated voltage is 270 v; the adjustment range of MPPT is 450–820 V. LC filter: L is 0.17 mH; C is 150 $\mu$ F, with the efficiency of 94%–98% (less than 10% output).
Transformer	Double split transformer: capacity 1,250/630/630 kVA; connection group D-Yn11-Yn11; the transformation ratio is 38.5 kV/0.27 kV/0.27 kV; Percentage of short-circuit impedance is 6%.
PV power station	It consists of 20 1 MVA power generation units connected in parallel with the rated capacity of 20 MVA; station load capacity is 0.5 MVA. 5.134 km LGJ-185 overhead line: line resistance is 4.42 $\Omega$ ; the reactance is 22.12 $\Omega$ ; electrical susceptance is $1.61\times 10^{-4}$ S.

Figure 4.

The relation diagram of harmonic voltage distortion rate and various factors.

Figure 5.

Modeling thought diagram.

5. Discussion on comparison between simulation and measure data

5.1 Simulation analysis of SPWM modulation

To verify the accuracy of the analysis conclusion in section 1.2, the MATLAB/Simulink simulation model shown in Fig. 6 was built. The three-phase inverter model had the DC voltage of 700 V, the fundamental frequency of the inverter of 50 Hz, and the three-phase RLC load module acting as the system load. The active load and reactive load were 1 kW and 500 Var, respectively.

Figure 6.

Simulation diagram of grid-connected inverter under bipolar SPWM modulation.

The output voltage harmonics under different parameters ( $m$ -modulation, $f_{c}$ -carrier frequency) were compared and analyzed by fast Fourier transform (FFT) in PowerGui. The four parameters were set as follows: $m=$ 0.4, $f_{c}=$ 1050; $m=$ 0.8, $f_{c}=$ 1050; $m=$ 0.4, $f_{c}=$ 2000; $m=$ 0.8, $f_{c}=$ 2000, and their corresponding FFT results are shown in Fig. 7.

Figure 7.

Harmonic analysis diagram of inverter output voltage under different parameters.

The following conclusions can be drawn by observing Fig. 7:

(1)

The comparison and analysis of the two groups of parameters (a), (b), (c) and (d) show that under the same carrier frequency, the larger the modulation degree, the larger the output voltage of grid-connected inverter and the less the output harmonic content. Further analysis shows that the output fundamental voltage amplitude of grid-connected inverter is proportional to the modulation degree $m$ , and the output harmonic content is inversely proportional to the modulation degree $m$ .

(2)

The comparison and analysis of the two groups of parameters (a), (b), (c) and (d) show that when the modulation wave is constant and the carrier frequency is small, the grid-connected inverter only outputs odd harmonics. When the carrier frequency increases, the grid-connected inverter starts to output odd and even harmonics, and the output voltage increases, but the output harmonic content decreases.

(3)

The simulation analysis is consistent with the theoretical analysis in Section 2.2.

5.2 Simulation analysis of dead zone effect

To verify the accuracy of the analysis conclusion in Section 2.3, a dead time module was added on the basis of Fig. 6, and the dead time was set to 4 $\mu$ s, as shown in Fig. 8.

Figure 8.

Simulation diagram of grid-connected inverter considering dead zone time under bipolar SPWM modulation.

The analysis method is the same as that in Section 5.1, and the output voltage harmonics under four groups of parameters were compared and analyzed. The corresponding FFT analysis results are shown in Fig. 9.

Figure 9.

Harmonic analysis diagram of inverter output voltage under different parameters.

Figure 10.

The measured environmental data were substituted into the figure.

Figure 11.

Diagram of measured results and simulation result.

As shown in Fig. 9, the amplitude of fundamental voltage of grid-connected inverter output is reduced to varying degrees, and the harmonic distortion rate of output voltage is increased to varying degrees considering the influence of dead time. The enlarged view of Fig. 9c shows that relatively obvious low-order harmonic injection occurs after the dead zone is added. The simulation results are consistent with the theoretical analysis in Section 2.3.

5.3 Simulation analysis and comparison of harmonic characteristics

According to the modeling thought in Fig. 5 and the parameters listed in Table 2, a grid-connected model of centralized PV power station generation units was built to simulate and analyze harmonic characteristics under the MATLAB environment. The input of measured environmental data is shown in Fig. 10. The measured and simulated results are shown in Fig. 11a and b.

Due to the large difference between the harmonic contents of the simulation data and the measured data, some harmonics in Fig. 11b are not obvious, so three representative moments of input data, namely 0.3 s (average irradiance), 0.5 s (lowest irradiance) and 0.8 s (highest irradiance), were selected for further comparative analysis, as shown in Fig. 12.

Figure 12.

Comparison of harmonic content of simulation and measured data under different environmental factors.

6. Discussion and deficiencies

Nowadays, the harmonic problem of centralized large photovoltaic power stations is lacking. Based on the data of the actual centralized photovoltaic power station in Tibet, we study the dynamic effects of the modulation mechanism, dead zone time, and output power on the harmonic voltage and build a specialized harmonic source model of Tibet for 1 MW photovoltaic power generation unit to analyze its harmonic characteristics.

As shown in Fig. 11a and b, the centralized PV power station mainly outputs odd-order harmonics such as 3, 5, 7, …, 43, 45 and 47, among which the high-frequency harmonics are mainly concentrated in the odd-order harmonics such as 21, 23, 25, 35, 37, 41 and 43, which are consistent with the theoretical analysis in Section 2.2.

Figure 12a–c show that with the increase of irradiance, the distortion rate of harmonic voltage increases, in which the content of higher harmonics (35, 37, 39, 43, 45, 47, 49, etc.) becomes larger, and a small amount of even harmonics is generated in addition to odd harmonics. When the irradiance is the lowest, the harmonic voltage distortion rate and the content of each harmonic wave are also greater than that of the irradiance, possibly because the inverter output power is low when the irradiance is the lowest. It is concluded that when the output power of the inverter is less than a certain proportion of the rated power, the harmonic voltage distortion rate will also increase.

Point-by-point comparison of (a), (b) and (c) reveals that the harmonic voltage distortion rate of the simulation data is obviously higher than that of the measured data, and the content of low-order harmonics in the simulation data is low, because the low-order harmonics in this model are mainly calculated according to the dead time, ignoring the influence of the background voltage harmonics on the inverter. The content of higher harmonics in the simulation data is high, because the modeling of higher harmonics filtering equipment in PV power plants is neglected in this model. In addition, it is found in the process of simulation that when the DC voltage fluctuates due to the change of irradiance, many non-integer harmonics-interharmonics are generated. In this study, the emphasis is on harmonic characteristics analysis of integer multiples of harmonics in PV power plants, so interharmonics are not studied and discussed.

According to the analysis of the harmonic characteristics of the centralized photovoltaic power station, there are two main ideas of harmonic suppression mode. The former, on the one hand, transforms the inverter itself, by improving the phase number of the rectifier device or using the high power factor rectification, so that it does not produce the harmonic itself, so as to solve the low harmonic problem. On the other hand, the energy storage device with a certain capacity is used to suppress output power fluctuations, thus solving the problems of high harmonic and interharmonic. The latter uses a filter device used to solve the low harmonic and high harmonic to make harmonic compensation to the centralized photovoltaic power station.

7. Conclusions

In this paper, the characteristics of grid-connected harmonics of the centralized PV power station in Tibet were analyzed theoretically and simulated under the plateau environment. The following conclusions are drawn:

(1)

First of all, the correlation analysis of the measured data reveals that the irradiance is the main factor affecting the PV output power and the harmonic voltage distortion rate of the drawing point, followed by the temperature and humidity.

(2)

According to the theoretical and simulation analysis, when the carrier frequency of grid-connected inverter is constant, the amplitude of the output fundamental voltage of grid-connected inverter is proportional to the modulation degree $m$ , and the output harmonic content is inversely proportional to the modulation degree $m$ . When the modulation m is constant, the output harmonic voltage of grid-connected inverter varies from odd harmonics to odd and even harmonics with the increase of carrier frequency. When considering the influence of dead time, the amplitude of the output fundamental voltage of grid-connected inverter decreases, and the harmonic distortion rate of the output voltage increases, which is mainly injected into the system with low-order harmonic components.

(3)

Through building the grid-connected simulation model of the centralized PV inverter, it is concluded that the irradiance is the key meteorological factor affecting the harmonics.

High irradiance will increase the harmonic voltage distortion rate, especially the higher harmonic parts, such as 39th, 43rd and 45th harmonics, and also will produce a small amount of even harmonics, such as 36th, 38th and 42nd harmonics.

Low irradiance will also increase the harmonic voltage distortion rate, especially in the high frequency odd and even harmonics, such as the 33rd-35th, 37th, 39th, 43rd-49th harmonics, etc.

Irradiance fluctuation will lead to voltage fluctuation on the DC side of inverter, resulting in many non-integer harmonics-interharmonics.

In practical application, it can be used as a reference to analyze the harmonic characteristics of grid-connected PV power plants in Tibet. At the same time, increasing the output power of the inverter to make it close to the rated output power and stabilizing the DC side voltage of the inverter to reduce the generation of interharmonics is also the focus of this paper in the future.

Footnotes

Acknowledgments

The work was financially supported by Interaction between Traction Load of Railway Connected to Power Network in Tibet of Sichuan-Tibet Railway (No. XZ202101ZR0107G).

References

Sheng

Qian

Luo

Song

Liu

Jiang

. Key technology of power equipment operation and maintenance for new power system and its application prospect. High Voltage Technol. 2021; 47(9): 3072-3084.

Kwofie

Mensah

Antwi

. Post commission grid impact assessment of a 20 MWp solar PV grid connected system on the ECG 33 kV network in Winneba. 2019 IEEE PES/IAS PowerAfrica; 2019. 521-526.

Chen

Wang

. Influence analysis of parallel interaction of single stage grid-connected multi-inverter photovoltaic system. J Southwest Jiaotong Univ. 2020; 55(4): 811-819.

Zhao

. Impedance modeling and characteristic analysis of three-phase LCL grid-connected inverter. Power Autom Equip. 2019; 39(7): 107-113.

Godinez-Delgado

Medina-Rios

Tapia-Sanchez

. A time-domain companion-circuit newton-based methodology applied to the harmonic analysis of electrical networks with PV systems. IEEE Access. 2020; 8: 129108-129122.

Kang

Chen

Bai

. Research and design of grid-connected photovoltaic inverter based on H∞ control in weak current network. Power Grid Clean Energy. 2021; 37(9): 108-117.

Wang

. Equivalent model of controlled source and resonance mechanism analysis of grid-connected photovoltaic inverter. Acta Energ Sol Sin. 2020; 41(1): 92-100.

Huo

Zhao

Zhang

Zhao

. Research on control strategy of photovoltaic inverter in high harmonic power grid. Renewable Energy. 2019; 37(11): 1679-1684.

Singh

Kumar

Singh

. Solar PV energy generation system interfaced to three phase grid with improved power quality. IEEE Trans Ind Electr. 2020; 67(5): 3798-3808.

10.

Bhunia

Subudhi

Ray

. Design and real-time implementation of cascaded model reference adaptive controllers for a three-phase grid-connected PV system. IEEE J Photovoltaics. 2021; 11(5): 1319-1331.

11.

Jahan

Biswas

Haq

Islam

Mahmud

MAP

Abbas

. An advanced control scheme for voltage source inverter based grid-tied PV systems. IEEE Trans Appl Supercond. 2021; 31(8): 1-5.

12.

Stonier

Murugesan

Samikannu

Venkatachary

, Kumar

Arumugam

. Power quality improvement in solar fed cascaded multilevel inverter with output voltage regulation techniques. IEEE Access. 2020; 8: 178360-178371.

13.

Babu

Guerrero

Siano

Peesapati

Panda

. An improved adaptive control strategy in grid-tied PV system with active power filter for power quality enhancement. IEEE Syst J. 2021; 15(2): 2859-2870.

14.

Faraji

Birjandi

AAM

Mousavi

GSM

Zhang

Wang

Guo

. An improved multilevel inverter for single-phase transformerless PV system. IEEE Trans Energy Convers. 2021; 36(1): 281-290.

15.

Naqvi

SBQ

Kumar

Singh

. Three-phase four-wire PV system for grid interconnection at weak grid conditions. IEEE Trans Ind Appl. 2020; 56(6): 7077-7087.

16.

Shen

Kang

Zhang

Sun

. Harmonic resonance suppression strategy for photovoltaic grid-connected system considering the influence of distributed capacitance of transmission line. Acta Energ Sol Sin. 2020; 41(1): 101-109.

17.

Zhang

Duan

Sun

Kang

. Research on harmonic resonance suppression strategy of grid-connected system based on long transmission lines. Acta Energ Sin. 2019; 40(9): 2561-2570.

18.

Zou

Zhang

. Harmonic and voltage control of high permeability distributed photovoltaic system. Smart Electr Power. 2020; 48(12): 40-45.

19.

Tan

Wen

. Research on harmonic comprehensive characteristics of feeder network with multiple power sources. Intell Electr Power. 2020; 48(2): 1-8.

20.

Sun

Jiang

. Output characteristics analysis and harmonic source modeling of grid-connected photovoltaic inverter. Electr Autom. 2020; 42(1): 71-74+78.

21.

Torquato

Tessmer Hax

Freitas

Nassif

. Impact assessment of high-frequency distortions produced by PV inverters. IEEE Trans Power Delivery. 2021; 36(5): 2978-2987.

22.

Deng

Dai

Chen

. Preliminary study on multiple constraints of photovoltaic access traction power supply system. Acta Energ Sol Sin. 2020; 41(8): 192-203.

23.

Deng

Dai

Chen

. Application and key problems analysis of photovoltaic in AC/DC traction power supply system under the background of rail transit energy Internet. Proc CSEE. 2019; 39(19): 5692-5702+5897.

24.

Chen

Wang

Han

Wang

. Review of photovoltaic power station access to rail transit traction power supply system. Power Grid Technol. 2019; 43(10): 3663-3670.

25.

Deng

Dai

Guo

Shi

Han

CBX

, et al. Harmonic interaction and adaptability analysis of photovoltaic access traction power supply system. Electr Power Autom Equip. 2019; 39(4): 181-189.

26.

Liu

. Harmonic characteristics analysis of photovoltaic power generation system and harmonic treatment technology research [Dissertation]. Hunan: Hunan University; 2019.

27.

Zhang

Zhou

Mao

Xie

. Modeling of inverter dead zone and its influence on large-scale photovoltaic grid-connected system. Acta Energ Sol Sin. 2020; 41(9): 161-168.

28.

Liu

. Effect of Dead zone on output error of voltage inverter and its compensation. Trans China Electrotech Soc. 2007; 22(5): 117-122.

29.

Xie

. Research on power quality analysis and compensation system of large photovoltaic power station [Dissertation]. Hunan: Hunan University; 2015.

30.

Cheng

Sun

. Influence of meteorological factors on photovoltaic power generation based on PRINCIPAL component regression analysis. Acta Energ Sol Sin. 2021; 42(2): 403-409.

Harmonic characteristics of power generation unit of centralized photovoltaic power station connected to Tibet power grid

Abstract

Keywords

1. Introduction

2. Generation units and harmonic generation mechanism in PV power stations

2.1 Structure of generation units in PV power stations

3.1 Factors influencing output power of PV power generation unit

Table 2 Table of PV power station parameters

5.1 Simulation analysis of SPWM modulation

7. Conclusions

Footnotes

Acknowledgments

References

Table 2
Table of PV power station parameters