Figure 8.37 shows a picture of the experimental rack containing the PV emulator. At the top of the rack the DC power supply can be observed.
The operation of the PV emulator has been tested by laboratory measurements. The used measurement system is composed by:
• a digital oscilloscope with a bandwidth of 1 GHz and a maximum sampling frequency of 2.5 GHz, 20 GS/s, i. e., the TEKTRONIX TDS7254B;
• a 100 MHz high voltage differential probe, i. e., the TEKTRONIX P5205;
• a current measurement system, including an amplifier, i. e. the Tektronix TCPA 300;
• a current probe, i. e., the Tektronix A6303.
First, the static I-V characteristics of the PV plant (described in Table 8.1 and Fig. 8.16) are determined under different uniform solar irradiance values. In particular, the irradiance levels of 350, 550, and 950 W/m2 are considered. The static experimental points on the I-V characteristics are obtained by imposing a constant solar irradiance value on the DSP board and by suitable values of the resistive load.
In Fig. 8.38 the I-V characteristics of the PV generator, deduced by the model with superimposed experimental points, obtained by the PV emulator, are shown. It is possible to observe that the PV emulator reproduces appropriately the theoretical operation of the PV generator. A slight deviation of the experimental points from the I-V curve is observed at higher irradiance.
Another test has been performed to show the ability of the PV emulator to reproduce the theoretical behavior of the PV generator even under partial shading conditions. In particular, the assembly drawn in Fig. 8.39 has been emulated, according to the method explained in Sect. 8.7.
Fig. 8.39 Representation of the assembly under partial shading condition, reproduced by the PV simulator
Figure 8.40 illustrates the I-V curves obtained by the model with the static experimental points superimposed. A good matching can be noted.
The transition between the two points belonging to the same I-V curve has been realized using the PV emulator. In particular, Test #1 described in Sect. 8.7 has been performed. In this test, solar irradiance is set equal to 550 W/m2, while the load is switched between an initial value of 50 X, nearly corresponding to the MPP, to a final value of 75 X. This load step variation is obtained by an abrupt commutation of the resistance load value using the resistive load bank.
The corresponding experimental current and voltage time domain waveforms are shown in Fig. 8.41. The obtained experimental results are in good agreement with those obtained by simulation with PLECS®. This is evident comparing Fig. 8.41 with Fig. 8.26. It should be noted that, to put in evidence the voltage variation, a scale of 20 V/div has been used in the oscilloscope and a DC offset has been set; for this reason, the reference level is not visible.
The transition between the two points belonging to different I-V curves has been realized using the PV emulator. In particular, Test #2 described in Sect. 8.7 has been performed. In this test, the load resistance is kept constant and equal to
Fig. 8.42 Current (top trace) and voltage (bottom trace) transition, corresponding to Test #2 obtained by the PV emulator
Fig. 8.43 Current and voltage AC components supplied by the PV emulator in grid-connected configuration
50 X, while a step variation of the solar irradiance from 550 to 950 W/m2 is imposed via software by the DSP-2 Terminal interface.
The corresponding experimental current and voltage time domain waveforms are shown in Fig. 8.42. Even in this case, the obtained experimental results are in good agreement with those obtained by simulation with PLECS®. This is confirmed by the comparison of Fig. 8.42 with Fig. 8.30, where the same rise time is noticeable.
A further assessment on the PV emulator operation is done to demonstrate that it is able to reproduce the typical fluctuations due to the power injected into the grid by a single-phase inverter. Such a power is expressed as:
p(t) = P0[1 + sin(2xt)] (8.46)
where x is the grid frequency in radians per second and P0 is the average power delivered at the fundamental frequency of the grid.
In order to perform this test, the PV emulator is connected to a grid inverter by a DC/DC boost converter, according to the scheme shown in Fig. 8.15. In particular, the boost converter is that described in Sect. 8.8.3; the grid inverter is a commercial device, i. e., the Sunny Boy 1100.
Figure 8.43 shows both the current and voltage AC components supplied by the PV emulator. It can be noted that the frequency of the two waveforms is equal to 100 Hz, i. e., twice the fundamental of the grid frequency f = 50 Hz). In order to
show the two waveforms superimposed, a suitable DC offset has been set in the oscilloscope. The voltage fluctuation is lessened by the presence of a DC link capacitor at the PV emulator output as it occurs in real operating conditions.
Finally, Fig. 8.44 shows the current and voltage fluctuations in grid-connected configuration, superimposed to a static I-V characteristic. It is shown in particular, the effect of the grid connection on the operating point, whose position varies with pulsation 2ю around the MPP. In this test, to highlight the voltage fluctuation, a lower DC link capacitance is used.