Location

This example model can be found in the software under the category "Renewable Energy" with the file name "Photovoltaic_Generation_System.ecf".

Description

This example shows the control operations of a PV generation system (PVGS) connected to a typical distribution feeder with active and reactive loads. The interface converter of the PV is an average model. Several tests are performed to show the PV generation both in curtailment and MPPT mode. The figure above shows the example of a distribution network with PV generation.

The PVGS has a nominal power of 750 kW and is rated at 1050 V DC link. The system features a PV array, a coupling inductor, a step-up transformer, a boost converter and an inverter. It is connected to a simplified three phase bus 22.9 kV, 60 Hz distribution system. The internal topology of the PVGS is explained in Photovoltaic Generation System (PVGS) - Average. 

The PVGS is connected to PCC via a 22.9 kV/480 V, 0.9 MVA star-delta distribution transformer. The transformer secondary is connected to the grid side converter with a 250 µH choke filter. The switching frequency is 1640 Hz (with average inverters, we do not simulate the actual switching events).  The PVGS features a PV array that is connected to an inverter via a boost DC-DC converter. The boost converter controls the PV array in two modes of operation: 1) maximum power point tracking (MPPT); and 2) curtailment. In mode 1 the boost converter duty cycle is calculated by a perturb-and-observe (P&O) MPPT algorithm to ensure maximum power extraction from the array. In the curtailment mode the duty cycle of the boost converter is controlled to follow the active power reference. The DC-DC converter is also an average model. The inverter’s function is to maintain and regulate the DC link voltage and the reactive power at their respective commanded values at the point of common coupling (PCC). 

For more details about the PVGS model please refer to Photovoltaic Generation System (PVGS) - Average.

Simulation and Results

The PV system can operate in two modes (MPPT or curtailment). When the curtailment input to the system is enabled by setting this input equal to 1. Conversely when the signal is 0, the system works in MPPT mode. When the irradiation in pu is lower than the reference active power, the system works in MPPT mode. Conversely, when the reference active power is lower than the power available in the sun, the system operates in curtailment mode when the curtailment input is enabled. If the requested curtailment power reference is larger than the power available from the PV array, the power reference is ignored. The control sequence displaying different scenarios is summarized in the table below.  The model was tested at a time-step of 50 µs.

The figures shown in results are the expected results of this example. The shown results can be obtained using the template in ScopeView (provided with the example). A point to be noted for the duty cycle results shown in the results section between the 5 s to 10 s, the duty cycle increases to curtail the output power to a reference value. The same active power can be obtained by reducing the duty cycle (from a typical power voltage curve of a PV array, a power can be obtained at two different voltages). If that is the case, the duty cycle will first reduce and then increase, while the output power of the system is the same. 

Inside the block “Model_Control” (shown in the figure here) the time delays and references values must be set as described in the table above. The curtailment mode means that the PV generation system has an available power to deliver higher than the required. Therefore, the controller must curtail the power to follow the commanded reference. The maximum irradiance of the PV Array is 1000 w/m² , this value corresponds to maximum output power available at the PV. The power available in the sun is assumed to be directly proportional to the irradiance i.e. the power available in pu is equal to the irradiance divided by the maximum irradiance.

In both modes the duty-cycle is adjusted to obtain a desired active power. It is observed in the results that the active power takes around 3 seconds to increase from 0.5 pu to 0.8 pu. It can be noted that the DC link is maintained at 1050 V. The reactive power measured is also shown against its reference. The settling time of the reactive power in following its reference depends on the reactive power regulator gains.

During the first MPPT window (operating points 1-3), the active power reference is ignored and the maximum available power at 800 W/m² is drawn from the PV array. As the curtailment signal goes high at 4 seconds (operating point 4), the active power reference is again ignored as the demanded power is greater than the available power. At 5 seconds (operating point 5), as the demanded power goes to 0.5 pu, the power drawn from the array is curtailed to meet the reference. At 6.5 seconds (operating point 6) the active power reference again goes beyond the available power, therefore, the system again operates in MPPT mode and starts to raise the active power. The active power jumps to 1 pu at 15 seconds (operating point 10) as the irradiation jumps to 1000 W/m², then as the temperature increases from 25°C to 35°C at 18 seconds (operating point 11), the active power reduces due to the temperature dependence of the model.

The measure of reactive power always follows the reference of reactive power  (operating points 1,2 and 5-9) except where the demanded power or the measured active power is high enough to violate the system ratings. That’s when the reactive power is reduced to respect the system rating (operating points 3,4, 10 and 11).

The reactive power has a limit of +/- 0.3 pu. The system follows the following PQ equation:

References

[1] J. Rocabert, A. Luna, F. Blaabjerg and P. Rodríguez, "Control of Power Converters in AC Microgrids," in IEEE Transactions on Power Electronics, vol. 27, no. 11, pp. 4734-4749, Nov. 2012.
doi: 10.1109/TPEL.2012.2199334

 [2] Gray, J.L, “The Physics of the Solar Cell”, in Handbook of Photovoltaic Science and Engineering, A. Luque, Hegedus, S., Editor. 2011, John Wiley and Sons.