# 9.4 Intermittency data analysis - UQ 1.2 MW PV system

To analyse PV power output ramp rates for a relatively large-scale solar system, 1-minute data was gathered for a 1.22 MW PV system located at The University of Queensland’s (UQ) St Lucia campus in Brisbane^{9}. This is currently Australia’s largest flat-panel PV solar power system comprising 5,004 panels on the rooftops of four of UQ’s biggest buildings. Solar data containing solar irradiance (global horizontal irradiance) and PV output power sampled every 1-minute for a period of five months in late 2011 was obtained for intermittency analysis. A breakdown of the peak power ratings for the solar arrays on the four different UQ buildings is:

- UQ Centre, St Lucia - 433.4 kWp
- Multi-level Carpark #1 - 338.9 kWp
- Multi-level Carpark #2 - 338.9 kWp
- Sir Llew Edwards Building - 89.8 kWp.

There is also a concentrating PV array of 8kWp on site, the data from which was not included in this intermittency analysis.

The power output profiles of the solar arrays on the four different UQ buildings on a day with many moving clouds can be seen in Figure 115. The output power is normalised against the individual peak rating. At around 12:35pm, the PV power outputs of all the four arrays are seen to drop by about 80% in a minute. The corresponding total power output is also seen to drop by a similar amount at the same instant. The percentage drops in magnitude of the total output power during the day is seen to be similar to each of the individual building PV output power. In this case, spatial diversity due to the spacing between the four UQ buildings did not contribute towards smoothing out the intermittency. This demonstrates a strong correlation between the outputs of all the four different installations. However, with the resolution of 1-minute, the effects of fast moving clouds and the corresponding correlation between the outputs cannot be analysed.

Figure 116 shows a similar plot of the power output profiles of the solar arrays on the four UQ buildings and also the total power output but for a period of 1.5 hours in the morning of the same day. As expected, the output power for the two car park buildings can be seen to be very strongly correlated (they are located adjacent to each other), while the correlation is not as strong for the other two buildings. In this zoomed-in plot, a reduction in the magnitude of output power variations is observed in the total output power profile compared to the individual building output profiles. For example, at around 9:43am a 70-80% drop in power output is seen to occur in about 2-3 minutes in the individual building solar output profiles and the corresponding drop in total output power is about 50%. The variances of the output power for the individual installations and the total output power, for both the whole day and 1.5 hour periods, are shown in Table 22. It can be seen that the variance for the total output power is significantly smaller for the 1.5 hour period when compared to the individual buildings. This is, however, not the case for the whole day period where the total output power variance is similar to those of the individual PV installations.

Site | Variance - whole day (%) | Variance - 1.5 hours (%) |
---|---|---|

Car park 1 | 5.5 | 4.8 |

Car park 2 | 5.6 | 5.0 |

Sir Llew Edwards Building | 6.9 | 5.5 |

UQ Centre | 5.9 | 5.2 |

Total output power | 5.4 | 2.8 |

Figure 117 shows the distribution of 964 ramp-up event occurrences over 1-minute periods with rates greater than 5kW/s, corresponding to output power increases of at least 300kW (24.6% of plant rating) in one minute. Ten of these events had ramp-up rates of greater than 12kW/s, i.e. output power rise of more than 720kW (59% of plant rating) in 1-minute periods. Similarly, Figure 118 shows the distribution of output power ramp-down events over 1-minute periods and 11 events of output power drops of more than 720kW in 1-minute periods were recorded in the five month period in which data was collected. The distributions of ramp events over the various timescales were obtained using the same method as that for the DKASC.

Figure 119 and Figure 120 show the distribution of ramp-up and ramp-down events respectively in 2-minute periods with rates of 4kW/s or more. A total of six occurrences of ramp-up events with rates of more than 7kW/s were observed and this translates to an increase in the PV system’s power output of at least 850kW (69% of rating) in two minutes. The corresponding number of occurrences of ramp-down events with rates of more than 7kW/s recorded was 3.

The distribution of ramp-up event occurrences over 3-minute periods with rates greater than or equal to 2kW/s is shown in Figure 121. The total number of such events in the **fi**ve month period was found to be 1,247 including two occurrences of ramp-up events with rates greater than 5kW/s corresponding to output power increases of more than 900kW (74% of plant rating) within three minutes. Figure 122 shows the distribution of output power ramp-down events over 3-minute periods for rates greater than 2kW/s, the total number of which was found to be 1,309. The output power was seen to drop by more than 900kW in 3 minutes once in the **fi**ve-month period.

Figure 123 shows the distribution of ramp-up event occurrences over 4-minute periods with rates greater than 2kW/s, which corresponds to output power increase of at least 480kW in four minutes. 14 events of increases in power output of at least 840kW (69% of plant rating) in minutes were recorded. Figure 124 shows the distribution of output power ramp-down events over 4-minute periods for rates greater than 2kW/s and 18 events of output power drops of more than 840kW (69% of plant rating) in four minutes were recorded, one of which was for output power drop greater than 960kW (79% of rating).

Figure 125 shows occurrences of various ramp-up rates for variations recorded in 5-minute periods. A total of 453 occurrences of ramp-up rates of 1.5 kW/s or more were observed, corresponding to increases in power output of at least 450kW (37% of plant rating) in **fi**ve minutes. The corresponding total number of ramp down occurrences for 1.5kW/s or more for 5-minute is 511, out of which three were for ramp-down rates of 3kW/s or more (greater than 74% drop in plant power output in **fi**ve minutes), as can be seen in Figure 126.

The distribution of ramp-up event occurrences over 6-minute periods with rates greater than or equal to 1kW/s is shown in Figure 127. The total number of such events was found to be 466, including one occurrence of ramp-up event with rate greater than 2.5kW/s, corresponding to output power increase of more than 900kW (74% of plant rating) within six minutes. Figure 128 shows the distribution of output power ramp-down events over 6-minute periods for rates greater than 1kW/s, the total number of which was found to be 525. The number of times the output power dropped at least 900kW in six minutes was found to be three.

The number of occurrences of ramp-up events over various time periods for the 1.22MW PV system is summarised in Table 23. It can be seen that the majority of the fluctuations observed over all time periods are relatively small variations with ramp rates of less than 1kW/s. A significant number of rapid increases in output power were observed with ramp rates of more than 6kW/s over 1- and 2-minute periods which correspond to increases of more than 360kW (29.5% of plant rating) and 720kW (59% of plant rating) respectively in a short time frame.

Timescale | 0-1kW/s | 1-2kW/s | 2-3kW/s | 3-4kW/s | 4-5kW/s | 5-6kW/s | 6-7kW/s | 7-8kW/s | 8-9kW/s | 9kW/s+ |
---|---|---|---|---|---|---|---|---|---|---|

1-min | 50930 | 3194 | 1716 | 985 | 680 | 416 | 231 | 144 | 89 | 84 |

2-min | 40951 | 2614 | 1332 | 684 | 300 | 132 | 65 | 6 | 0 | 0 |

3-min | 33566 | 1784 | 801 | 331 | 113 | 2 | 0 | N/A | N/A | N/A |

4-min | 28620 | 1098 | 407 | 130 | 0 | 0 | N/A | N/A | N/A | N/A |

5-min | 25023 | 544 | 221 | 1 | 0 | N/A | N/A | N/A | N/A | N/A |

6-min | 22366 | 389 | 77 | 0 | N/A | N/A | N/A | N/A | N/A | N/A |

*Note: ‘N/A’ refers to ramp-rate events that are not feasible - such ramp rates over those timescales will exceed the capacity of the plant.*

Table 24 presents a summary of occurrences of ramp-down events over various time periods for the 1.22MW PV system. Similar to ramp-up events, majority of the total number of fluctuations over all time periods are small variations with ramp rates of less than 1kW/s. A considerable amount of rapid drops in output power with ramp rates greater than 6kW/s, corresponding to at least 360kW (29.5%) and 720kW (59%) drops in output power for 1- and 2-minute time periods respectively, were observed.

Timescale | 0-1kW/s | 1-2kW/s | 2-3kW/s | 3-4kW/s | 4-5kW/s | 5-6kW/s | 6-7kW/s | 7-8kW/s | 8-9kW/s | 9kW/s+ |
---|---|---|---|---|---|---|---|---|---|---|

1-min | 49681 | 3233 | 1588 | 1052 | 651 | 434 | 244 | 173 | 72 | 73 |

2-min | 39579 | 2526 | 1376 | 696 | 306 | 134 | 50 | 3 | 0 | 0 |

3-min | 32172 | 1720 | 858 | 345 | 105 | 1 | 0 | N/A | N/A | N/A |

4-min | 27197 | 1105 | 482 | 120 | 1 | 0 | N/A | N/A | N/A | N/A |

5-min | 23569 | 688 | 238 | 3 | 0 | N/A | N/A | N/A | N/A | N/A |

6-min | 20891 | 457 | 68 | 0 | N/A | N/A | N/A | N/A | N/A | N/A |

*Note: ‘N/A’ refers to ramp-rate events that are not feasible - such ramp rates over those timescales will exceed the capacity of the plant.*