4.3 Load-following and regulation requirements

Analysis in the CAISO report [10] examines the load-following (5-minute) and regulation (1-minute) requirements to meet the new variability in net load introduced by wind and solar. A metric was developed using synchronised 1-minute data of wind, solar and load. A 15-minute rolling average of the 1-minute data was calculated and the 5-minute delta was calculated on this rolling average. The 1-minute delta is the difference between the 15-minute rolling average and the 1-minute measure. Figure 46 illustrates an example of this metric.

Figure 46 Example for load-following (5-min) and regulation (1-min) metric definition [10]

Using this metric, 5-minute and 1-minute deltas on the 15-minute rolling average for the load, net load, wind and solar data were produced. These are illustrated in Figure 47 and Figure 48 respectively. Both the 5-minute and 1-minute deltas are for the same 3-hour period one morning in July, 2003. The left y-axis scale applies to load and net load while right y-axis scale applies to the wind and solar. From Figure 47, the load following requirement ranges from 120 MW to 400 MW per 5 minutes. Over the 3 hour period the average load is 40 GW, 600 MW for wind (1.5% penetration) and 700 MW (1.75% penetration) for solar. Examining the 5-minute analysis, the standard deviation for wind is 18 MW (3% of the average) and 13 MW for solar (1.8% of the average). As expected, at such low penetrations there is little impact on the net load. What is of interest is the difference in variability between wind and solar for 5-minute and 1-minute intervals. From Figure 48, the 1-minute wind delta stays within 20MW/minute while the solar delta ranges between ±60MW/minute. Looking at the 1-minute analysis, the standard deviation for wind is 9 MW (1.5% of the average) and 22 MW for solar (3.1% of the average). So the 1-minute standard deviation increases relative to the 5-min by 50% for solar and decreases by 50% for wind. Based on the comparisons made above, something for utilities to consider is that a greater penetration of solar is likely to imply greater regulation requirements while a greater penetration of wind will require greater load-following requirements. However, it should be noted that solar and wind deltas at different timescales are situation dependent, and are dependent upon such factors as number and diversity of solar versus wind plants, and weather characteristics of where they are located.

Figure 47 5-min delta over three hours one July morning, 2003 [10]

Figure 48 1-min delta over three hours one July morning, 2003 [10]