New data shows benefits of renewable energy

July 25th, 2017, Published in Articles: Energize


The Renewable Energy Data and Information Service (REDIS) has been updated with new data up until June 2017. REDIS provides project data for 121 renewable energy generators, and now features more than one million hourly electricity production data records from 83 plants in operation, from April 2012 to June 2017.

In this REDIS update, the hourly performance of wind power and photovoltaics (PV) for a statistically average day of production in South Africa across all years and provinces was analysed and compared. The results show that average wind power output is relatively stable throughout the day, observing a peak between 17h00 to 19h00, which correlates with the demand peak.

Even though the average load factor for single-axis tracking PV is 19% higher than the load factor for fixed PV without tracking, the analysis shows that fixed PV without tracking performs 6% better than PV with single-axis tracking in the period between 11h00 and 14h00.

In the analysis below, a technical explanation for this phenomenon is provided, and seasonal differences are studied. The findings have significant value in power system planning for the integration of renewables, and may also be used as a basis for improved feasibility studies of investments in renewables.

For this analysis, a statistically average day of hourly production for renewables in South Africa spanning April 2012 to July 2017 and all nine provinces was compiled. Furthermore, seasonal variations in hourly production were analysed. While the interactive dashboard available on REDIS holds data for all renewables (wind, PV, CSP, hydro, biomass, landfill gas), the analysis here focuses on wind power and PV with and without tracking.


Fig. 1: Seasonal and time-of-day output from wind and PV installations.

Average day of hourly production

Fig. 1 illustrates that wind power output is relatively stable throughout an average day, observing a peak between 17h00 and 19h00. The wind production peak correlates fairly well with the demand peak, which appears around 19h00 in winter and 20h00 in summer according to data from Eskom [2].

In REDIS Update No. 3, it was shown that the average load factor for PV with single-axis tracking is 19% higher than the load factor for PV without tracking. However, when analysing the hourly variations, it was found that PV without tracking performed 6% better than PV with single-axis tracking in the 3-hour period between 11h00 and 14h00.

PV with single-axis tracking produces more electricity than PV without tracking from 05h00 and 11h00 and again from 14h00 and 19h00. While this results in a higher overall load factor for PV with single-axis tracking, the load factor in the middle of the day is actually higher for PV without tracking.

In order to understand this phenomenon, it is necessary to understand that:

  • The performance of PV depends on the angle of incidence: the lower the angle, the higher the performance.
  • The majority of PV installations with single-axis tracking in South Africa are so-called horizontal single-axis systems, which tracks the sun from east to west.

PV tracking systems

A horizontal single-axis system tracking from east to west gradually tilts the panel to minimise the angle of incidence. However, at midday, the horizontal single-axis PV panel is horizontal to the ground. This results in a higher angle of incidence, and thus lower production, than that of a PV panel without tracking, which is fixed tilted north towards the sun.

A tilted single-axis tracking system would be able to match the fixed tilted PV without tracking at midday. However, tilted single-axis tracking systems are more expensive than horizontal single-axis tracking systems, mainly because they cast longer shadows and therefore require more land-use.

Seasonal variations
Significant seasonal variations have been observed. For example, while wind power output is virtually constant on an average winter day, the afternoon peak is double than that of the morning output on an average summer day.

Fig. 1 shows that the statistical representation of hourly production has been normalised to the installed capacity in each hour, where the value 1 represents maximum output at full capacity. The term “load factor” describes the actual production as a fraction of the theoretical production at full capacity. For example, if a 1 MW-peak PV installation in a given hour produces 0,75 MWh, the load factor in that hour is 0,75. Subsequently, load factors can be averaged for days, months, and years.


Fig. 2: Hourly production and annual load factors.


Please refer to REDIS Update No. 3 for an analysis of annual load factors.


[1] B Sørensen: “Renewable Energy: Its physics, engineering, use, environmental impacts, economy and planning aspects”, 2004.

[2] Eskom: “State of the power system quarterly update”, September 2014.

[3] GM Masters: “Renewable and Efficient Electric Power Systems”, 2004.


REDIS would like to extend its thanks and appreciation to Jarrad Wright and his colleagues from the CSIR’s  energy centre for their assistance in this report.

Contact Renewable Energy Data and Information Service, Department of Energy,



Subscribe to our leading email newsletters


CLICK for other EE Publishers information products