The below article is a follow-up of Dr. Asfaw Beyene’s commentary published on OPride.com on June 14, 2013 (click here to read the previous article). Both articles examine the engineering merits of the “Grand Ethiopian Renaissance Dam.”
By Asfaw Beyene*, Ph.D.
This is continuation of my previous commentary entitled “Reflections on the Grand Ethiopian Renaissance Dam,” dated 14 June 2013. Here, I will try to address the issues of size and capacity of the dam which were missing from my previous article. Thank you all who brought this issue to my attention, Dr. SK in particular.
Based on Internet data, the peak flow rate of the Blue Nile is 5663 meter cubed per second (mcs). Data from Bashar et al., Water Balance Assessment of the Roseires Reservoir (Khartoum, Sudan; Ministry of Irrigation and Water Resources, Sudan) give a flow rate of the same range: 6944 mcs in 1985, 5208 mcs in 1995, 5787 in 2005. The numbers for Roseires Dam in South Sudan are more reliable since the dam is close to the Ethio-Sudanese border.
Based on a British article, the Grand Ethiopian Renaissance Dam (GERD) is reported to have 145 meters height. The flow rate and the GERD dam height fix the maximum possible theoretical power output from the dam at about 7250 MW, assuming some 90% efficiency for the Francis turbine. The annual peak flow rate varies, but the above average value 5663 mcs can be assumed for further analysis. If the dam were designed to use this peak flow, most of the turbines of the 7250 MW have to idle when the flow rate drops below 5663 mcs, and there would be no storage required, if not for the required head (elevation). In fact, even after years of initial storage time, the reservoir can’t possibly produce the peak flow rate for extended period of time.
The average flow rate of the Blue Nile is reported to be about 2350 mcs. This average for the same height of the dam would provide about 3000 MW power output. Reverse calculation yields 4700 mcs flow rate for the proposed design of 6000 MW, which is way above the annual average of 2350. In fact, the 16 turbines with 350 MW each can only produce a total of 5600 MW, not 6000 MW. This corrects the design flow rate to 4400 mcs, not 4700 mcs.
There are many possible input and design scenarios which we simply don’t know. One thing is true however. Given the height of the dam and the flow rate, there is no way the dam can operate at the level of 5600 MW output throughout the year even if the dam stores the difference between peak-flow and design-flow rates. If we assume the design engineers factored in two turbines – 700 MW downtime (allowing equivalent water for bypass or storage) for a more realistic 4900 MW output, the flow rate has to be about 3800 mcs, which still remains way above the average flow rate (see Figure). Simply put, the peak covers too few months to fill the reservoir in the summer when the discharge drops significantly.
According to my calculations, allowing the average flow rate of 2830 mcs to pass to downstream countries during peak months will fill the reservoir in about 5 years (a number talked about a lot), but the refill per annum will be too small to sustain annual operation even close to 4800 MW. The only scenario under which the power supply will be consistent, and the refill can be sustained for summer at about the same power output level is if the hydroelectric dam is designed for a mean flow, which is about 1456 mcs (Internet sources). This will provide just less than 2100 MW (say, 7 turbines with 350 MW each). Assuming 700 MW (two turbines for maintenance downtime), the appropriate design target in my opinion would be 2800 MW, still larger than the 2100 MW at Aswan High Dam – to please those who like to compete. This assures year-round supply of electricity at almost constant level, also requiring a shorter period for initial refill. Such consistency offers high rate on investment. The total price at $800/kW rate will be about $2.3 billion dollars – much less than the $4.7 billion for the 6000 MW. More importantly, Egypt may be happy, making it easy to borrow money for the project.
Of course, the input values are not known and flow rates also vary from year to year, rendering the calculations here a bit tentative. Regardless, there is little doubt that the system is designed for near-peak flow rate. The question then is, should one design a system for near-peak flow, i.e., near the theoretical maximum power generation, or for the mean flow rate? This is a common topic in system design, and the question arises whenever input resources or supply demands vary. Hydroelectric dams are best designed to provide maximum kilowatt hour (not kW), which means we target mean flow values. This decision is fairly trivial for the Blue Nile that has very low flow rate during the dry season. Targeting near-peak or peak flow rate makes no economic sense. The remaining question is then, why is it sized for 6000 MW? I really don’t know.
* The author, Asfaw Beyene, is a Professor of Mechanical Engineering and Director of the Center for Renewable Energy and Energy Efficiency at San Diego State University. He can be reached at email@example.com.