2016/2017 Botswana Energy Update – Part 3: Electricity by Mike Mooiman and Karen Giffard

Following up on Part 1, in which we looked at the big picture of energy supply and use in Botswana, and Part 2 where we examined coal resources, we turn our attention in this post to electricity production, the Botswana Power Corporation, electricity tariffs, and appraise developments over the last few years.

Botswana Power Corporation (BPC) is the government-owned electrical utility and is the only entity that is responsible for the supply, transmission, and distribution of electricity throughout Botswana. As a result, discussions about electricity in Botswana often center on BPC and its operations and performance. Much of the electricity situation in the country is summarized in the figure below, where we have plotted data for the past 21 years. The blue bars show how the total annual amount of supplied electricity (generated locally and imported) has doubled over this time period. The orange line shows the percent of electricity generated in-country by BPC, largely from their Morupule operations. (Data for 2017 are estimates based on data from the first two quarters only.)

  Data Source: 1997 to 2004; Adjusted BPC annual report results. 2004-2017; Statistics Botswana

There are two important details to note from this chart. The first is the remarkable turnaround since 2011 in the amount of electricity generated in-country. It has increased from less than 10% in 2011 to an estimated 77% in 2017. This is a result of the improved operability of the Morupule B operation and it is likely that, within a few years, that Botswana could be generating enough power internally to supply all of its domestic needs. Remarkably, in July 2017, there was a report that Botswana had, for the first time, exported (!) electricity into the South African Power Pool for a short period.

The other point of note is that the amount of electricity supplied from both local generation and imports appears to have peaked and levelled off since 2014 at just below 4000 GWh. The data from the first two quarters of 2017 suggest that there might even be a further drop in electricity supply in 2017. This decrease in supply could result from several reasons: less demand from the mining industry due to the closure of the BCL copper and nickel mine, reduced diamond production, energy-efficiency drives, and perhaps more temperate weather that required less heating and cooling.

For many years, electricity generation in Botswana was a disappointment. The Chinese-built Morupule B 600 MW coal-fired power plant experienced operating problems right from the start. There were many problems with the contractor and operability of the equipment that impacted reliability and limited availability. A damning report released by the World Bank highlighted many project errors, including selection of an unqualified contractor, safety issues on site that resulted in five deaths during construction, and poor project overview by BPC.

According to the 2015/2016 BPC annual report, Morupule B only had availability of 55.8% compared with the expected 89%. Although progress has been made in improving the operability of this plant, problems with the contractor, China National Electric Equipment Company, remain and cost overruns have ballooned. The investment in Morupule B and the electricity imports to cover for its lack of operation are now of the order of 12 billion pula ($1.2 billion) and the Botswana government is negotiating to sell Morupule B back to the contractor, following which BPC will buy back the electricity though a power purchase agreement.

At the same time that Morupule B was being commissioned and was experiencing operating problems, the ageing Morupule A plant was shut down in 2013. To compensate for the Morupule B problems, a decision was subsequently made to refurbish the Morupule A operation. This is being undertaken by Doosan, a South Korean company. The scheduled completion date is December 2017, but delays have been experienced.

Despite the many problems associated with the Morupule B plant, the recent good news is that its operability has improved and, according to the chart above and data from a recent Statistics Botswana report, local generation of electricity has improved. In the first two quarters of 2017, more than 75% of the power needs of the country were locally generated. The equivalent numbers a year earlier were ~50%.

Even as the refurbishment of Morupule A is taking place and the reliability of Morupule B is being improved, plans for further expansions of Botswana’s generating capacity are underway. A 300 MW expansion of Morupule B to build two more 150 MW generating units (Units 5 and 6) was agreed to by a joint venture formed by Marubeni from Japan and Posco from South Korea. The joint venture was to act as an independent power producer and would design, build, and operate the plant to sell power to BPC under a negotiated power purchase agreement. However, progress on this project has now been halted owing to issues of loan guarantees for $800 million and contention regarding the proposed power purchase rates.

Negotiations are also underway with Kepco and Daewoo from South Korea to build another 300 MW (Units 7 and 8) of generating capacity. If all the units at Morupule are built and the refurbishment of the 132 MW Moropule A plant is completed, this would give Botswana a coal-fired generating capacity of 1332 MW. This would be well in excess of its present peak needs and will allow for electricity exports and growth in local demand.

It should be noted that the potential expansion of Morupule B and the award of contracts to foreign companies has been criticized by several local energy companies that have coal resources and that had made progress in developing their projects in order to bid competitively for greenfield 300 MW generating sites, in line with a 2013 Botswana government tender.

Generally speaking, there are a lot of electricity-generation projects on the drawing board and under discussion, including the mostly unexplored potential of solar generation. The table below lists the existing generating facilities (highlighted in yellow) and those in planning that we are aware of (highlighted in green). It is unlikely that all these projects will be developed, but it is encouraging that there is so much focus and interest in developing electricity generation in Botswana.

As Botswana’s only electricity utility, BPC has struggled with insufficient income and financial deficits for many years, although the annual report for 2015/2016 did show the deficit shrinking. Operating deficits have been made up by cash infusions from the government. As a result, electricity in Botswana is heavily subsidized. Recent BPC data show that ratepayers are paying only about 60% of the cost of electricity: the average cost for procured electricity in Botswana (local generation and imports) is 139 thebe/kWh (US$ 0.14 kWh), whereas the average selling price to ratepayers is 82.1 thebe/kWh (US$ 0.082/kWh).

BPC is working hard to transform itself and to deal with its operating deficit. It has recently appointed an ex-patriot CEO from Germany and has prepared a turnaround plan called MASA 2020 that focuses on cost cutting, operational efficiency, selling of assets, and working with independent power producers. In line with many power companies across the world, BPC is looking to transform itself into a transmission and distribution utility, as opposed to a fully integrated generation, transmission, and distribution entity. The plan to sell problematic Morupule B back to the contractor that built it is part of this strategy. BPC is attempting to address the deficit/government subsidy issue by increasing electricity prices and improving operational efficiencies. In April 2017, electricity prices were increased between 8 and 13%, with the largest increases leveled at medium and large businesses.

Another challenge for BPC is that not all of the electricity generated or purchased from the South African Power Pool makes its way to customers. There are expected losses in the transmission and distribution grid due to the resistance of the wires, but there are also losses due to theft or meters not correctly recording actual electricity usage. It is notable that the losses from the BPC grid are large. The chart below shows that the recorded electricity losses, as determined by the difference between supply and customer sales, generally exceed 10%; based on the most recent data, these losses have risen to 15%. Such losses are considered large: global average transmission and distribution losses numbers, as determined by the World Bank, are 8.3%. Losses at Eskom in South Africa are typically 9% or below. Electricity losses in other neighboring countries are also high, e.g., Zimbabwe at 16.4%, Zambia at 15%. This represents an important opportunity for improvement.

In summary, the electricity situation, at least in terms of local electricity generation, has much improved during the past few years and as a result coal mining has also increased. It seems that load-shedding (rolling blackouts), as experienced in 2013/2016, could be a thing of the past due to local generation improvements and reduced electricity import costs with the economic downturn in Southern Africa. However, dealing with electricity blackouts, due to unexpected or scheduled maintenance inconveniences, is still very much part of life and doing business in Botswana. Nevertheless, we are encouraged at the progress that has been made.

In Part 4 of this series, we turn our attention to renewable energy projects and various government programs directed at the energy sector.

In the meantime, remember to turn off the lights when you leave the room. 

Mike Mooiman and Karen Giffard

mooimanm@franklinpierce.edu

karen@sosolar.co.bw.

History of Electrical Energy Generation in Botswana

Dumelang*. In this post, I take a look at historical trends to see how electricity generation in Botswana has changed over time. I noted in an earlier post about Sankey diagrams that Botswana generated only 7% of its electricity needs in 2012. This seemed an extraordinarily low number, so, in my research for this post, I also took a close look at the source data and uncovered some interesting discrepancies.

Utility-scale electricity generation in Botswana began in 1970 with the commissioning of a small oil-fired station in Gaborone. This provided for the needs of the capital and surrounding areas until its decommissioning in 1989. In 1974 a 65 MW coal-fired operation was built in Selebi-Pikwe to service the mining industry. 1985 saw the start-up of a large coal-fired plant at Morupule near Palapye and the shutdown of the Selebi-Pikwe plant a few years later in 1989.  The Morupule operation consisted of four 33 MW air-cooled units, providing a total generation capacity of 132 MW. This operation, known as the Morupule A plant, served Botswana’s needs well for a time. However, with increasing population, electrification of the country, and industrialization, demand rose and increasing quantities of electricity had to be imported from neighboring countries. The construction of the Morupule B coal-fired plant, consisting of four 150 MW air-cooled units, commenced in 2009. The deadline for completion was 2012, but many startup and operational problems have been encountered and today—three years past the scheduled completion date—the new plant has not lived up to its promise. There have been frequent breakdowns and, as a result, a great deal of Botswana’s electricity is still sourced from the Southern African Power Pool (SAPP).

Location of Morupule power plant near Palapye

In the figure below, I have plotted two sets of data for the past 18 years. The orange line shows the percent of electricity generated in-country by the Botswana Power Corporation (BPC), largely from their Morupule operations. The blue bars show the total annual amount of electricity (generated locally and imported) supplied in gigawatt hours (GWh). Data for 2015 are for the first three quarters only.

Data Source: 1997 to 2004; Adjusted BPC annual report results. 2004-2015; Statistics Botswana

An examination of the data shows that, from 1997 to 2008, the amount of electricity supplied steadily increased, with a compounded annual growth rate of 5.7%. During this same period, the proportion of local generation, which was as high as 60% in the late 1990s, decreased. This is to be expected from a fixed source of local generation in the face of increasing demand. However, the decline in local generation accelerated because of operational problems at the aging Morupule A power station. In 2011, an extreme situation was reached, when only 9% of the electricity supply was locally generated.

The difference in supply was made up of imports from the SAPP. However, since late 2008, supply shortages in the entire SAPP region have become acute and electricity imports to Botswana were reduced. As a result, rolling blackouts, also known locally as load shedding, were introduced across the country to curtail demand. Since then, the annual electricity supply has been up and down and there have been some years (specifically 2009, 2011, and 2013) in which energy supply was actually lower than the previous year. Load shedding, combined with the introduction of demand reduction initiatives, prepaid electricity, smart metering, and hot water load control, has slowed the rate of electricity supply growth since 2008 to a compounded rate of ~3% per year.

Since 2012, with the startup of the Morupule B operation, the local supply situation has improved and there have been year-on-year increases in electricity output from this operation. In 2014, Botswana again generated 60% of its electricity demand, significantly reducing the amount of electricity that it needed to import. However, the first quarter numbers for 2015 are disappointing because they indicate a similar level to that of 2014. It was anticipated that this value would be higher, but the performance of the Morupule B plant has been problematic, with only 300 MW of the specified 600 MW generation capacity currently available. We are hopeful for an improvement by the end of the year.

(In my post, The Big Picture, I noted that Botswana only generated 7% of its electricity needs in 2012. However, the historical data in the figure above indicate that this low actually occurred in 2011. My earlier statement was based on data from the International Energy Agency, IEA. A careful analysis of the IEA data led me to conclude that they were using information from the BPC annual reports. Because the BPC financial year ends in March, use of their data requires adjustments to take into account that their annual numbers actually incorporate the last three quarters of the previous year’s data. As a result, IEA reports of low levels of energy generation in 2012 actually correspond largely to the previous year, 2011.)

In my last post, I noted that energy students sometimes confuse capacity factor with conversion efficiency. The concepts are quite different: capacity factor is a measure of how much of the theoretical capacity of an energy-generating device was utilized over a time period (typically one year), whereas conversion efficiency is a measure of the effectiveness of the conversion of one form of energy to another, such as the conversion of energy in coal to electricity. To get a sense of conversion efficiencies, let’s take a closer look at the Morupule coal-fired plant.

According to the data provided by the IEA, the Morupule operation burnt 169 kilotons (kt) of coal and generated 250 GWh of electricity in 2012. To calculate conversion efficiency, we need to compare the input energy in coal and output energy as electricity in the same energy units by converting the energy in coal to GWh equivalents. If we assume that the Morupule plant burns coal with a heat content of 24 MJ/kg (a typical value for Botswana coal), we can calculate that 169 kt of coal contains 7.1 billion MJ of energy. Considering that there are 3.6 MJ in a kWh, we can calculate that the energy input was equivalent to 1126 GWh, which generated 250 GWh of electricity.

With this common set of units, we can now calculate that the conversion efficiency is Input/Output x100 = 250/1126 x 100 = 22%.

In fossil fuel plants, a different measure of conversion efficiency is often applied. This is heat rate, which is the amount of input energy (usually measured in kilojoules (kJ)) that is needed to produce one kilowatt hour (kWh) of electricity. One kWh of energy is equivalent to 3600 kJ, so if a fossil fuel plant is 100% efficient, it would have a heat rate of 3600 kJ/kWh, i.e., the plant would take 3600 KJ or 1 kWh of coal-based energy and convert it into 1 kWh of electricity. A plant operating with 50% efficiency would have double this heat rate or 7200 kJ/kWh (= 3200/0.5). At 22% conversion efficiency, the heat rate would be 16,363 kJ/kWh (= 3200/0.22).

Compared with other coal-fired plants around the world, which have conversion efficiencies of 30 to 40%, this value of 22% seemed extraordinarily low. To understand this discrepancy, I undertook a careful review of the IEA data and determined that they used a standard conversion efficiency of 22% every year to back calculate the amount of coal burnt in producing electricity in Botswana, i.e., their coal consumption numbers are not based on actual coal consumption data! The correct way to do this calculation would be to use the actual coal consumption numbers for the Morupule coal-fired operations. Unfortunately, I have not yet been able to source this data.

Further research led to some old 2006 data from SAPP which indicates that conversion efficiencies for the old Morupule A plant were actually more of the order of 30%. Additional evidence for higher conversion efficiencies was provided by data from the World Bank, which indicate that the heat rates at the Morupule plant were 11,621 kJ/kWh. This is equivalent to an efficiency of 31% (= 3600/11,621 x 100). As a result of this analysis, I am forced to conclude the IEA numbers for coal consumption in Botswana have a high bias.

A conversion efficiency of 30% (although significantly better than a value of 22%) still means that only 30% of the energy in coal ends up as useful electricity. The remaining 70% is lost due to process inefficiencies and heat losses.

To understand why this occurs, we need a better understanding of how the Moropule coal plant works. Coal is burned and the heat produced is used to boil water to generate steam. The steam is used to drive a turbine which then drives an electrical generator. In the process, we have the conversion of the chemical energy in coal, to thermal energy in the steam, to the kinetic energy of the turbine, to the electrical energy leaving the generator. In this process, there are heat losses in the hot off-gases that leave the combustion chamber and then exit those tall stacks at the Morupule operation. Another problem with a steam plant is that steam can only be used once to turn the turbine: it then needs to be condensed into water so that the water can be again be boiled to generate steam. The Morupule plants are air-cooled so the energy that was in the hot steam is lost to the atmosphere during the cooling process. The challenge with air cooling is that this is an energy-intensive operation itself, because large air fans need to be powered to drive air past heat exchangers. A significant portion of the energy generated by an air-cooled plant, typically 10%, is used to run the cooling units, which reduces the amount available for distribution. 

There are certainly coal-fired steam plants that have higher conversion efficiencies. Many modern coal-fired power plants have conversion efficiencies greater than 35% and water-cooled plants, such as those located near rivers or oceans, have even higher efficiencies. The world’s most efficient coal plants have conversion efficiencies as high as 47%. One hopes that, as the operational problems are solved at the Morupule plant, attention will be focused on improving its conversion efficiency to above 30%.

Until next time, remember to turn off those lights when you leave the room

Tsamayang Sentle**
Mike Mooiman
mooimanm@franklinpierce.edu

Click here to be notified of new Energy in Botswana blogposts.

(*Greetings in Setswana)
(**Go well or Goodbye in Setswana)

The Big Picture – Energy Supply and Consumption in Botswana

Dumelang.* The best way to approach most country- or statewide energy considerations is to understand the big energy flow picture. To do so, we use Sankey, or flow, diagrams which are often used in the energy industry, an example of which is shown below. In these diagrams the magnitude of the flow of energy is indicated by the width of the arrow. These diagrams were first used in the energy field by an Irish engineer and captain in the British army, Matthew Henry Phineas Riall Sankey, in 1898, to illustrate energy flows in steam engines. The modern version below neatly shows how input energy into a steam engine, in the red flows, is lost to smoke, friction, and heat loss via the steam condensor. Useful energy as forward motion of the steam engine, a small recirculating flow of energy that comes from the recirculated condensed water and a small amount going to the alternators are shown as the blue flows.

The great thing about Sankey diagrams is that they are not restricted to energy flows. They can be applied to quantities of many types. For example, material and cost flows are often depicted. One of the most famous of these flow diagrams is that prepared by the French engineer, George Charles Minard, in 1869, and shown below. This illustrates the fate of Napoleon's army in 1812 – 1813 as they progressed through their disastrous Russian invasion. The width of the lines shows the fate of the invading army. Napoleon crossed into Russian with 422,000 men and, through attrition, minor skirmishes, and some great battles, he entered a largely abandoned Moscow with about 100,000 men under his command. He then turned back to return to France: on the way back, starvation, battles, and incessant harassment by guerilla forces decimated his army to 10,000 survivors. The harsh winter also took its toll on his men - the line graph below the flow diagram shows the decreasing temperatures encountered on the army's return from Moscow. The diminishing width of the flow is a skillful, albeit harrowing, representation of what was happening to the army in the field, the prisoners that were taken, and the lives that were lost. 

  

But I digress… Let's return to energy flows. I have prepared the following Sankey diagram for energy flows in Botswana, based on 2012 data from the International Energy Agency (IEA). The IEA have prepared Sankey diagrams for most countries in the world (including Botswana), and, although this is a wonderful source of data, their flow diagrams are two-part figures showing a complicated spaghetti of flows. I have taken their data and distilled it down to the more easily understood diagram shown below.

Although I am highly impressed with the data that the IEA has collected for most of the world’s countries, we need to be cautious about some of the numbers. In my previous post, I pointed out some discrepancies regarding biofuel usage. I am, as yet, not in a position to verify all the Botswana-related data that the IEA has collected, but I hope to do so over the next year.

There is a lot of data incorporated into this diagram, so let’s work from left to right to get a sense of some of the larger flows. On the left, we have the energy sources for Botswana. There are only four types: local biofuel (wood), imported electricity, locally mined coal, and imported oil products. For 2012, the total energy supply available in Botswana was 109,165 terajoules (TJ), as shown in the pie chart below.

Almost one half of Botswana’s energy supply is in the form of oil products, such as petrol and diesel. The green flows in the Sankey diagram show that these are mainly used in transportation and industry. The small flow to non-energy applications show use largely as lubricants, oils and greases, for industry and transportation.  Biofuel, largely wood, is still very important in Botswana: the brown biofuel flows show that all wood is used in household applications, most likely for cooking and heating in the rural and poorer areas.

As shown by the black Sankey flows, Botswana produces a good amount of coal. Much of this was stockpiled in 2012, but a fair quantity was used by industry and about 11% was used for electricity generation at Botswana’s only coal-fired station, Morupule, which is near the town of Palapye. However, Botswana does not generate sufficient electricity for its own needs. If we refer to the center of the Sankey chart and we follow the red flows, we can see that, in 2012, Botswana only produced a small portion, 7%, of its domestic energy needs. The bulk of its electricity was imported from the Southern African Power Pool grid and came from Mozambique and South Africa. The bulk of this electricity is used by industry, followed by households, and the commercial and public sector.

In my last post I noted that energy consumption for 2012 was 86,206 TJ. The large difference between the 109,165 TJ of supply and the 86,206 TJ of consumptions accounted for by several reasons:

• More coal was mined than was actually used. The difference ends up in stockpiles to be used in the future.
• Electricity generation in coal-fired power plants is very inefficient. Only about 23% of the energy in coal ends up as electricity.  
• Transmission and distribution of electricity to end users results in line losses of the order of 8 to 10%.
• There are small differences due to statistical variations, own use, and storage of petrol and diesel at various locations.

The overall usages of energy by the main categories, as shown on the right of the Sankey chart and the pie chart below are split as follows: transportation uses just over one-third, followed by households, and then industrial use.

So here you have the overall energy picture for the country of Botswana, at least for 2012.  It will interesting to see over the next year or so, as I collect current data, how much this picture will change.

Until next time, feel free to comment on my post and remember to turn off the lights when you leave the room.

Tsamayang Sentle**

Mike Mooiman

mooimanm@franklinpierce.edu

Click here to be notified of new Energy in Botswana blogposts. 

(*Greetings in Setswana)

(**Go well or Goodbye in Setswana)