Power system dog maths

29th January 2021 By: Terry Mackenzie-hoy

Dog Mathematics’ is mathematics you could teach a dog to do. The dog may not get it right, depending on the complexity, but it’s all real cute all the same.

In 1998, I wrote to Eskom and said that, as far as I could see it, the utility would run out of power in about nine years – in 2007. Eskom wrote back (the utility actually replied to letters in those days) and said, in effect: “Hey, not to worry, we got it all nailed.” In point of fact, they did not have it in hand, since the President at the time, Thabo Mbeki, would not authorise any further construction of any power resource. I did hear some alarming reasons for this, but I think government ignorance was one of them; it seemed to me that government thought that the power system was some sort of huge dam in which energy is stored so that you could just keep it topped up, somehow. In point of fact, as I have explained in the past, there is no storage of any significance in the Eskom power system. If the demand is 35 000 MW and the generated power is 34 900 MW, then the whole system crashes and the lights go out.

Now, with a different President, the whole system is going to suffer equal stupidity. It was published in the Integrated Resource Plan (IRP), an electricity capacity plan that aims to provide an indication of the country’s electricity demand, how this demand will be supplied and what it will cost.

The plan, recently published, is a blueprint of how to mess around with things until this country runs out of power in a planned fashion. So, how is this? Firstly, in the plan is a table which lists the proposed capacity of all types of generation: coal, nuclear, hydro, storage, photovoltaic (PV), wind, concentrated solar power (CSP), gas and diesel. We read that the plan is to reduce coal generation from 37 149 MW to 33 364 MW (reduction of 3 785 MW), add 5 000 MW of storage capacity, increase PV to 8 288 MW, increase wind to 17 742 MW, increase CSP to 600 MW and increase gas and diesel to 6 380 MW. Let’s look at all this. Firstly, ‘5 000 MW storage’. What is this? Is it some humongous battery? If so, based on other batteries worldwide, it is ten times larger than the Australian Tesla battery and probably only available for a few hours. No good as baseload. Then wind and solar PV give a total of 26 030 MW or zero megawatts if it’s a still, windless night. What all this means is that the Eskom power system will be able to support a load of 38 000 MW by 2030, after which it will not be able to do so.

One can see that having a new capacity of 14 400 MW of wind generation is going to use 7 000 km2 of land. Since there is no wind on the Highveld, one imagines that the wind turbines will fill a strip of coast 20 km wide and 350 km long from Jeffery’s Bay to East London, which is a bit depressing for tourists.

In essence, the IRP clearly has been cobbled together by a group of self-interested parties who have spoken into the ear of government and convinced them to call for the building of wind turbines on a large scale, when these hardly help anything. Money is the motivation. The fact is, any megawatt of wind power has to be backed up by a megawatt of firm power, and that is an inescapable fact. The IRP will not serve South Africa very well. Assuming government gets the economy together, we can look to a power requirement of more than 38 000 MW by 2030, and Eskom just won’t have it. What is needed is big investment in solar PV and hydro pump storage. A programme of power generation from wave power would also be a good idea.