If Taiwan's 4th Nuclear Power Plant Is Not Built (And No Other Plants Are Built), How Long Would It Take Before We Saw "Power Shortages"?

The decision last week by the government to hold a referendum on halting construction of the Longmen 4th nuclear power plant in Gongliao, "New Taipei City" (i.e. Taipei county) is bewildering. Construction of that plant has been ongoing for the best part of sixteen years already, with an horrendous amount of money spent; if, in the end, the plant is left unfinished then, stipulating to the ostensible purpose of the plant, all of that time, money and effort will have been wasted. That is probably the only point on which common agreement can be found.

Among the arguments for continuing construction of the power plant will be the claim that, if it is not built, there will be power shortages. Although this claim will be made by politicians (which is sufficient reason alone to doubt its' accuracy), the denunciation of this claim as "lies" will also be made by politicians and the kind of people who describe shale gas as "a witch's brew". So in the absence of credible sources, I thought I would try to evaluate the claim for myself.

What I want to know, is, how many years of increasing demand can be absorbed by the reserve margin capacity in the absence of new power stations?

Of course, this question is somewhat academic, because new power stations will be constructed, even if this particular power station is not (and also of course, no electrical grid in the world operates without reserve capacity). But the point of asking it is to get some rough idea of how things stand with regards to whether electricity production is adequate or not. I will not consider other extraneous variables such as transformer and transmission losses simply because I haven't got time to look into that.

To begin with, some basic numbers should be noted:

1) According to the Bureau of Energy's "Energy Statistics Handbook" for 2011 (which contains the government's latest official statistics), total electricity production in 2011 was 252.2 TW hours with total electricity consumption at 242.2 TW hours. In the previous year, 2010, total electricity production had reached 247.0 TW hours and total electricity consumption had reached 237.6 TW hours. So from 2010 to 2011, electricity production increased by 5.2 TW hours (2%) and electricity consumption increased by just over 4.6 TW hours (1.8%). However the electricity consumption figures for both 2011 and 2010 represent an average increase of approximately 5% over the twenty year period from 1990 and 1991. (These percentage increases in consumption are probably not an accurate reflection of demand, but I will assume they are roughly accurate for my present purpose).

2) According to the 2011 handbook, the peak load record for the electricity grid in 2011 was 33,787 MW with the average load given as 24,320 MW. The reserve margin is given as 20.60% which presumably is in reference to the peak load rather than the average load (although it may be taken in reference to some value between the two). If this margin is in reference to peak load, then that means the reserve electricity capacity in 2011 would have been 6,960 MW. In the previous year, 2010, the peak load record was 33,023 MW with the average load at 23,674 MW and the reserve margin was 23.40%, meaning again that reserve electricity capacity in 2010 would have been somewhat higher at 7,727 MW.

OK, so let's work it out. If we assume a reserve margin of around 20% or 7,000 to 8,000 MW, and if we assume an electricity demand increase every year of either 5 TW hours (2%) or 12.5 TW hours (5%), then what we need to know is how many TW hours per year could be delivered by the reserve capacity. To answer that question however, we also need to know what kind of power stations make up the reserve capacity and, more specifically, how efficient they are (i.e. what proportion of their power capacity (their MW rating) will make up the produced energy over the course of a year). Again, I am oversimplifying this due to lack of knowledge (I have some of the information but not the time to delve into sufficient detail), because, of course, no power station is going to be run for the entire 8760 hours per year - but that's the assumption we're going to make nonetheless. If we assume that the reserve capacity comprises mostly thermal (i.e. gas and coal) power stations, then we can probably assume an efficiency of between 30% to 50%. Let's assume three values for the sake of comparison: 30%, 40% and 50%. Let's also assume two values for the annual increase in electricity demand: 2% (the increase from 2010 to 2011), or about 5 TW hours and 5% (the average annual increase over the last twenty years), or about 12.5 TW hours. 

***

At 30% efficiency....

1,000 MW (30% = 300 MW) x 8760 = 2,62,8000 (or 2.62 TW hours). 5/2.62 = 1.908. So...

1,908 MW (30% = 572.4 MW) x 8760 = 5,014,224 (or 5.01 TW hours). 12.5/2.62 = 4.770. So...

4,770 MW (30 % = 1431 MW) x 8760 = 12,535,560 (or 12.5 TW hours).

At 40% efficiency...

1,000 MW (40% = 400 MW) x 8760 = 3,504,000 (or 3.5 TW hours).  5/3.504 = 1.426. So...

1,426 MW (40% = 570.4 MW) x 8760 = 4,996,704 (or 4.9 TW hours). 12.5/3.5 = 3.57 So...

3,571 MW (40% = 1,428 MW) x 8760 = 12,512,784 (or 12.5 TW hours).

At 50% efficiency...

1,000 MW (50% = 500 MW) x 8760 = 4,380,000 (or 4.3 TW hours). 5/4.38 = 1.141. So...

1,141 MW (50% = 570.5 MW) x 8760 = 4,997,580 (or 4.9 TW hours). 12.5/4.38 = 2.853. So...

2,853 MW (50% = 1,426 MW) x 8760 = 12,491,760 (or 12.4 TW hours).

***

From that little set of sums, we can see clear minimum and maximum values: in the "best" case scenario (i.e. with thermal stations operating at 50% efficiency and an increase in electricity demand of only 2%), the necessary capacity is only 1,141 MW. So if we divide our assumed reserve margin of 8,000 MW by that figure, we can see that our reserve margin will run out in just over 7 years under those "best" case conditions; by contrast, in the "worst" case scenario (i.e. with thermal stations operating at 30% efficiency and an increase in electricity demand every year of 5%), then the necessary capacity is 4,770 MW. So again, if we divide the 8,000 MW reserve by this figure, we can see that it will run out in 1.67 years.

So in the absence of new power stations, the existing reserve capacity is worth anywhere between one year and seven months, and just over seven years - depending on the type of power plant responsible for producing the reserve capacity and depending on the annual % increase in electricity demand.

My suspicion would be that many of Taiwan's thermal power stations are already quite old, and are therefore older designs that likely have a lower efficiency (e.g. closer to the 30% figure). However, since the economy is not growing at the rate it used to, it might be more reasonable to expect a growth in electricity demand closer to the 2% figure. According to the rough calculations above, the necessary capacity to meet the 2% increase at 30% efficiency is 1,908 MW. So divide our 8,000 reserve by this number and we get 4.19 years, or four years and nearly three months. That would be my guess as to how much time the current reserve margin would buy us in the absence of new power stations.

As for the fourth nuclear power station up in north Taiwan, with a capacity of 2,700 MW and a likely efficiency of at least 60%, it would produce about 14.1 TW hours per year or nearly three times the 2% growth figure of 5 TW hours - so at most, this power plant would give Taiwan an extra three years in which electricity production could exceed demand. That's not much considering how much money has been spent on it.