A Quick Look at the National Grid’s Future Energy Scenarios

National Grid has just published its 2015 Future Energy Scenarios report, which gives  four different NatGrid visions of what the UK energy mix might look like in 2035/36. The scenarios are documented in a long report which is heavy on projections and assumptions and light on engineering – in fact I think it would be true to say that there isn’t any. The level of detail (which goes down to “number of appliances by type”, “take-up of cavity wall insulation” and “shorter distance EV charging profile”) also defies analysis in any reasonable time frame. So here I document the results of a quick-and-dirty review conducted by subjecting Gone Green – the most aggressive of NatGrid’s scenarios – to the “February 2013 treatment”. Does NatGrid’s February 2015/36 generation mix fill February 2035/36 electricity demand if weather conditions in that month are the same as they were in February 2013? As might be expected it doesn’t:

Figure 1: National Grid “Gone Green” scenario for February 2035/36, generation deficits factored from February 2013 hourly National Grid data.

A very brief account of how I came up with this result. Figure 70 of the NatGrid spreadsheet shows NatGrid’s 2035/36 Gone Green generation mix, which according to NatGrid cuts CO2 emissions by 87% relative to 2013/2014 levels. (It is not, however, a high-renewables-penetration scenario. Renewables in fact make up only 50% of total generation.) I factored Gridwatch February 2013 generation and demand to match this generation mix. The Table below gives specifics:

Note: I prorated French solar generation for the month to match NatGrid’s 1.3TWh of projected solar generation in February 2035/36. Gridwatch supplies no solar generation data for the UK.

After applying these factors NatGrid’s generation by source for February 2035/36 looks like this:

Figure 2: National Grid “Gone Green” scenario for February 2035/36, generation by source factored from February 2013 hourly National Grid data.

Comparing February 2035/36 total generation from all sources with February 2035/36 demand gives this:

Figure 3: National Grid “Gone Green” scenario for February 2035/36, total generation versus demand factored from February 2013 hourly National Grid data.

And subtracting demand from total generation gives these power surpluses and deficits (the deficits are shown in more detail in Figure 1). They are caused dominantly by the large fluctuations in wind generation. (I’ve made no attempt to estimate how much storage would be needed to balance them out, but once more it would be up in the TWh range):

Figure 4: National Grid “Gone Green” scenario for February 2035/36, generation surpluses and deficits factored from February 2013 hourly National Grid data.

So we can add the NatGrid Gone Green scenario to the DECC and Centre for Alternative Technology scenarios which as discussed in recent posts also won’t work.

NatGrid also has three other scenarios; Low Carbon Life, Slow Progression and No Progression. Would they fill 2035/36 demand? Rather than repeat the Gone Green exercise I short-circuited the process by taking the generation mix graphics in the NatGrid report and replotting them in modified form (Figure 5). I whited-out wind and solar, the two non-dispatchable generation sources, imports, which probably won’t be available when needed in 2035/36, and CCS, which at the present rate of progress may never be commercialized. This leaves only generation from dispatchable sources, which are the only ones that can be relied on to fill demand:

Figure 5:  National Grid’s four future energy scenarios with non-dispatchable and other speculative generation sources whited out.

Low Carbon Life and Slow Progression meet emissions targets but have substantially the same generation mixes as Gone Green, indicating that they too will almost certainly not meet demand. No Progression (top right) has a higher proportion of dispatchable generation and therefore a better chance of working, but it’s essentially a business-as-usual scenario that doesn’t meet emissions targets.

The question now has to be asked; why can’t anyone come up with a scenario for a low-carbon, low-emission, renewable energy future that works? Because there isn’t one. At this time we simply don’t have the ability to replace conventional dispatchable generation with large amounts of non-dispatchable renewable generation and still deliver power when needed, which is what the published scenarios attempt to do. The best the UK can hope to do at this point is develop a plan that works and which gets it at least some way down the road towards a green energy future. What might such a plan look like? I’m working on one.

This entry was posted in Energy and tagged , . Bookmark the permalink.

16 Responses to A Quick Look at the National Grid’s Future Energy Scenarios

  1. Luis says:

    After having read this and other posts such as http://euanmearns.com/the-decc-pathways-calculator-a-false-prophet/ in your blog regarding the problem of non-dispatchability which renewable energies have and how demand could be met by intermittent energy sources, I became interested in finding a way to obtain the cheapest solution that could supply the electricity demanded by the society. In order to do so I developed an excel macro and applied it to the Spain’s 2014 data although with the proper data (load and wind and solar production per hour throughout a year) it could be applied to any country.

    Here is a detailed explanation on how the excel sheet works


    and here the the excel sheet


    I did the following assumptions:

    Energy storage plants have the following characteristics:

    • Power 600 MW
    • Energy stored 30 GWh
    • Lifespan 50 years
    • Cost 1,200 millions of euros
    • Interest rate 5%
    • Annual cost 66 millions of euros
    • Losses of energy 30%

    All energy must be paid whether it is used or not

     Hydro 0.04 €/Kwh
     Wind 0.07 €/Kwh
     Solar 0.09 €/Kwh

    Hydro annual production 33,970 Gwh (completely dispatchable)
    Hydro power 17,000 Mw

    Optimal solution
    – Cost 0.0844 €/Kwh
    – Solar factor 6 (currently 5% energy)
    – Wind factor 3 (currently 21% energy)
    – 25 storage plants
    Non dispatchability accounts for a cost of 0.0129 €/Kwh, approximately an increase of 18%
    Approximately 5% of the energy produced is wasted.

    I would really appreciate any comment in order to improve the program, and finally I would like to congratulate you for this blog, really interesting stuff.

    • Euan Mearns says:

      Luis, this looks very interesting. I am to be away this weekend and today I’m trying to complete the post for Monday and so don’t have time to engage in this right now. Hopefully Roger will be able to. I’d like to pursue this. I haven’t downloaded your spread sheets. You say 17 GW of hydro. How much does Spain have today? And what does the generation mix look like? How many GW wind and solar? Is this a 100% renewables option?

      Sunny and warmer Spain has less of a problem with annual storage requirements. A recurring theme of mine is that one size cannot fit all. Norway and Iceland for example have run on 100% renewable electricity for decades. The Dutch are just not trying 😉 The UK has 1.6 GW of hydro.

      Your storage seems under-dimensioned to me, but I’ve not looked at your sums. One of the problems we have with seasonal storage in the UK is how to pay for it. I guess what you are saying here is that in Spain an 18% storage levy would be required?

      • Luis says:

        Hi Euan, thanks for your quick repply
        – Spain has currently 17,791 + 2,101 MW (small hydro) http://www.ree.es/sites/default/files/downloadable/the_spanish_electricity_system_2014_0.pdf (page 10)
        – The 2014 generation mix cand be seen in the same document and also in this link for any given day http://www.ree.es/en/balance-diario/peninsula/2014/12/31
        – in 2014 22,845 MW of wind were installed which produced 50,630 GWh (20%)
        PV: 4,428 MW and 7,794 GWh (3.1%)
        CSP: 2,300 MW and 4,959 GWh (2%)

        I did the program in order to get a 100% renewable option although I think it could be modified in order to make possible some nuclear ( this would be easy, just subtracting nuclear from load) or another kind of source.

        You are right regarding the issue of small hydro in the uk. Spain fortunately has a lot of power installed. In fact I think that without it the solution provided by the excel sheet would have been more costly, but I have not tried so I am not sure about it.

        The 18% figure is a bit tricky,
        First I work out how much of solar and wind would be necessary if they were dispatchable, (goal: getting K)

        K·(SF·(SP 2014) + WF·(WP 2014)) = Load 2014 – Hydro
        K = (Load 2014 – Hydro)/(SF·(SP 2014) + WF·(WP 2014))

        Solar Ideal = K · SF·(SP 2014)
        Wind Ideal = K · WF·(WP 2014)

        Then I calculate the total cost for this “ideal Solution”

        Cost Ideal solution = (Solar Ideal ·Cost Solar + Wind Ideal · Cost Wind + Hydro · Cost Hydro)/ Demand

        In the best solution I got a cost of 0.0715 €/Kwh
        However, given that solar and wind are intermitent, storage and surplus of wind and solar are needed so the real cost is 0.0844 €/Kwh
        That makes a different of 0.0129 €/Kwh (0.0844/0.0715 -> 18% increase

        The storage needed is 720 Gwh (30 Gwh · 24) .It’s just 110% of the average daily consumption but it seems enough given the great deal of hydro that Spain has.

        Regarding the issue about storage payment, I assumed that the utilities or whoever would build the infraestructures and then they would received an annual fixed payment regardless of the use, so they could make a profit.

        It’s true that Spain is warmer, specially the south and the mediterranean cost, but the UK could partly solve this issue by building district heating networks. They usually count on heat storage plants so they can use energy surpluses and the store it for a few days. Moreover district heating networks usually have more than one energy sources so in case that there is not enough electricity biomass could be employed.

        I’d like to apologise for the mistakes I probably have made writing this.

    • Luis: Thank you. It’s refreshing to know that other people are working with actual generation data rather than pie-in-the-sky future energy scenarios.

      Like Euan it will take me a while to go through your results, but you obviously have access to Spanish grid data that I don’t. If you have the data available and can spare the time I would be very interested to see hourly generation from all sources (wind, solar, hydro, thermal, exports/imports etc.) for Spain 2014. I need a break from UK data anyway.

      You’ve come up with a storage requirement of 750,000MWh = 0.75TWh, correct? This is lower than the estimates Euan and I have made for the UK, but this could be because you have allowed for storage in existing hydro plants (at least I think you have) while we haven’t. And Spain has a lot more hydro than UK.

      • Luis says:

        Hi Roger, I used the data provided by the Spanish Grid REE
        they are available here
        I have manually downloaded all 2014 data, although I have not translated it into English I think the different categories are easily understandable.

        Probably hydro was fundamental to get such a low figure. I am going to run another simulation without hydro, As soon as I get a soluton I’ll tell you but the process is very time consuming so I don’t know how long it will take.

        • Luis: Apologies for the delay in replying but I’ve been running errands most of the afternoon.

          I”ve managed to download your 2014 data and convert it into hourly averages. Thank you!. Y no hay problema con el idioma – hablo Español. Pero I do have one question. How do I allocate the special regime generation between wind, solar, biomass etc? I read that about 50% of it is wind, but I don’t know what the rest is.

          What I might do at some point is repeat your analysis from scratch to see if I get the same result. Stand by.

          • The plot thickens. I find that the Resto reg. esp. column is actually load-following generation.

          • Luis says:

            Hi Roger, no problem. I had forgotten but the entire data was downloaded from REE site except from an hour (2 to 3 am 30/3/2014) in which I employed a cubic interpolation to get the production of that time.
            With regard to the Regimen especial, the “resto” includes mainly cogeneration and renewable thermal (probably biogas from landfills and other sorts of biomass)

            The whole Regimen Especial includes the following:

            Rest of the Hydro (Small Hydro)
            Solar PV
            Solar thermal
            Thermal renewable
            Cogeneration and the rest of the technologies

      • Luis says:

        I didn’t use current hydro as means of taking advantage of previous surpluses of energy, I mean, if there is an eventual surplus of energy, that surplus goes to storage (if there storage plants are not full). I used hydro just as I could have used gas, dispatchable source of energy, with two limits, maximum power and maximum annual energy.

        In order to decide which source is better to use I envisaged the Storage coefficient. This ratio was introduced to simulate the behaviour of the grid operators.
        Let’s explain it with an example in which SC is 60 (%). If in a given time demand were higher than solar and wind production, two different options would be possible. If the energy stored were higher than SC, the energy stored would be used first, were it not be enough, hydro would also be employed. Conversely, if energy stored were lower than SC, hydro would be used first and energy stored would only be used if the previous one were not sufficient.
        This coefficient was introduced as an automatic decision device to answer the question of what source of dispatchable (but finite) energy is better to use, normal hydro or pumped hydro. If pumped hydro plants are close to maximum capacity it may be better to utilise their energy in order to leave space for prospective surpluses. On the contrary, if pumped storage is hardly empty it may be better to use hydro and leave pumped storage for future moments when demand exceeds what can be provided by hydro.

  2. jacobress says:

    Spain has a steady supply of better that 20% of generation from nuclear plants. That surely helps – but their plants must be aging, like all plants, and I don’t know how much longer they will stay in use. I’m not aware of any new plants being built or planned.

    • Luis says:

      Yes, that’s true but I didn’t take them into account, although it could be possible

  3. Luis says:

    There was a small bug in the program, If you have already downloaded it please do it again

  4. peter2108 says:

    I wrote a brief piece (http://monicol.co.uk/ukClimateAndEnergyPolicyIn4Charts.html) on future UK energy policy based on DECC projections. The gulf between DECC and National Grid Gone Green is huge.

    For 2035 DECC’s primary energy demand projection (low fossil fuel prices) is 38 (nuclear), 21 (renewable), 74 (gas), 71 (oil) and 6 (coal). The units are million tonnes of oil equivalent (Mtoe). Excluding oil this comes to 139 Mtoe. 1 Mtoe = 11.6 TWh so DECC projects primary energy demand excluding oil of 1617 TWh.

    National Grid’s Gone Green scenario for power generation (Fig 57 in spreadsheet) has total of 447TWh. In their annual supply of gas (Fig 74) they estmiate 46 bcm. 1bcm = 0.9 Mtoe = 0.9*11.6 TWh. So 46bcm = 482 TWh. So Nat grid total energy projection for power + natural gas for home heating etc is 480+447 = 927 TWh.

    It would be neat if DECC was projecting twice the energy demand that Gone Green was but its only 75% more. Sure, DECC is projecting on present policies (whatever that means exactly) while Gone Green is … well what is it? Accordfing to Roger it does not keep the lights on, but it a;lso seems so far from reality to justify the overused adjective “delusional”

    • Peter: Thank you. Your point about the difference between electricity demand projections is a good one. No one can predict what future electricity demand is going to be, and the differences between DECC and NatGrid illustrate the degree of uncertainty. That’s why I always use actual 2013 or 2014 generation numbers. These almost certainly won’t be representative of demand in future years but at least they give me a fixed point of departure.

      If the UK wants to decarbonize all of its energy sector and not just electricity generation (which supplies less than half of total energy consumed) then the only option I see is to expand electricity generation to the point where it produces enough kWh to run the entire transportation sector on “clean” electricity rather than oil. This would require roughly a doubling of current generation, which seems to be more or less what DECC has in mind. Scenarios that involve powering millions of vehicles with hydrogen from electrolysis and/or methane from cow manure are not going to work.

      • peter2108 says:

        Roger: “cow manure” – yes exactly. The whole decarbonising enterprise seems so wildly unrealistic that I cannot understand how it is taken seriously. But it is, and that I find quite scary at times.

  5. Pingback: Decarbonizing UK Electricity Generation – Five Options That Will Work | Energy Matters

Comments are closed.