Inside a network operator: How solar changed the game

energex

We’ve been hearing a lot about the solar revolution recently: from the largest generators and utilities in the US, the entire German power industry, from the world’s biggest solar companies, and from leading energy analysts from the likes of Macquarie Group, UBS, Deutsche Bank, and Citigroup who trackĀ the implications for listed electricity and energy technology assets around the world.

The basic synopsis is that the arrival of rooftop solar as a cheaper alternative to grid based power – as is now the case in more than 100 countries, including Australia –Ā is changing the nature of the energy game forever.

Last week, at the Solar 2013 conference in Melbourne, we got a rare insight into how this is happening in Australia – from the point of view of one of the major network operators. We’ve had plenty of graphs to suggest the changing consumption patterns and on the impact on pricing, but how exactly is it affecting the day-to-day operations of the people charged with delivering that service?

The Queensland utility Energex is interesting because no network operator has had to deal Ā with quite as much of a rush on solar as this network ā€“ centred around the state capital Brisbane and the high growth areas of south-east Queensland, including the Gold Coast, the Sunshine Coast and their hinterlands.

Energex already has installed 212,000 systems and is still adding them at 3,000 a month. Even in the first three months 41MW have been added ā€“ and it was continuing despite the fall in incentives. Talk of time of use pricing and further increases in electricity prices ā€“ and the continuing fall in the cost of solar – was the cause of that.

Mike Swanston, who goes under the title of “consumer advocate” for Energex, began his presentation by delivering a few home truths to the solar industry. The first was that the industry was kidding itself if it thought the network operators were not going to try to recover their regulated return on investment – even if solar was subtracting demand from the grid in a significant way. (This is an issue we raised on Thursday with our suggestion that perhaps the only solution is for the value of networks to be written down, particularly in light of some established over-spending, but more on that below.)

Swanston also sought to dispel a few of the myths that have been circulated in the media. One was that solar had not been having an impact on coal-fired generation. He noted that some 700MW of coal-fired generation had been sidelined in Queensland, a comparable amount to the amount of solar capacity in the state. ā€œYou donā€™t need to be Einstein to work out where that energy has come from,ā€ he said.

He also noted there was no evidence that solar households on juicy export tariffs (now ended) had shifted their use of grid power to the evening peaks. Nor for that matter, was there any significant growth in peak demand, particularly in residential areas. ā€œPeak demand is not an issue in the next five to six years,ā€ he said. ā€œThe network is in pretty good condition.

But he gaveĀ some fascinating insight into just what impact high penetrations of solar are doing to the shape of the load curve ā€“ and on the habits of household energy users.

Consider this load curve (in the graph below) from one of the network’s key feeders. There was a time when night-time electricity demand was so low that governments had to come up with proposals such as off-peak hot water to provide the generators some demand. Now, the night-time use on certain parts of the grid is virtually more than the daytime. Partly this is due to the enormous number of appliances that are on stand-by or charging during the night – plasma TVs, laptops, mobiles, and games.

Swanston said electricity from rooftop solar kicks in after the morning peak ā€“ to such an extent that on sunny days some feeders run backwards.Ā Then, hey presto, the sun goes down, householders return, turn on the plasma TV and the air-con, and the load rises from 3 per cent to 100 per cent in the matter of a few hours. Swanston said there was no doubt that this was a network challenge.Ā But it was not insurmountable. ā€œWe will fix it and we will deal with it.ā€

energex feeder 2

To give an illustration of what has happened over the Ā past four years, here is a graph (below) plotting the load on the same day (2nd Tuesday in October) as solar is being added to the grid.

As Swanston summarised it – midnight hasnā€™t moved, the morning peak is down slightly, the midday peak is ā€œgoing down the gurglerā€ and the evening peak hasnā€™t moved an inch.Ā ā€œThat presents technical issues,ā€ he noted. ā€œIt presents much larger commercial issues.ā€

Ā energex feeder comparison

And here are two more graphs (below) to add to the overall picture. Not only is the intraday load being reshaped, overall demand is decreasing from the domestic sector – mostly because households have added vast amounts of rooftop solar, but also because they have become more judicious about the way they use energy. The first graph below illustrates the seasonal changes – there was once a peak in winter and then in summer. Despite the huge additions of air-con, that summer peak has faded dramatically.

Ā energex consumption

The second graph below indicates the decline in the annualised use of energy drawn from the grid. Solar households have achieved substantial falls, but even non-solar households have reduced demand as the result of various energy efficiency measures, and as a response to rising electricity prices.

energex domestic annual

So what does this mean for networks? The essential problem from the network point of veiw is that they need to recoup their investment – which has been approved by regulators. Networks across the country had been investing huge amounts in the last few years – a total of $40 billion. Energex itself had spent around $600 million a year in the last four years, but this would drop off sharply.

But Swanston said the ā€œvast majorityā€ of urban networks are ā€œsolid as a rock.ā€ ā€œThe introduction of solar from a technical point of view is not a problem,ā€ he said. That should not be a surprise, because it echoes a CSIRO study last year that suggested that networks should be able to absorb up to 40 per cent with little problem. If there was an issue, it was not likely to be the fault of solar.

As mentioned above, the issue are not so much technical issues as commercial ones. Swanston vigorously defended the right of the network operators to recoup their investment, but he made clear that over the long-term it would require a changeĀ to the way that customers were billed – continuing on the kilowatt-hour basis would accelerate the “death spiral,” while raising fixed charges was also an unlikely long-term solution. (Others have pointed out that it would likely cause a backlash from consumers, and reduce incentives for energy efficiency).

Just how those new charges will evolve is yet to be seen.Ā Swanston said Energex was spending around $20 million installing new metres to solar households, and in regional areas, the influx of solar was having more of an impact because the ā€œthinnerā€ and ā€œweakerā€ networks were less able to cope.

He said it was clear that more households would turn to solar, and new financing options would also attract owners of rental properties. This would continue to reduce demand from the grid. ā€œThis is a huge problem for us,ā€ he said.

Swanston said battery storage would play a key role.Ā The eventual solution would have to incorporate demand management ā€“ be that in storage (such as batteries), or in traffic control.

He suggested that there would need to be a revolution in the way that electricity and grid access was charged. Battery storage would be important because it could flatten the remaining peaks. Home storage, where people had options and ā€œcould make their own decisionsā€, was the likely future. How that pans out for tariffs and networks was not yet clear. At the moment, the opportunity for battery storage was a straight arbitrage. ā€œHow cheap can I charge it, and how expensive can I spend it.ā€

He didnā€™t see much interest in networks using battery storage on a ā€œutility scaleā€, although he said networks in rural areas , where the ā€œthin networksā€ were much more susceptible to increased penetration of solar, and faced higher costs, may find it cost-effective to invest in larger storage facilities rather than having to upgrade those network lines.

 

 

Comments

27 responses to “Inside a network operator: How solar changed the game”

  1. wideEyedPupil Avatar
    wideEyedPupil

    What if storage was built into the generation facilities as it is with the Concentrated Solar Thermal plants with Molten Salts Storage as promoted by Beyond Zero Emissions and others. These plants are operational in SPain and the USA and China and India have such plants on the drawing board ATM.

    1. Louise Avatar
      Louise

      I predict the next big thing in solar power is to install a small battery for emergencies and buffering.

      A home owner could install some electrical timers available at many hardware stores and thus turn their (not so smart) washing machine or dishwasher into a smart appliance, which will start operating when the next solar power producing period starts.

      Super energy efficient fridges and freezer only need to be cooled down during the solar power producing periods.

      Laptops consume less energy than desktop computers.

      By time shifting the home owner could do with a small battery of say 2 – 4 kW/h and dramatically reduce their impact on the electricity grid.

      And people who buy a car like the Mitsubishi Outlander Hybrid, available in 4th quarter in Australia, but already on sale overseas, could plug their house directly into the car and could could power their house when the sun does not shine.

      Thus avoiding to have to buy a large battery array that could power the house in case the sun does not rise for a week.

      The car could also be used as a generator should the battery run out of electricity.

      .

  2. Rob Avatar
    Rob

    It would make a lot of make for utilities to strongly encourage the uptake of electric vehicles in order to support their business model. I drive one, is a brilliant car and has had a massive price drop recently. Not to mention the clean air benefits and health benefits that flow. Also much cheaper to fuel than buying petrol. And even 100% energized by coal fired power it produces fewer emissions due to zero consumption while stationary. A utility could justify subsidising the EV purchase cost by locking the owner into an electricity supply plan that would still be cheaper than petrol. I can just see Origin opening up a car dealership network. Different skills needed, much easier to maintain so perfect time to upend the existing car dealer model.

  3. Warwick Avatar
    Warwick

    Given that Energex clearly shows that solar PV has not affected the network peak demand over the last 4 years and the costs of a network relate largely to capacity, why should a PV owner get a cheaper network bill?

    1. taiyoo Avatar
      taiyoo

      Warwick, is the network peak demand on second Tuesday of October ? If not, then the graph does NOT clearly show this.

      The network costs the utilities are trying to recoup are for the infrastructure required to service the real peak demand in the mid-afternoon in January. Why didn’t they show this graph and how it has changed over the years? Possibly because it doesn’t suit the narrative that Warwick has swallowed hook line and sinker.

      1. picoallen Avatar
        picoallen

        Good point. That may very well be a cherry pick. Another day might show a higher peak in 2012 or a dramatically lowered one. What does a typical day look like?

      2. Warwick Avatar
        Warwick

        Taiyoo, if it is the case that network maximum demand is really reduced by solar then why doesn’t the PV industry get behind a demand charge for homeowners? PV owners would be better off (if they actually reduce peak demand)! Perhaps the PV industry knows this?

        1. taiyoo Avatar
          taiyoo

          Warwick, it is a fact that PV has reduced network maximum demand. Demand charges, on the other hand, have nothing to do with reducing network maximum demand and are all about utilities trying to recover some of the poor investment decisions they made in gold-plating the network based on demand estimates that failed, as you do, to understand the effect of PV on the network.

          Tell me how does slugging households and businesses every month, even winter, with a peak demand charge reduce a network maximum demand that occurs for a few hours in the height of summer?

          1. Warwick Avatar
            Warwick

            Taiyoo, please prove that PV has reduced network demand. Any citation from any network that shows in Australia that maximum network demand is consistently reduced would be sufficient….

            Unfortunately, demand charges have EVERYTHING to do with reducing network demand. If you actually took the time to understand the networks you’d observe that Energex customers above 100MWh/pa are already charged for their maximum demand. This means if you have a business and install PV that reduces your maximum demand or make other measures such as energy efficiency, power factor correction (where the demand is in kVA such as NSW) etc, your network bill will be less. Proactive businesses have been managing their maximum demand for decades…

            Charging or “slugging” people or businesses with a maximum demand charge based on their annual maximum demand actually encourages people to reduce their maximum demand as their bill will be less, which is usually a suitable reward. Put simply, reduce your maximum demand (possibly for those few hours at the height of summer or whenever it is) and your bill is less. Those who use the network capacity should pay for it…this should be obvious!

            If PV reduces network demand then those PV owners have nothing to fear from a demand charge and will be relatively better off. It will encourage storage and penalise inefficient airconditioning load. The only logical reason why someone would rally against this is if they aren’t reducing their maximum network load.

          2. taiyoo Avatar
            taiyoo

            Warwick, you are confused between NETWORK Maximum Demand and INDIVIDUAL Maximum Demand. You are also confused about how demand charges work – the maximum demand charge for an INDIVIDUAL business or household is NOT based on their “annual maximum demand” as you said – it is measured and charged in each monthly billing period.

            How does slugging an INDIVIDUAL household or business with a demand charge in June reduce the NETWORK Maximum Demand in January?

            The other side of demand charges, as i’m sure you do know, is a too low energy tariff for the actual energy consumed. This distorts the market as it takes a chunk off the real cost of each unit of energy and loads it into the maximum demand charge. On a NETWORK scale this does nothing to encourage overall energy efficiency as there is no incentive to reduce energy consumption for the rest of the billing period, and in fact encourages energy consumption by its artificially low price. Coincidentally, behind the meter PV is most financially viable when the true cost of each unit of energy is reflected in the energy tariff.

            The utilities aren’t confused, they know this. They also know the PV industry has installed 2.25GW since 2009 and their NETWORK Maximum Demand forecasts, which they use to justify gold-plating, have been getting further from reality over this time.

            You can do your own research – the AER has some graphs on the performance of the Australian Energy Industry, there are plenty of sources showing the installed PV capacity. The AER graphs showing the reduction NETWORK Maximum Demand and the corresponding reduction in >$5000/MWh trading intervals since 2009/10 might help you understand why the utilities are fighting PV so aggressively.

          3. Warwick Avatar
            Warwick

            Taiyoo, I am in now way confused and your ad hominem attacks on being “confused” or networks “slugging” people are inappropriate. Everyone’s individual demands matter as in aggregate they add up to the total maximum demand in the distribution system. The annual maximum demand charge on Energex’s tariffs is referred to a “capacity” charge, some networks call it “annual maximum demand” or “12 month rolling demand charge” or a capacity charge. It doesn’t matter what you call it but an annual “capacity” charge is appropriate so that users pay for the capacity they require from the network.

            I’m not sure why you argue that a lower energy charge increases energy consumption but then fail to see why a higher demand or capacity charge would increase the motivation to reduce demand on the network. More efficient use of the network reduces the need for additional investment or defers it, potentially benefitting all consumers.

            “Gold plating” is a somewhat subjective view and it does occur in some cases but the networks have also been accused of underinvestment in the past such as Energex which used garden sprinklers to keep the transformers cool in summer about 10 years ago when the network was struggling to cope.

            I must correct you on another error. The AER deals with spot prices that are settled at the regional reference node (Southpine in the case of QLD) and have nothing to do with the demand at the distribution feeders which is what Energex is talking about. I might also point out that January this year in Queensland was in fact very volatile with a number of prices above $5000/MWh and was one of the It’s a common misconception that total regional demand is coincident with the distribution system serving homeowners. Obviously there are some loads that are distribution connected that have a consumption profile similar to the regional demand and often these businesses can benefit significantly by PV such as factories or warehouses as they are metered on demand charges amongst others so get a real saving.

            The real issue is that PV owners should have nothing to fear from a demand or capacity charge as if they are actually reducing network load they will save money compared to the average consumer. Airconditioning users will likely pay more under this scenario and battery storage will become more compelling. If the charging arrangement is maintained PV owners will pay a disproportionate share of the network bill. This is simply unfair and this should be aware to everyone.

            There’s no point attacking individuals just because they point out the inequity in the current system nor is right to assume a “David and Goliath” struggle because the network businesses are large, spend a lot of money on advertising and often waste money on football sponsorship. Hmmm, I notice big PV companies spend a lot on advertising and some sponsor football teams… Perhaps they’re there to make money too…hence you can’t trust them either?

  4. Bob_Wallace Avatar
    Bob_Wallace

    I don’t understand the “thin network” issue. I’ve read something about the grid not being able to deal with the input from end-user solar but that doesn’t make sense to me.

    If we look at loads the grid would normally support – a clothes dryer (5 kW), 3 ton AC (5 kW), dishwasher (2.4 kW), microwave (1.5 kW) – all of these and other devices such as refers, TVs and lights can easily be on at the same time. An individual house could, at times be pulling 15 kWh. But the grid can’t handle 3 to 5 kW flowing the other way from solar?

    1. Concerned Avatar
      Concerned

      Perhaps not everyone is home when the sun is out?

      1. Bob_Wallace Avatar
        Bob_Wallace

        You miss the point (could be the way I stated it).

        If the grid is capable of delivering 15 kW of electricity from “coal plants” to individual houses then why would the grid struggle to deliver 5 kW of solar from houses to where demand exists?

        Copper is an equal opportunity carrier.

        1. Concerned Avatar
          Concerned

          It was never designed that way.

        2. ArchCC Avatar
          ArchCC

          Hi Bob-Wallace. You’re mistake is that you’re equating the flow power (kW) to that of water. Electrical power isn’t a fluid and wires aren’t pipes. Although it can, at some level, be thought of in the same terms, there are additional constraints on how electricity can be transported and distributed.

          For example, one major problem is “voltage rise”, which is more sever on thin networks. It’s cause is as follows: In order to export energy onto the grid, household inverters have to output power at a higher voltage than the local distribution line so that they “push” power back on to the network. When this happens on the end of long or thin feeder lines, the impedance (resistance) on the line causes voltages close to the end-user generator to rise above safe level. That is, the voltage rises to levels at which it can be dangerous to run normal household appliances (and outside the statutory limits).

          Another consequence is that voltage rise can force the inverter up to its overvoltage level, at which point it disconnects from the grid. If several households in the same area do this at the same time, there can be rapid changes in voltage which is also dangerous and unstable.

          Hope that helps clear up why some parts of the grid are not able to deal with the input from end-user solar.

          1. Bob_Wallace Avatar
            Bob_Wallace

            Sorry, I don’t buy that.

            No roof-top solar system or neighborhood group of roof-top systems would be capable of creating the sort of “push” you speculate. Plus output voltage and frequency are locked to grid levels.

            The grid deals with very large generators such as coal and nuclear plants going on or off line very rapidly. It deals with very large demands suddenly appearing or disappearing.

            The potential supply coming out of a neighborhood is less than the demand potential for that neighborhood. A 5kW rooftop solar system produces the same amount of demand that a clothes dryer creates. The grid is designed to supply a clothes dryer, AC , microwave, refer, lights, etc. all at once which can easily be 3x the power created by a roof-top system.

            You’re making an argument along the line of a fly lighting on a passenger plane wing causing it to tip sideways.

            “Another consequence is that voltage rise can force the inverter up to its overvoltage level, at which point it disconnects from the grid. If several households in the same area do this at the same time, there can be rapid changes in voltage which is also dangerous and unstable.”

            That’s a load of bunk. Grid-tie inverters match their output to grid voltage and frequency. Don’t know who’s feeding you that stuff, but it’s just not factual.

            And – electricity does flow both directions in a wire as do liquids in a pipe.

          2. Concerned Avatar
            Concerned

            Bob ,he is really correct.

          3. Bob_Wallace Avatar
            Bob_Wallace

            Give me a link to a technical paper.

          4. ArchCC Avatar
            ArchCC

            What, even in the middle of the day, when no one is at home consuming power, and all the rooftop PV in a district is generating at its peak?

            You might be thinking in terms of water flow again.

            Look, I’m not going to give you a link to a technical paper because there are journals full of ways to address this problem. Instead, read this http://en.wikipedia.org/wiki/Ferranti_effect and then consider what happens when voltages are applied at _both_ ends of the line.

          5. Bob_Wallace Avatar
            Bob_Wallace

            Even when no one is home refrigerators and air conditioners run.

            The grid deals with power spikes. A large motor turning off can create a spike. Take a look at your link.

            And one certainly can get pressure spikes in pipes/water flow. Same-same.

          6. ArchCC Avatar
            ArchCC

            This is getting tedious, Bob. You’ve switched to talking about spikes; I’m talking about voltage
            rise. If the voltage across a district are maintained at a higher-than-safe level for a
            long-enough period of time, it will do damage to equipment, including grid infrastructure. Network companies take requests to test for over-voltage seriously because they are really keen to protect their own assets and not be fined by the regulators for delivering power outside of the statutory voltage limits.

            There is no debate on this issue, it happens.

          7. Bob_Wallace Avatar
            Bob_Wallace

            The only way that rooftop solar is going to do damage on the grid by long term high voltage is if inverters are being installed with too high cutout settings.

            If that is actually occurring then adjustments need to be made.

            Have you seen any data to support the charge that lots of end-user inverters are set higher than grid standards?

            If you go back and read ArchCC’s wiki link it’s about voltage spikes.

            I agree, the conversation has become tedious. It seems that a few, for reasons they haven’t made public, are looking for reasons to oppose roof-top solar. They are taking what seems to be unlikely and at best very uncommon events and blowing them into something large.

          8. Concerned Avatar
            Concerned

            Where I live, Energex must be wrong. You should contact them.

    2. Catprog Avatar
      Catprog

      To use a water analogy.

      Imaging a dam that feeds a river. the river can handle lots of water. Try to fill the dam from the houses and you run into problems.

      I believe the problem is the transformers have been designed to take power from the station and put it onto the distribution grid.

      The transformers were not designed to put it into reverse.

      1. Bob_Wallace Avatar
        Bob_Wallace

        Obviously enough hoses will create more flow than a river. “Enough” is a key word in the sentence.

        If the river flows 1,000 gallons a minute and a single hose flows 1 gallon a minute then 1,001 hoses will “fill the dam” faster than the river.

        As far as I know transformers are can be either “step up” or “step down” depending on whether you feed into the larger or smaller coil. If there’s some other type of transformer that grids use I’m not aware of it.

        1. Catprog Avatar
          Catprog

          My point is not about the amount of flow. My point is the level of water in the dam is higher then the river level. It won’t fill the dam until the river is higher then the dam.

Get up to 3 quotes from pre-vetted solar (and battery) installers.