The Economics behind Rooftop Solar and Wind Farms
A comparison of the economics behind the two biggest renewable generation technologies in Australia, today. As can be seen in the graph below of accumulated install, Australia's wind fleet has increased steadily over the past ten years, while solar took just five years to expand as much.
Note; this article is based on grid connected, Western Australian experience, May 2018.
First, a look at wholesale electricity prices and retail prices. The difference between the two is the difference in both time and space between the 'gate' of an electricity generator and where and when you flick a switch in your home. It is the sticks and wires and transformers and substations in between. It is the ancillary services such as frequency control, spinning reserve, phase balancing, metering, power factor correction and balancing. Not to mention billing.
Then there is the difference in how all of this is paid. In WA, there is a near real time market, the Short Term Energy Market (STEM) whose price can be viewed at wa.aemo.com.au along with an awful lot of other near real time data. However, this is only 5% of the market, the rest is tied up in Power Purchase Agreements between retailers and generators. They can be very long term and are usually confidential. Retail, on the other hand, is regulated and fixed for a year.
There are reasons why retailers pay less than they charge. Synergy is offering 7.1c/kWh (under their Renewable Energy Buyback Scheme) whilst charging 24.067c/kWh (A1 Homeplan tariff, excluding GST). When the system generates more than the household uses, which is certainly the case for most people too much of the time, the excess electricity spills into the grid. This is unscheduled. When consumption is higher than generation, and overnight, electricity is imported. As mentioned above, there are a number of services that are provided by the network system to make it all work. They need to be paid for but anxillary services are actually a very small cost.
Of course, neighbouring houses will absorb the 7c/kWh exported from a house with solar on the rooftop and be charged by Synergy at 24 c+GST/kWh. The difference of 17c/kWh to take electricity from one house to it's neighbours is no way reflective of the cost of ensuring this transfer can happen.
Any decision to increase the fixed costs of electricity to the customer because of the drop in volume of electricity they purchase, needs to balanced by an increase in the payment for exported electricity.
Interestingly, GST on the electricity per unit is now greater than the carbon tax. Riots anyone?
There is a fixed daily charge (equivalent to $346/year excluding GST) which, despite what is said on the Synergy brochure, goes nowhere near to covering the cost of maintaining the network and paying for any outstanding installation costs of new infrastructure. As consumption appears to have plateaued over the past 5 years (actually declined in the eastern states), the pressure is on network operators to charge differently. For example, charge according to maximum consumption during peak times rather than on volume. How this is actually done will be the issue.
Why install projects?
The primary incentive for installing solar on the roof, ie behind the meter, is to reduce electricity bills, saving at retail prices if possible. Whereas a windfarm is built to generate electricity for 100% export into the grid.
The income for a rooftop PV system comes from saving at retail prices and earning 7.1c/kWh ($71/MWh) for any electricity exported into the grid. The upfront cost quoted to the householder will usually have the Federal Government renewable energy certificates already removed from it. These certificates are called Small Technology Certificates (STCs, 1 STC equals 1,000kWh or 1 MWh) and the amounts are based on an estimate of now 14 years' worth (reducing to 13 years, next year, 12, the year after that etc) of generation at the location, reducing the capital cost of the system by 30 to 40%. By comparison, windfarm operators can only claim their green certificates (Large Generator Certificates, LGCs, 1 LGC equals 1 MWh) as they generate them with payment in arrears.
A final point about both STCs and LGCs, they are NOT paid by the taxpayer, state or federal. They are paid for by retailers who then pass the costs of purchasing the certificates to their customers. User pays - as it should be .
Unique to WA is the capacity market which, although currently under review, allows for an additional source of income. For ‘non schedulable' generators such as wind and solar, it depends on how much generation, measured in MW, based on past experience/monitoring+modelling, occurs at the peak load times. These load peaks occur mid to late afternoon after an accumulation of very hot days during working days.
For wind farms, the value of the capacity market mechanism is roughly 5 to 20% of income.
Rooftop solar is not eligible for this support, as individual projects are too small. The Australian Energy Market Operator (AEMO), who administers electricity on the SWIS (WA), estimates the benefit of rooftop solar as reducing the peak by 7% on the 1st of March, 2017 when the system had it's maximum load over the summer 2016/2017. The combined impact for pushing the peak forward into the evening and the reduction in demand from PV at the time was estimated at 265MW from an estimated total of around 700MW. This is a benefit to the system that owners of rooftop installations do not get any credit for.
Capital and Expenditure
Householders have to fork out an extra 10% on the price of an installation because of GST. Windfarm operators, being businesses, simply have to bear the cost of the administration of it.
Householders, however, do not have to pay for the capital cost of their existing connection apart from some minor charges. Windfarms (and any other new generator not behind the meter) will have substantial connection costs; the switchroom and the switchgear inside it, switchyards and power transformers, ancillary equipment such as communications, monitoring and fire alarms, and the power line out to the grid.
The existing connection for the householder also means that they do not have any extra network charges beyond an initial once off meter charge. In WA, there are no less than seven components to network charges (six are proportional to the maximum output or Declared Sent Out Capacity (DSOC)) for a windfarm connected at the distribution level, which have to be paid every month.
As PV has no moving parts, there is very little
maintenance, though triannual inspections are an excellent idea. Scheduled maintenance, depending on the turbine type, may involve a two man service crew putting in a full day, four times a year per turbine. On top of this, there are unscheduled maintenance visits when turbines send crews messages of the various errors that require attention. Depending on the fault, these events can be ignored or they can require the turbine to be out of action until the replacement part and, if really unlucky, cranes, turn up.
Most solar systems are paid for outright but projects as large as windfarms will have some debt/finance, however, the interest on this will be tax deductable. While the householders pocket the savings from their PV systems, windfarm operators running a company will be paying company tax. Though, this means shareholders will enjoy franking credits with their dividends.
Control and metering
Both PV solar and wind are at the mercy of the weather. Rooftop solar is not generally monitored centrally. As with individuals using vacuum cleaners or running pool pumps or any other large load, the individual impact on the grid is too small to worry about. For any generator larger than 1 MW (and possibly, some smaller), the network operator, Western Power, requires the Power, Voltage and Power Factor to be sent to their central control office in east Perth. This needs to happen every two seconds, otherwise the generator is turned off. This means some very reliable and very expensive radio communications.
For householders, electricity bills come bimonthly and have just the one number for consumption and for export and for generation. For bigger generators, these numbers are created every half an hour, so monthly administration consists of datasets of 1500 numbers or so. This is the kind of detail required to deal with different wholesale values based on time of day or the Short Term Energy Market. Thank goodness for spreadsheets!
Some comments on rooftop solar versus large scale solar
Currently most rooftop solar is where most people live in WA, in Perth. Better solar resources exist on the northern and eastern edges of the grid. Installing solar farms in such places as Geraldton and Kalgoorlie where substantial mining loads exist, transmission will not be such an issue until such projects exceed the local loads and look to supplying Perth.
There is a benefit for the grid to having solar generation spread over wide geographical areas but it is not as significant as it is with wind.
Use of resource
Locating solar panels on rooftops can mean access to the sun is quite comprised by shading from trees and neighbouring buildings and roof structures, and roof slopes not always facing north. Ground mounted panels can have optimum orientations in tilt and azimuth, and run a little cooler as heat is not trapped between the panel and roof. Single axis tracking is also possible, increasing output by 20 to 30%.
From the website of the Australian Photovoltaic Institute, the WA rooftop solar fleet would appear to have an aggregate Capacity Factor of around 16%. This compares to 25% for Greenough River based on it's 10MW of rated capacity. The Capacity Factor is 21% if based on the number of panels, 12MW. Also, being just south of Geraldton, the solar resouce is better than Perth.
With rooftop panels, there is no land loss but with fixed ground mounted panels (note this ground mounting can be quite expensive), the land cannot really be used for grazing or cropping. This is unlike wind, when livestock can graze right up to the base of the turbines. There are some innovative concepts such as growing wild flowers for honey bees (UK), in rice fields (Japan) and over water (Australia) where an additional benefit is the reduced loss of water from evaporation.
Sensitivity to change
As rooftop panels are distributed over a very large area and face different directions, the aggregate impact from clouds passing over is smaller and smoother. With panels concentrated in one area and all facing the one direction, passing clouds have a much bigger impact on big solar. Such big systems will have communication links with the network operator who may still require some kind of short term storage or other solution to deal with rapid changes in output.
Risk has two parts, consequence and likelihood. Smaller systems may have a greater likelihood of failure but beyond the household concerned, there are negligible consequences. If a household fails to follow through with a commitment to install, the issue is contained to the installer and their customer. If a committed 150MW solar farm fails to get built, outside those involved directly in the project, there are also consequences for the planning people in Western Power.
Once in operation, there is the risk of equipment failure. If your neighbour's 3kW inverter dies during a heatwave, you are not going to notice. However, if a 150MW project were to fail, the impact would be significant.
There are, of course, measures in place to deal with big plants failing to be built or tripping when in operation. They are simply not required for small rooftop plants.
Smooth versus Lumpy
Yet another benefit of small and many versus few and big is that the smaller projects are quicker to install and can start generating straight away. Bigger projects usually involve years of planning and can only start generating once the connection and at least one generation unit are commissioned.