In reading current electric energy publications, it seems that a new study is always being released touting the competitive and declining cost of solar PV. A recent Greentech Media article cited a North Carolina Clean Energy Technology Center report that said “a fully financed solar PV system costs less than the energy purchased from a residential customer’s local utility in 42 of the 50 largest cities in the United States.”
And last week, a PV Magazine article cited a study performed by a leading Middle Eastern bank that said “solar will be at grid parity in 80% of the globe by 2017.” The study further stated that a recent Dubai Electricity and Water Authority solar auction shows that solar is “competitive with oil at $10/barrel and gas at $5/MMbtu.” Many more could be cited. If these numerous studies and reports are to be believed, then solar is currently or will soon be competitive with grid electricity in most places. Some see the solar industry in such good shape that they have boldly declared solar tax incentives bad for the long term-health of the solar industry and have called for the incentives to be phased out.
Yet when a utility proposes to end the net metering incentive for rooftop solar, solar industry representatives cry foul, accusing the utility of trying to kill the industry. Even in Hawaii — where rooftop solar competes against some of the highest retail electric rates in the country — solar advocates are characterizing Hawaiian Electric’s plan to change net metering incentives as “…following the national utility playbook of trying to stop competition from solar.”
I am more than a little confused by all this. Solar companies claim to have a low-cost product, and more often than not, they compete against utilities that are prohibited by their regulators from competing in return. Yet, their actions seem to indicate that perhaps they’re not as competitive as they claim.
Here are the questions: How can the solar industry be weaned off the “incentive bottle” so solar can become the competitive energy resource the industry says it is? And specifically, what needs to be done about solar tax incentives and net metering?
There is no doubt that the cost of generating solar energy is coming down. Just in the last few years, we have seen prices that have dramatically come down to about one-fourth of what existed earlier. This does not mean that we are there; it needs to come down a lot more. This will happen in several ways. Either existing technologies will need to reach the next level of scale so that the cost of manufacturing can come down dramatically, or new technologies will need to be developed with a greater yield and better performance, or most importantly, we will need to find avenues to incorporate solar generation into existing materials such as roofing materials, window glass and other common construction components. We believe that a combination of the points above will accelerate the acceptance of solar into the mainstream much more than it is now.
However, it is very important to note that regardless of the reduction in costs or acceptance into the mainstream, certain limitations of the solar cell needs to be accepted. These limitations are well known and all associated with exposure of the PV-cell to sunlight and the unpredictability that comes with it. New mechanisms need to be thought through in how to integrate solar into the power system. We believe that New York and California in combination are looking at new and innovative mechanisms to do just this. Not all of us can (or should) hope for storage to be the white knight to bail us out.
The business case of solar is also complicated and varies wildly from locations such as in sunny parts of Arizona when compared to cloudy parts of Washington. This makes it complicated for solar to make broad statements or conclusions about the efficacy of this technology to be the panacea for all of our energy woes. The business case also gets complicated because, unlike other existing sources of generation, solar is not dispatchable — we have to take what is being generated when it is being generated.
Very often, solar implementations are also closely interconnected with the discussions on “net metering.” Net metering is a complex subject, right now focused on making the broader population pay for key changes to how the system is monitored based on the abilities of a few wealthy people who are able to place solar installations on their properties. However, in the long run, net metering needs to be the norm not for supporting the few that have solar installations in their homes, but more to support a systemwide flow of two-way power that is quite fundamental to a smart grid.
Lastly, while much brouhaha has been made about incentives to solar, they are not the only industry getting tax incentives. Every form of generation gets tax incentives just in different forms. This is another of those not-so-easy solutions. We believe that incentives need to be analyzed in their entirety for all generation technologies (fossil-fired, hydro, nuclear, wind, solar and others) so that an apples-to-apples comparison can be made.
In response to John (and citing Europe’s experience), the reality is that PV from roof-mounted household systems flows to neighbouring houses.
There are two ways to account for this flow: net metering that pays for the electricity at some rate or the households can build their own mini local energy market (LEM). A modest DUoS (Distribution Use of System charge) could be levied on energy flows in such a LEM.
In the UK and other locations, there is in any case a fixed daily charge for the network — regardless of energy use — so perhaps the DUoS charge could be avoided.
When you look at energy-only charges in households, most are somewhat lower than LCOEs for PV (although these vary considerably within Europe); this makes it problematic supplying next door with an energy-only kWh, if you see what I mean.
One way to address this could be to add something to cover avoided network losses, although this is likely to be rather small.
I suppose it comes down to this: although PV equipment costs have dropped, installation costs have not shown the same "trajectory," so you still have residential roof top systems that have LCOEs well north of 12 euro cents/kWh in Europe
Amusingly, Cyprus, Malta, the UK and Ireland all have energy-only costs well north of 12 euro cents/kWhr. Of course, none in a million years would ever consider allowing the locals to have a local energy market; after all, that would provide competition to incumbents, and we can't have that, can we?
Solar installations can have better economics (i.e., price points) at the customer level (i.e., distributed resources) based on the following:
1. The tried-and-true way that prices will likely drop as the PV equipment and installation market develops
2. Energy storage will interact to firm up the solar energy better and provide less nameplate need, and
3. If aggregators (or utilities, if allowed) develop a business model by which they reach thousands of solar devices (including storage devices if applicable) in a market enabled by robust transmission, using law of large numbers, they can play in ancillary markets and lower individual price points. Let's not forget that if aggregators can reach these devices, they can also reach devices such as hot water heaters to enhance the play. In addition, there may be leasing arrangements by these aggregators (or utilities) that lessen the burdens of customer cash outlay, on-going maintenance, etc.
What we must recognize is that reliance on the grid will be needed for the foreseeable future and that grid (i.e., wires) cost must be recognized and not totally offset by energy injection into the grid. In addition, the standards by which these devices connect must provide robust ride through (so that all do not trip out due to a grid fault) at least, and be able to utilize the inherent voltage regulating capability of these devices. Let's develop a win-win grid-customer conversation rather than only seeing the one-sided view that connection costs and standards are barriers to entry, and the wires are not relevant.
I have a follow up question. Mani mentioned New York and California are "looking at new and innovative mechanisms." At this point, are any states headed in the right direction?
I'm not sure which state is doing the right thing, but when many states come forward with ideas, the right one in time will reveal itself. We need time and a host of ideas to discern. It is obvious though, that states with relatively higher rates will be first movers, such as California and New York. Then once we find the "right" way, someone will come along with a better idea. That is the American way after all.
In this next step — focusing on redesigning how we do business in the grid — I believe New York and California are leading the effort. Now I am not sure I can say one state is moving in the right direction or not. I believe we will learn from all the steps, and somewhere in the near-to-mid-term future, we will figure out the right approach. In the end, it may not be either California or New York or something else. It may be a combination of all of the above.
In my role as a Trustee for the Long Island Power Authority (LIPA), I have had a ringside seat in witnessing what is going on in New York with respect to revising the electric supply structure of the state.
Speaking as an individual, it appears that many actions are being taken and much pressure exerted by the state's utility regulatory body before anyone has really answered the hard question of whether or not this new distributed paradigm is more economic and reliable than the current system. Maybe I missed it, but, for me personally, I have yet to see any hard numbers and analysis transcending all the philosophical and social justifications. In a rush to make history, the state may be putting the cart before the horse.
As much effort and cost are being expended by LIPA to cooperate with this effort, it should be kept in mind that the Authority, as a public utility, is not regulated by the New York PSC like the investor-owned utilities, and as such, if it is discovered at some point that embracing the new paradigm is not in the best interests of its customers, then the utility should not hesitate to pursue a different direction. This also may be something the investor-owned utilities should consider as they deal with the regulators.
I have issue with the statement of cost as it does not address compatibility to the system.
The 2014 CIGRE symposium in Brussels had multiple presentations on impact of PV and renewables to the grid. PV and wind create both voltage and frequency issues that create great cost in the grid. A market was not established in Europe or England to provide the necessary voltage and frequency mitigation particularly at the distribution level. Utility are required to spend a fortune installing SVCs and STACOMs with no return and negative revenue. Not to mention, no way to return the captured investment with negative flow. The picture painted was ugly. The transmission market is better, but the same problems exist: who pays to make renewables work?
The second issue is developing models to predict the generation market for renewables. The tools right now are not accurate and significant spinning reserve needs to be running just to keep the system falling down. Examples of recent failure are the 2009 Texas blackout and 2009 Spain blackout. You can call this scenario the day the wind stopped. Without good day-ahead modeling and the proper amount of spinning reserve a non-adaptive market in place will create many such blackouts. Generation models for large integration of PV are yet to show great promise. PV models are a work in progress.
Not to be an alarmist, but the technology is not mature at the distribution level yet. Energy storage may be able to help with some of problems but will not be cost effective in small amounts at the distribution level due to conversion coast. Use of EVs is not practical in most of the U.S. as the service drops and transformers are too small to get the power stored, back in to the grid. Significant upgrades in distribution assets will have to take place before EV grid solutions will be much help. Demand response maybe some help but not for mitigation for voltage issues.
Most of my concern is distribution since transmission does have some of the rules in place to make the generator pay for the impact to the grid. The first cost of the renewable device is not a true cost of installation to the grid. Somebody has to pay for what renewables do to the grid. For the most part, right now, the burden is on the utility.
I agree as alluded to in my last paragraph. We must capture grid costs as these go in, and the standards by which they connect must be considered as well.
In response to Mr. McVey's posting, I will make the following comments with respect to Europe. Two EU member states installed large amounts of PV in the period 2007 – 2014 (Germany 25 GW and Italy 14 GW) changes in costs due to transmission and distribution have been +1% rise in Germany and -6.4% fall in Italy. In that period, Denmark saw costs rise by 23% at a time when it was building new off-shore wind farms (cost rise due to transmission). PV causes significant network cost rises? Not in Europe.
With respect to the comment "a market was not established in Europe or England to provide the necessary voltage and frequency mitigation," this suggests a misunderstanding of how things "work" in Europe. Certainly in the UK, RES network costs are NOT socialized. If you want to connect PV or wind, you pay for the connection and any impacts that has. However, the situation varies from country to country. Germany has socialised off-shore costs, for example. The case of frequency has produced no problems. Monthly auctions for the provision of frequency response — and there are plenty of providers all across Europe — is a highly competitive market. If Mr. McVey's argument is that RES gets frequency response for free, well so do plenty of other generators. The provision of frequency response is made, mostly by (using UK terminology) "balancing mechanism units" – BMUs. That everybody gets a free ride from these large synchronous units is not a big deal because the cost is trivial.
With respect to SVCs/STATCOMs, I only wish Euro DNOs had spent a fortune. I worked with a supplier of such products. Sadly, many Euro DNOs are not very interested, even at semi-transmission voltages such as 132 kV where an SVC could be used to adjust power flows. I have one STATCOM project running in Europe; it is likely to be the last. With respect to the statement "utilities are required to spend a fortune installing SVCs and STACOM’s with no return and negative revenue," name the Euro ones. I'm absolutely fascinated who they might be.
In the case of transmission, the statement "significant spinning reserve needs to be running just to keep the system falling down" again with respect to Europe is wrong as follows.
Most Euro TSOs have incorporated day-ahead weather forecasting into the operation of power system control and dispatch. National Grid in the UK at the end of 2013 stated that it had no control problems with the existing RES installations. In February 2015, Elia/50Hz (North Germany) noted that it had radically changed its view of how much RES its network could handle (in the 2000s, 4%; now they regularly handle 30% to 40%) given developments in forecast software. In real time, the Danish power system control centre compares the actual output of RES against the prediction made the day before. The error of actual vs. predicted is then used to forecast the output of RES some hours ahead. It is claimed that this approach almost eliminates errors in system operation. The box below gives a National Grid view (UK) as of end 2013.
Nigel Williams, head of electricity systems operations at the National Grid, “"When we first got wind on the grid we had a few choppy days, but we are managing it very well now. Our forecasting is very good."
The grid gets useful wind forecasts 10 days ahead, and 24 hours ahead can predict the amount of electricity that will be generated to within 4%. Earlier in December (2013), a wind power record was set of 6 GW accounting for 14% of all electricity at the time. The grid managed a 2.5 GW drop-off over a few hours as winds surpassed safety limits in many places.
"I don't see an upper limit to how much wind we can accommodate [on the grid]," said Williams.
There are some new coal stations coming on stream in places such as Germany (a number already exist in Denmark) that can ramp at 3% to 4% of rated output per minute and can cycle down to 10% to 20% of rated output, which of course means that "significant spinning reserve" is not needed; synchronous systems can respond quickly.
"The tools right now are not accurate." Sorry, but not according to Euro TSO experience (and I speak to them).
And again "Generation models for large integration of PV are yet to show great promise. PV models are a work in progress" Germany does just fine with 38 GW on its system. Actually, the problem will come if/when there is lots of residential PV (of similar size) and everybody has battery storage (of similar size). Think about it. Ditto Italy.
In conclusion, European networks have some problems, but for the most part have adapted well to RES. I have generated a long paper on this subject, and if Penton is interested, they are welcome to publish. My responses to Mr. McVey's e-mail are based on my understanding that they included European networks. If that was not the case, then feel free to ignore the above.
I believe distributed energy resources have a major role to play in the future of our integrated energy systems. And rooftop PV will be a major part of this. Our recent utility benchmarking study of the largest U.S. and European electric utilities confirms that utilities agree on this point fully and are implementing strategies for this.
I have to second Mark's comments with regard to the need to ensure a safe and fair approach to integration. But the question here is less about the technology at this point — while there is still work to be done around storage of course — and I believe is more about proactive and fair rate setting. Whether it's some form of distribution/delivery charge or charges for necessary integration expense, equitable rate redesign must be on the table over time.
Meanwhile, there can be no individual free lunches, and additional attention needs to be given to both the collectively shared value of rooftop PV in system utilization benefits and the very real costs of ensuring the very mundane but real issues including efficiency issues created by phase imbalance on distribution systems as the result of unpredictable and, as Mark said, still hard to model distribution power system effects.
In theory, net metering and two-way power flows sound great, but each utility and each utility jurisdiction needs to create rules and rates that work for their individual communities best interests based on a tailored approach.
Some communities and utilities may find excellent business cases for encouraging distributed PV and net metering aggressively right now. For others, based on things like their position in the capacity markets and past asset and system investments not yet fully realized, such incentive policies may end up being less optimal near term. This is where considerations such as specific PV-related fees should be fair game and not subject to solar advocate "histrionics" and accusations of nationwide "conspiracy" by the "big bad" utility companies bent on avoiding competition. All of this has actually bothered me a great deal.
Suffice to say anyone who wants to go fully "off grid" given today's power economics should be told to have at it!
But folks can't both rely on the grid capacity to be there when their panels break and also bitch about equitable approaches to ensure fairness for all. This whole line of discussion by solar advocates actually has been upsetting because utilities are in fact driving massive growth in solar in places like here in Georgia where I live. So the accusations of conspiracy against solar by private vested interests like those of Elon Musk and company must be confronted with reality and rebuttal by the men and women in utilities, both public and private, who serve their neighbors and communities day in and day out ...and when the sun is shining AND when it's not! Suffice to say the linemen who put their lives on the line also deserve safety first and foremost, as well!
I like the discussion so far. I too find it interesting that many of the suppliers talk about reaching grid parity but none are willing to support reduction of incentives. I guess that is true, in most cases, people do not like to see diminishment of their economic models, even when they can and should project that they will change.
I believe that my colleagues have admirably covered many of the key issues and areas of concern. There is one, however, that I do not believe has been adequately addressed and that is the issue of planning for failures of the PV systems.
I was speaking with a utility executive some months back and asked him the question: “What happens if we see a ‘type failure’ in a certain vintage of panel or inverter?” He replied that he had been worried about the same thing and that when he brought it up their external consulting firm (name brand) said that it was not an issue, that even if they had several hundred MW of failures over a 12- or 18-month period, by that point in time there would be other technologies that could be deployed to replace the lost resources. At a macro level, this may be true. What happens at the feeder level? I believe that this is an area that is largely overlooked.
I have noticed a trend among many of the people I know that have significant rooftop solar installations. Over time, they export less energy even when there is net metering in place. I don’t believe that this is the result of reductions in output of the panels. I have noticed that for many of these people they have become less sensitive to energy usage. They have added more load to their homes, and they are less energy conscious. I know that this is not true for everyone, and I do not mean it to be a criticism. They have a right to have the technology and appliances that they want, and they have a right to run them the way they wish. The problem for the grid is the potential for unseen load growth. In most, if not all, cases, the utilities do not get to see behind the meter, so they cannot tell how much load growth is occurring as a result of the low-cost energy from rooftop units.
So what happens if that rooftop system fails? If there has been unseen load growth no one has planned to cover that load. No one in the utility knew it was there. What happens if there are multiple failures on the same substation or the same feeder? Everyone served by that portion of the system is then at risk. The utility really has no means to plan for those failures, nor does it have a means to demonstrate the level of investment that it needs to put into distribution system renewal to cover that possibility/eventuality.
The regulators and utilities are not engaged in this discussion. Should they be? We have seen throughout our history that equipment/technology tends to reach its end of life in a normal distribution. Often that normal distribution curve occurs sooner or later than expected. A few weeks ago, there was a significant hail storm in the San Francisco Bay area. Could severe weather exacerbate the risk of failures?
Looking at the history of PV installations, the early adopters were taking advantage of tax breaks and other incentives. This resulted in PV being concentrated in specific portions of the system at specific periods in time. This suggests that failures are likely to occur in similar concentrations. If I were still running the distribution asset investment program, or still accountable for distribution system reliability, I would be very worried about the risk that is being created. It could be supported by better visibility behind the meter and installation of distributed resources including distributed PV. Ultimately, the regulators, the utilities and the PV industry need to address the issue of who is accountable for long-term reliability as the fleet of PV beings to fail due to age or weather events. They will need to address the questions of system planning and probabilities as well as the investment levels required to support some amount of backup capability from the distribution system. I believe that the conversation is overdue, and it would be good for the industry (regulators, utilities, PV suppliers, customer advocates, and policy makers) to be out ahead on this issue rather than playing catch-up.
To join the DC power conversation, please comment in the box below.