Solar modules. Author: Oregon Department of Transportation. License: Creative Commons, Attribution 2.0 Generic.
Today's still young solar power industry more than often turns a blind eye to long-term operation and maintenance (O&M) costs, but ignoring reality in the hopes of “big short” returns is a dangerous gamble.
While the financing period for most renewable energy facilities is 20 years or less, well-built solar infrastructure has the very attractive potential to continue to pay dividends for decades after the original project debt has been paid off.
The catch is that in today’s young solar industry, long-term costs such as O&M for these assets have so far largely been ignored, potentially threatening expected long-term returns.
With project ownership designed to maximize the return on investment (ROI) with the added benefit of tax equity, the chain of ownership is extremely complex. Developers contract an engineering, procurement and construction (EPC) company to build the project, then flip it to a long-term owner, who in turn bundles the project into a portfolio before it is sold to yet another third party. The current practice is to turn a blind eye to the elephant in the room which is long-term O&M costs, rather than mitigate the looming economic risk it presents.
The industry is guilty of perpetuating two fallacies that threaten solar's upward trajectory: 1) Long-term O&M costs are going to be much lower than they will actually be and 2) O&M costs are the identical across all solar systems, and therefore need not be considered in the decision making process.
If we’re to learn anything from other industries that have faced their own economic reality checks, it’s that ignoring reality in the hopes of “big short” returns is a dangerous gamble.
The realities of long-term O&M costs
Consider the case study that RayTracker provides. RayTracker popularized distributed tracker technology in the early 2000s. (The company was acquired by First Solar in 2011.) RayTracker touted that their system would be maintenance-free for 20 years. Instead the actuators have failed on average after 25 months! Even if the actuators are easily replaced, the cost in labor and downtime is enormous.
If actual costs aren’t anticipated, the solar farm becomes much less profitable than anticipated. However, who is going to confess to that mistake? Hence the error continues to propagate.
With more and more solar developers electing to install distributed solutions, it is essential to accurately model two decades or more of operation and maintenance costs up front.
Distributed systems have multiple orders of magnitude in additional points of failure than centralized systems. For example, while a centralized tracker will have as few as 80 drives per 100 MW capacity, a distributed tracker will have more than 4300. Similar math applies to centralized, string and micro-inverters. In a distributed system, even a simple visual inspection is a significant and expensive undertaking much less rolling a truck to investigate a potential failure.
The reality is that the thousands of parts of an electromechanical system will fail at a certain rate, and in a time period often far shorter than the useful life of a utility scale generating station. Therefore, the industry needs to be much smarter about collecting and analyzing performance data and leveraging that data for both better modeling and more cost-effective O&M strategies.
What’s at stake
Imagine what will happen when O&M providers begin to realize that their contracts are under water. Will they honor what’s stipulated in the contract? Possibly not! How often have you heard about someone walking away from a bad contract?
Take that a step further. In an industry where profitability is predicated largely on the ability to buy and sell solar assets, what will happen if owners are faced with a cost structure that is entirely different from what they anticipated? And what if these unattractive O&M economics reduce or eliminate the buyer pool?
I can’t help but think of a friend’s recent experience with a vacation club membership. He wanted out, but was locked into the club until he found someone to buy his shares. In the interim, he was on the hook to pay the escalating dues. He was stuck between not wanting to take a loss on his initial investment and continuing to spend more and more on a membership he did not want as the dues continued to mount. Imagine having to pay for something you no longer want, but have no choice perpetually until you find a buyer. Ultimately, this friend paid someone to take over this vacation club membership just to get out from under the dues. If we’re not careful, solar asset owners could find themselves in a similar bind which would be a further detriment to the industry’s growth.
The opportunity to optimize O&M costs
I live by the motto “show me something that hasn’t changed in six months, and I’ll show you something that can be improved.”
The standing line in the solar industry is that there is nothing that can be done to lower O&M costs or accurately model them for the long term. Nothing could be further from the truth! In fact, when it comes to O&M, solar is still in the dark ages.
As with most challenges of this nature, the answer lies in recognizing the problem, quantifying it and doing something about it.
Internet of Things (IoT) technology and a new era of data analytics can be a mitigating element of the solution. By simply adding inexpensive sensors in the field to collect real-time data, and then having tools to analyze that data to inform O&M, we change the equation.
Like any industry, O&M costs become predictable by considering the sum of the mean time to failure (MTTF) of the components times their quantity. A very common term employed by the technology industry is “RAS” or Reliability, Availability, Serviceability. Through the RAS characterization process, every system is dissected down to its individual components and a meantime to failure analysis is developed for each to determine life span, early life failure projections, and this analysis provides informed O&M protocols.
Shame on solar if we don’t follow their lead. There is no excuse for not having a thorough understanding of how every component we employ on a project will perform over 20 years or more. It should impact our investment decisions and influence how we can optimally maintain our assets.
The insurance industry knows statistically how long you will live, BMW knows how long its cars will last. They don’t put their finger in the air or consider everything equal. They are constantly collecting data and updating their models. It’s time for solar to do the same! Let’s avoid “big short” misfortunes in the solar industry.
Michael Maulick is a tech-industry veteran and currently the CEO of energy solutions company SunLink Corporation. His work is focused on facilitating the rapid convergence of the technology and energy sectors on an international scale, enabling more solar projects to succeed -- even in the most challenging global environments -- while also enabling vital improvements to grid security and stability.