The first installment of this 6-part series provided a broad overview of the challenges facing the American energy landscape. More specifically, rapid onboarding of distributed energy resources (DERs) like solar photovoltaic (PV) systems and wind turbines is making utility grids harder to manage and more expensive to maintain.

In addition to cost overruns, this decarbonization movement paradoxically leads to increased reliance on fossil fuel combustion – precisely at a time when the goal is to green the grid with more renewable power.

This installment explores:

  • The growing adoption of wind and solar energy 
  • The main challenges of intermittent energy generation 
  • The impact AI can have on wind and solar energy 

 The Growing Adoption of Wind and Solar Energy

There exists near universal agreement that the world must decarbonize, with solar and wind representing two of the most promising technologies to wean us off fossil fuels like oil, coal, and natural gas – i.e. the primary energy sources that currently power most of the planet.

Both solar (and to a lesser extent wind) are scalable technologies – ranging from residential rooftop PV installations to multi-acre solar farms that are capable of generating many megawatts of clean and renewable energy.

Better still, both technologies are growing increasingly affordable, with wind experiencing a 70% price drop from 2010 to 2020 and solar prices going through an even more dramatic 90% decline over that same period.

Moreover, there now exist a range of green incentives and financing vehicles designed to make distributed energy resources like solar even more affordable. And more recently, the Biden Administration signed the Inflation Reduction Act, which is the most ambitious climate change legislation in US history. Experts predict that the $370 billion allocated for green energy projects could reduce American greenhouse gas emissions by up to 43% by 2030 (compared to 2005 levels).

In addition to carbon reductions, private investments into green technologies also deliver substantial financial savings to system owners, with most residential solar PV installations paying for themselves in under 10 years due to utility cost avoidance.

Even larger savings await truly energy-hungry organizations that invest in solar and wind farms at scale. This explains why companies like Apple, Google, and Amazon  have all announced multi-megawatt renewable energy plants to power more of their operations.

At the micro level, going green is a no-brainer. Once installed, for example, a rooftop PV installation instantly delivers carbon offsets and monthly savings. Moreover, the system owners enjoy peace of mind in knowing that their investments are helping to protect the planet.

At the macro level, however, these benefits are often slow to emerge. And in fact, widespread investment in renewable energy technologies sometimes leads to higher costs, more CO2 emissions, and a less uncertain future for the planet.

The Main Challenges of Intermittent Energy Generation

Although solar and wind represent some of the best tools in our fight against climate change, both technologies have struggled with reliability and profitability. In addition to facing sudden surges and dips in power production, the main challenge is that neither wind nor solar are consistent sources of energy: 

In all of these cases, grid operators must scramble to smooth out energy peaks and valleys, either by:

  • Bringing on additional load power as quickly as possible to cover unexpected shortfalls. Unfortunately, this strategy requires using energy sources with fast response times like fossil fuel, which defeats the whole purpose of using renewable power.
  • Managing excess power whenever renewable energy assets push supply above demand. Grid operators normally accomplish this by selling extra electricity to neighboring utility markets (at a loss) or simply dumping this unused electricity (also at a loss).

One increasingly popular option involves using on-site battery energy storage system (BESS) technology to capture excess renewable power when it’s abundant and discharges this stored power whenever shortfalls exist within the electricity grid.

When done correctly, batteries can help smooth out peaks and valleys, leading to greener, cheaper, and more reliable electricity delivery. But as promising as battery storage technology is, the path to a greener grid is still beset with major challenges. Chief among these is the real-time coordination needed to ensure that all grid-connected DERs work together in concert:

  • Many privately owned distributed energy resources operate in black-box environments, outside the direct control of those tasked with managing the grid. 

When a user’s solar PV and battery system lacks sensors, for example, it’s unable to share production and storage data with grid operators. And in the absence of receivers, that same renewable energy system is unable to execute instructions sent by utilities.

  • Even when DERs are equipped with sensor and receiver technology, the aggregate data they generate and send every second is overwhelming. It simply isn’t possible to parse through the many terabytes of real-time production, storage, and weather data spread across the millions of privately owned wind turbines, PV panels, electric vehicle charging stations, and battery systems that might exist within a single utility market.
  • Lastly, lags exist throughout the system due to slow human response times not to mention the time required to onboard additional load power whenever battery storage proves insufficient. These lags lead to inevitable shortages throughout the grid and are often due to something as innocuous as a cloud that temporarily blocks out the sun.

However, artificial intelligence (AI) and iterative machine learning (ML) are proving incredibly effective at turning this deluge of real-time data into actionable results.

The Impact AI Can Have on Wind and Solar Energy

The first installment in this series already explored the most direct and impactful way that AI can overcome renewable power intermittency. In short, AI can use historic data in real-time to generate accurate forecasts about future production, consumption, storage, and weather patents. AI can then use these predictions to find the optimal balance among all grid-connected energy sources. Through iterative ML, AI can then match these forecasts against actual results to refine its prediction models and become even more accurate over time.

Better still, reaction times are almost instantaneous, leading to minimal disruptions and optimal delivery. What results is the greenest, cheapest, and most reliable electricity possible at any given moment – thanks to real-time, grid-wide coordination powered by AI.

However, the benefits of the technology aren’t limited to this use case. AI and ML can also help us extract more overall value from green investments by:

1. Analyzing Weather Patterns 

By analyzing wind data, turbines can be repositioned to optimize energy production based on real-time weather models at any given location. Alternatively, AI can predict solar radiation levels minutes, hours, and days into the future, using this information to autonomously adjust energy production and storage schedules accordingly.

2. Extending Device Lifetimes

Reducing surges and shortages can help extend the lifetimes of grid-connected DERs,  many of which aren’t designed to withstand sudden shocks to their internal circuitry. However, modern technology can also be used to detect potential malfunctions before they mushroom into even larger problems. For example, many utilities are now turning to AI, autonomous drones, and infrared imaging to scan large solar PV plants for hotspots and panel defects.

3. Reducing the Cost of Green Investments

The same drone imagery used to identify solar panel defects can also be used to analyze potential sites for future renewable energy projects. When combined with AI, this imaging allows machines to generate multiple configurations for a given solar PV installation to identify which design will allow the system owner to produce the most energy at the lowest cost.

Conclusion

Nearly every discipline plagued by large data sets is increasingly turning to the power of artificial intelligence and machine learning. And we’re excited to see this technology have such a positive impact on the future of renewable energy. AI allows us to continue improving the safety, profitability, and reliability of solar and wind projects at scale.

And this is only the beginning.

The next installment in this 6-part series will explore how artificial intelligence is helping to make us more energy efficient, leading to increased savings, decreased waste, and a cleaner planet.

If you would like to learn more about how our own AI is helping to make the grid greener and more reliable, request a free demo from us today.