Increasing Solar Panel Energy Production Using AI


The comedian George Carlin once observed that “dusting is a good example of the futility of trying to put things right. As soon as you dust, the fact of your next dusting has already been established.” 

Though the fact of our next dusting might be inevitable, at least we get to choose when we get around to doing it. In most cases, it’s a matter of convenience. But when it comes to large-scale solar panel operators, determining the optimal schedule for cleaning their units holds much higher stakes. 

Dusty, dirty solar panels produce less electricity, curbing revenue. Furthermore, soiling—the natural accumulation of dust, pollen, dirt, algae, bird droppings, and any other form of debris on solar panels—can create hot spots that eventually damage sensitive components, shortening their lifespan and reducing return on investment. One recent study estimated the global cost of soiling might rise to as much as $6.4 billion by 2023

While the logical response to soiling on solar panels is to clean them regularly, the billion-dollar question is: on what timeframe? Large-scale solar panel units are frequently hard to access and expensive to clean. Deciding exactly when and how often to clean them is a balancing act: operators don’t want to clean too often, which wastes maintenance resources, nor do they want to go too long between cleanings, limiting energy efficiency. Further complicating matters, the traditional method of monitoring solar panel soiling with soiling stations to determine a proper cleaning schedule is not only costly but often excessive or even ineffective as a mitigation strategy. 

The fast-growing solar industry needs a new paradigm for mitigating the damaging effects of soiling. SparkCognition enables solar power providers to increase energy production and reduce operational costs with our AI-powered asset management platform that removes soiling stations from the equation. Our solution provides greater asset visibility and actionable insights to discover the optimal cleaning schedule based on all contributory data sources, combined with the actual cost for cleaning the panels.

Putting the problem of solar panel soiling into perspective

As the world acts with urgency to reduce the impact of climate change by, among other efforts, transitioning from fossil-fuel power generation to renewable energy power generation, the solar industry is scaling at a record pace. For example, the U.S. solar industry grew at an average annual rate of 42% in the last decade. This steep uptrend comes from more than only increased demand. Solar panel photovoltaic (PV) technology has steadily improved over time, making solar energy more economically viable.

Yet, with more solar panel power operations deployed at a faster rate than ever, remember that many of the best places to harvest solar energy are in arid and remote areas. These are typically sun-baked environments, where the hot wind carries and deposits local contaminants and dust particles constantly. 

While rain and wind may naturally remove some amount of dust from PV panels, weather events are more of a contributing factor in measuring the overall soiling accumulation rather than a reliable cleaning method by themselves. Also, PV panels cool down at night and collect morning dew. The dust on panels mixes with this moisture resulting in a process called cementation. This caked-on dust is much harder to remove, even by an intense rain shower.

Scientists studying the soiling problem for solar panels estimate that energy loss annually can exceed as much as 7% in parts of the United States. In the Middle East and parts of Asia, energy output reductions resulting from soiling can reach up to 50% per year. Although each site location requires a site-specific analysis, soiling is clearly a persistent challenge for solar panel production and profitability, which works against the global effort to mainstream renewable energy on the fastest timeline possible.

Soiling stations vs. AI-powered soiling mitigation solutions

Operators have commonly relied on soiling stations to measure the degree of soiling occurring on their panels. In this method, sensors installed on a PV panel are regularly cleaned, while ‘control’ panels are allowed to become soiled. Comparing the soil build-up on the two panels yields information on the level and frequency of cleaning intervention likely to be needed.

The problem with the soiling station approach is twofold. First, the soiling stations add unnecessary costs to the project. Second, different parts of a large-scale solar panel array collect soil at different rates. Soiling stations don’t always account for this variance well. 

As a result, the analysis produced by soiling station data runs the risk of over or underestimating actual levels of soiling. And unless the soiling mitigation strategy accounts for expected and actual precipitation, an unnecessary cleaning would only add cost but not increase production.

There’s a better way to do this, using AI—no soiling stations required. SparkCognition’s EnsembleTM platform is the answer for solar panel operators looking to optimize the vast streams of data flowing from their assets and quantify the effect of energy losses from soiling against the actual cost of cleaning their panels more, less, or differently. 

By continuously processing the SCADA data already available from their assets in the field, historical weather data and forecasting, and project-specific criteria like PPA details, capacity limits, and curtailment factors, SparkCognition’s Ensemble platform leverages the power of AI and machine learning to optimize maintenance tasks, like cleaning solar panels, for maximum ROI.

For example, using AI and machine learning to interpret weather forecast data, operators can better determine when a precipitation event of sufficient magnitude to clean the panels will occur. Thus, all contributory data sources combine with the actual cost for cleaning the panels at the project. Ensemble provides the project operator with the optimal time to perform a panel cleaning for maximum project profitability.

AI-powered asset management is the new paradigm for solar panel soiling mitigation

SparkCognition’s AI-powered asset management platform for clean energy allows solar power providers to maximize production improvement at the lowest cost with an optimal soiling mitigation strategy to effectively help restore photovoltaic performance and extend asset lifespan. It’s the future of solar panel soiling mitigation, enabling the solar industry to scale up and meet the challenge of transitioning our world to renewable energy with speed and efficiency.


Latest blogs

Abstract depiction of our Generative AI Platform created by DALL-E
Campbell LeFlore

What’s Inside our Generative AI Platform?

Perfected over ten years of real-world engagements serving many of the world’s largest brands in energy, manufacturing, transportation, utilities, financial services, and other industries, SparkCognition’s

Read More
Campbell LeFlore

The Top Challenges HSE Managers Face Today

HSE managers are the behind-the-scenes heroes enabling today’s high-throughput industrial environments to operate at peak performance. Their determined efforts ensure the well-being and protection of

Read More
SparkCognition is committed to compliance with applicable privacy laws, including GDPR, and we provide related assurances in our contractual commitments. Click here to review our Cookie & Privacy Policy.