Hazleton engineering professor's solar solution cuts costs for energy provider

November 7, 2024

HAZLETON, Pa. - As an engineer, Associate Professor of Engineering Joseph Ranalli is a problem solver. When he learned a major energy provider, Southern Co., was facing a potentially costly and labor-intensive issue, he got to work figuring out how to fix it.

The resulting solution has since been put into practice by Southern Co. and served as the focus of an article Ranalli, along with co-author Principal Research Engineer Will Hobbs from Southern Co., recently published in the May 2024 issue of the Institute for Electrical and Electronics Engineers (IEEE) Journal of Photovoltaics titled "PV Plant Equipment Labels and Layouts Can Be Validated by Analyzing Cloud Motion in Existing Plant Measurements." The software that includes this solution, authored primarily by Ranalli, was covered in an article by Ranalli and Hobbs in the Journal of Open Source Software (JOSS) titled "SolarSpatialTools: A Python package for spatial solar energy analyses." These materials are all freely available - the articles are both open access, and the software is open-source.

Moreover, Ranalli said, it serves as another example of the valuable real-world industry experience Penn State Hazleton faculty share with students who might one day encounter similar opportunities to address a challenge.

"I really try to emphasize to our engineering students how much data and computers and analysis are becoming a big part of business, and I think this project is a great demonstration of that for them because it shows we didn't go to any sites or do any experiments, we looked at data that was provided to us and used that to help a company make a decision," said Ranalli, who also serves as discipline coordinator for engineering for Penn State's 14-campus University College.

Impact on industry

Ranalli and Hobbs first met at a solar conference in 2014 where Ranalli was presenting on solar site assessment tools, and Hobbs had been looking at similar techniques to improve accuracy of shade estimates. In 2022, Ranalli was presenting on solar variability and cloud motion at another conference, and Hobbs sparked a discussion with him about ways his methods could be adapted to solve labeling challenges Hobbs had run into in ongoing research at Southern Co.

"Southern Co. has hundreds of data streams coming in from every segment of the solar plant, and there are times that they believe communication wires may be crossed in the data, where what they think is coming from one section might actually be coming from somewhere else," Ranalli said.

As such, a malfunction in a section of the plant could lead to workers being sent to the wrong area, or even create an inability to determine where a remedy is needed, Ranalli explained. Addressing the issue would typically involve sending crews to the plants to manually identify and relabel affected areas, which can be a tedious and costly process.

Ranalli realized he might have a solution. While on sabbatical in Germany in 2019, he worked with researchers at the University of Oldenburg and the German Aerospace Center on several projects related to energy meteorology.

Inspired by a cloud tracking system the two organizations used to forecast where and when sunlight would be strongest, Ranalli developed a method that describes the timing of cloud shadows in different areas of a solar plant.

Ranalli offered to test out his method, and based off operational data provided by Southern Co., was able to determine when cloud coverage created a delay in signal responses at one of the company's solar facilities. By observing the delay between the motion of the clouds, this method was able to pinpoint mislabeled combiners - centralized points where wiring from multiple solar panels connect.

"In my time doing research, this is the project that has had the most direct application to a real-world problem," Ranalli said.

"The traditional approach to solving this issue involves tedious field verification, while the approach Ranalli developed could be faster, less expensive and safer," Hobbs said. "We appreciate Ranalli's collaboration with our team through the development process ... turning this work into something that is directly applicable in our industry."

Hobbs added that Southern Co. continues to work with contractors to deploy Ranalli's method at its solar facilities throughout the United States.

"To see that my research mattered to this company and helped them solve a real-world problem was very satisfying, and it's very motivating to keep working on this and hopefully find other companies that could benefit as well," Ranalli said.

Student-centered approach

Although Ranalli is thankful for the recognition from Southern Co., he is most appreciative of the opportunity the partnership provides for students in Penn State Hazleton's Bachelor of Science in Engineering Alternative Energy and Power Generation program.

Like other Penn State Hazleton faculty, Ranalli often aims to incorporate his real-world experience in the field into his teaching and mentorship of future engineering graduates.

"As an engineer, I'm always grateful for these kinds of opportunities to use our expertise and be able to make valuable research contributions that I can in turn bring into the classroom," he said.

Though no students worked directly on this project, Ranalli regularly invites students to perform undergraduate research with him during the academic year. He discusses this project as an example of how current research trends can have a profound industry impact.

"I get to say, 'Let me tell you about this awesome thing I've been working on,'" Ranalli said. "It helps to bring excitement into the classroom."

Ranalli discussed his project and its results in detail during a presentation at the IEEE Photovoltaic Specialists Conference back in June. He presented alongside Hobbs, who co-authored the published article.