The team applies the power of AI to forecast energy supply and demand

0


Researchers at the Center for Applied AI and Machine Learning at UT Dallas worked with Vistra on a project to help predict prices for its Moss Landing energy storage facility, where these battery modules are part of a solar and wind energy storage system. The aim of the project is to help Vistra buy electricity to recharge batteries at the lowest price and sell at the most convenient time.

The increasing use of renewable and fluctuating energy sources, such as wind and solar power, has made it harder than ever to predict electricity prices.

In partnership with Vistra, one of the largest competitive power producers in the United States, researchers at the University of Texas at Dallas are using artificial intelligence (AI) techniques to help the Irving-based company make more precise price projections.

Research for Vistra is the inaugural project of the University Center for Applied AI and Machine Learning (CAIML) in the Erik Jonsson School of Engineering and Computer Science. The center, established in 2019 to work with industry partners to apply advanced research in AI and machine learning to solve practical problems, will be located in the Richardson Innovation Quarter.

In machine learning, computers analyze large amounts of data in order to learn to recognize patterns.

“Machine learning and deep learning technologies provide valuable information that can help an energy supplier make more accurate predictions that can save the business and its customers money,” said declared Dr Gopal Gupta, co-director of the center, professor of computing and President Erik Jonsson. “Our partnership with Vistra is an example of how universities can work with business to solve real world problems. “

The researchers completed the work in August 2020 and held courses from December to February to train Vistra employees in basic technologies.

“Machine learning and deep learning technologies provide valuable information that can help an energy supplier make more accurate predictions that can save the company and its customers money. Our partnership with Vistra is an example of how universities can work with business to solve real world problems. “

Dr Gopal Gupta, Professor of Computer Science and Erik Jonsson Chair at the Erik Jonsson School of Engineering and Computer Science

The project, which was funded by Vistra, will help the company set prices for its Moss Landing energy storage facility in Monterey County, California. The future 400 megawatt energy storage system, which is one of the largest of its kind in the world, can power up to 300,000 homes during peak periods. Such battery storage facilities offer the possibility of storing energy from renewable sources, such as wind and solar power, for later use.

“This project, carried out jointly by the Center for Applied AI at UT Dallas and Vistra Corporation, was crucial to optimizing electricity prices as our Moss Landing battery bank came on stream in early 2021,” said Rachit Gupta, vice president of Vistra and main sponsor of the project. “The project was a huge success and we are extremely happy to have been able to benefit from a great source of expertise present locally.

Wholesale energy prices are constantly fluctuating based on supply and demand. For example, extreme weather conditions, such as the winter storms that Texas experienced in February, can cause demand and costs to skyrocket. Weather conditions also affect supply – solar panels produce less electricity on cloudy days and nothing at night, while wind farms often produce most of their energy at night, when demand for electricity is lowest. .

Vistra’s Moss Landing energy storage facility in Monterey County, Calif., Stores energy from renewable sources, such as wind and solar power, for later use.

“AI can help a company like Vistra forecast future generation and demand for load, wind and solar power, and optimize bidding, planning and energy deployment for improve profitability and market participation, ”said Dr Feng Chen, Associate Professor of Computer Science at UT Dallas and Principal Investigator of the project.

Researchers applied statistical and machine learning methods to create models capable of predicting buying and selling prices in near real time. The goal is to help the company buy electricity to recharge batteries at the lowest price and sell when the most convenient.

In addition to Gupta and Chen, the project involved Dr Doug DeGroot, director of CAIML and professor of computer science education. Other team members included Farhad Shakerin MS’16, PhD’20, now a senior software engineer at Microsoft Corp .; Abhijit Balaji, a graduate student in computer science; and PhD students in computer science Yibo Hu MS’17 and Changbin Li. Dr Anurag Nagar, teaching assistant professor, and Ashutosh Senapati, master’s student in computer science, led the training.

DeGroot said members of the center are proud to have been selected for the project and look forward to teaming up with Vistra and other industry partners in the future.

“Vistra’s innovative Moss Landing battery energy storage facility will play a critical role in supporting California’s energy grid,” said DeGroot. “Our machine learning technologies to capture and deliver this energy have helped optimize the way this battery system is used. “

AI Center helps InfoVision develop drone-supported inventory system

When InfoVision envisioned a drone-based system for tasks like inventory management, the Richardson, Texas-based tech company turned to a new center at the University of Texas at Dallas that helps industry partners to apply artificial intelligence (AI) to cutting edge scenarios.

Researchers from Center for Applied AI and Machine Learning in the Erik Jonsson School of Engineering and Computer Science co-created a 3D software module with InfoVision that allows drones to navigate autonomously in a warehouse and identify products. It is an emerging method of inventory management that can reduce operational costs and complexity and dramatically increase efficiency.

“This is an opportunity for the University and an industry partner to work hand in hand on applied AI problems by matching theory and algorithm design with real world applications,” said Dr Ovidiu Daescu, Professor of computing, former site director of the National Science Foundation (NSF) Collaborative Industry-University Research Centers Program, and principal investigator of the project.

The center opened in 2020 to research, train and apply the latest AI technologies to make real-world inroads with industry partners.

“InfoVision believes that university partnerships bring expert knowledge and new perspectives, and given our physical proximity to UT Dallas, this joint project is a no-brainer – a win-win for both,” said Sean Yalamanchi, president from InfoVision.

Chithrai Mani, chief technology and innovation officer at InfoVision, said the company worked with researchers at UT Dallas to benefit from the centre’s deep AI expertise and knowledge of differentiated AI applications.

“Working with a team of computer science researchers from the University gives us the advantage of applying the latest research to overcome the limitations of traditional GPS-based autonomous drones to create a solution that can be exploited for a variety of applications,” said Mani said.

Dr Douglas DeGroot, director of the center and professor of computer science education, and Dr Gopal Gupta, co-director of the center, professor of computer science and Erik Jonsson chair, are the co-principal investigators of the project, which began in January. The team is complemented by Ankur Yadav, a PhD student in computer science working on the project under the supervision of Daescu on the design and implementation of algorithms.


Share.

About Author

Leave A Reply