NOV 16, 2023
New GE Digital Software Designed to Optimize Gas Turbine Performance
JAN 20, 2022
A new product from GE Digital will improve the operation of gas turbines through artificial intelligence (AI) and machine learning (ML), resulting in lower fuel consumption and reducing emissions, the company announced Jan. 20.
GE Digital in its announcement said turbines “require seasonal adjustment of flame temperatures and fuel splits, which is generally a manual process performed by an expert after an outage and may take a few days to complete. However, manual seasonal tuning is only efficient for the precise conditions in which it was completed and does not respond to changes in ambient temperature or fuel properties.”
GE Digital said its Autonomous Tuning software uses AI “to build a machine learning (ML) Digital Twin model of a gas turbine to continuously find optimal flame temperatures and fuel splits to minimize emissions and acoustics. The on-premises software senses changes in ambient temperature, gas fuel properties, and degradation, and sends real-time automatic adjustments to the controls every two seconds.”
“By moving customers from optimizing their combustion systems manually, Autonomous Tuning unlocks additional emissions and fuel savings for accelerating their decarbonization goals,” Martha Saker, Digital Product Manager of GE Digital’s Power Generation and Oil & Gas business, told POWER. “What’s more, applying AI/ML eliminates frequent intervention, moving customers a step closer to realizing the autonomous plant.”
Before and after Autonomous Tuning software: Using information from the software, the CO emissions were significantly reduced. The CO profile is now controlled and fully compliant with regulations. Source: GE Digital
The company said power plants using the software will have an improved emissions profile, with reductions of 14% in carbon monoxide; a 10% to 14% reduction in nitrous oxide; and fuel and carbon dioxide reduction of between 0.5% and 1%.
“With Autonomous Tuning, GE Digital has introduced a practical industrial example of the use of machine learning in closed loop supervisory control, and all running at the edge,” said Joe Perino, Principal Analyst at LNS Research, in a news release. “This is a real-world application of AI for decarbonization with tangible reductions in emissions and fuel for gas turbine operators. This, and other building block sub-systems, are a step toward autonomous operations.”
GE Digital said the goal of its Autonomous Tuning software is to allow for tracking of what it calls a turbine’s “sweet spot,” where it operates with low acoustics and low emissions, “in response to changes in environmental conditions, fuel properties, or physical degradation, and reduce the need for seasonal remapping.” The company said the software is applicable to any original equipment manufacturer (OEM) gas turbine platform, and “also fully bound by the turbine controls system’s safety-critical programming,” to ensure it does not harm the turbine.
“Gas turbines are becoming increasingly critical as the world looks to produce energy from lower-carbon sources,” said Linda Rae, General Manager of GE Digital’s Power Generation and Oil & Gas business. “While they offer a better alternative to coal, they can be made more efficient with software. Digital solutions like Autonomous Tuning are no longer optional. The energy transition demands we employ every measure for efficiency.”
GE Digital said it expects users of the software will have a lower total cost of ownership, and more operational flexibility with their turbines. They also said improved productivity from the turbines “can result in payback in under one year.”
The company in Thursday’s announcement said it expects power generators that will most benefit from the new software “are located in highly regulated regions or with constrained emissions, such as Europe, the United States and Canada, or in any location that does not have consistent weather patterns.” GE Digital said the software specifically would help power plants “subject to fuel-quality variability issues or sites looking to reduce their Operations & Maintenance (O&M) cost by reducing manual tuning and fuel consumption … customers will enjoy full-service deployment of the on premises solution and calibration of the software to run autonomously without plant personnel intervention.”
—Darrell Proctor is a senior associate editor for POWER (@POWERmagazine).