Domain-Informed Operation Excellence of Gas Turbine System with Machine Learning

2025-07-11

AI Summary

To address the physical infeasibility and poor generalizability of optimization solutions arising from insufficient integration of domain knowledge with AI’s black-box nature in gas turbine systems, this paper proposes a domain-aware, data-driven optimization framework—Mahalanobis Distance–Constrained Optimization (MAD-OPT). MAD-OPT innovatively encodes thermodynamic principles and operational constraints as Mahalanobis distance metrics, embedding them into both machine learning modeling and Monte Carlo simulation workflows to ensure physically realizable solutions and robustness beyond nominal operating conditions. Validated on a 395 MW industrial gas turbine unit, the method achieves significant improvements in thermal efficiency and reductions in heat rate. Optimized results align closely with plant-measured data, exhibiting an average absolute error of less than 1.2%, thereby demonstrating strong engineering feasibility and broad applicability across similar power generation systems.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Optimization
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

The domain-consistent adoption of artificial intelligence (AI) remains low in thermal power plants due to the black-box nature of AI algorithms and low representation of domain knowledge in conventional data-centric analytics. In this paper, we develop a MAhalanobis Distance-based OPTimization (MAD-OPT) framework that incorporates the Mahalanobis distance-based constraint to introduce domain knowledge into data-centric analytics. The developed MAD-OPT framework is applied to maximize thermal efficiency and minimize turbine heat rate for a 395 MW capacity gas turbine system. We demonstrate that the MAD-OPT framework can estimate domain-informed optimal process conditions under different ambient conditions, and the optimal solutions are found to be robust as evaluated by Monte Carlo simulations. We also apply the MAD-OPT framework to estimate optimal process conditions beyond the design power generation limit of the gas turbine system, and have found comparable results with the actual data of the power plant. We demonstrate that implementing data-centric optimization analytics without incorporating domain-informed constraints may provide ineffective solutions that may not be implementable in the real operation of the gas turbine system. This research advances the integration of the data-driven domain knowledge into machine learning-powered analytics that enhances the domain-informed operation excellence and paves the way for safe AI adoption in thermal power systems.

Problem

• Incorporating domain knowledge into AI for gas turbine optimization
• Maximizing thermal efficiency while minimizing turbine heat rate
• Ensuring robust and implementable AI solutions in power plants

Innovation

• Mahalanobis distance-based constraint integration
• Domain-informed optimal process condition estimation
• Data-driven domain knowledge in ML analytics