Abstract:Condition monitoring and diagnostics of a combined cycle gas turbine power plant has become an important tool to improve its availability, reliability, and performance. However, there are two major challenges in the diagnostics of performance degradation and anomaly in a single shaft combined cycle power plant. First, since the gas turbine and steam turbine in such a plant share a common generator, each turbine's contribution to the total plant power output is not directly measured, but must be accurately estimated to identify the possible causes of plant level degradation. Second, multivariate operational data instrumented from a power plant need to be used in the plant model calibration, power splitting and degradation diagnostics. Sensor data always contains some degree of uncertainty. This adds to the difficulty of both estimation of gas turbine to steam turbine power split and degradation diagnostics. This paper presents an integrated probabilistic methodology for accurate power splitting and the degradation diagnostics of a single shaft combined cycle plant, accounting for uncertainties in the measured data. The method integrates the Bayesian inference approach, thermodynamic physics modeling, and sensed operational data seamlessly. The physics-based thermodynamic heat balance model is first established to model the power plant components and their thermodynamic relationships. The model is calibrated to model the plant performance at the design conditions of its main components. The calibrated model is then employed to simulate the plant performance at various operating conditions. A Bayesian inference method is next developed to determine the power split between the gas turbine and the steam turbine by comparing the measured and expected power outputs at different operation conditions, considering uncertainties in multiple measured variables. The calibrated model and calculated power split are further applied to pinpoint the possible causes at individual components resulting in the plant level degradation. The proposed methodology is demonstrated using operational data from a real-world single shaft combined cycle power plant with a known degradation issue. This study provides an effective probabilistic methodology to accurately split the power for degradation diagnostics of a single shaft combined cycle plant, addressing the uncertainties in multiple measured variables.

Bayesian Calibration of Performance Degradation in a Gas Turbine-Driven Compressor Unit for Prognosis Health Management

Implementation of Condition-based Maintenance Decision-making Model and System for Heavy-duty Gas Turbine

Real-time Novelty Detection of an Industrial Gas Turbine Using Performance Deviation Model and Extreme Function Theory

Prognostics of Gas Turbine: A Condition-Based Maintenance Approach Based on Multi-Environmental Time Similarity

Multidimensional analysis and performance prediction of heavy-duty gas turbine based on actual operational data

Towards Predictive Maintenance of a Heavy-Duty Gas Turbine A New Hybrid Intelligent Methodology for Performance Simulation

Bayesian Calibration For Power Splitting In Single Shaft Combined Cycle Plant Diagnostics

Research on Performance Degradation Prediction Method of Heavy-Duty Gas Turbine Based on Data-Physics Fusion under Uncertainty

A Novel Data-Driven Approach for Predicting the Performance Degradation of a Gas Turbine

A sequential model-based approach for gas turbine performance diagnostics

Simulation Model Uncertainty Quantification and Model Calibration for Natural Gas Compressor Units

Physics Informed Machine Learning Approach for Performance Degradation Monitoring of Gas Turbine

A Generalized Remaining Useful Life Prediction Method for Complex Systems Based on Composite Health Indicator.

Estimating Recoverable Performance Degradation Rates and Optimizing Maintenance Scheduling

Preventative Maintenance Decision System For Gas Turbine Based On Operation Status Information

Enhancing Turbine Performance Degradation Prediction with Atmospheric Factors

Performance-Based Gas Turbine Health Monitoring, Diagnostics, and Prognostics: A Survey

A Deterministic Calibration Method for the Thermodynamic Model of Gas Turbines

Performance-based health monitoring, diagnostics and prognostics for condition-based maintenance of gas turbines: A review

Application of A State Space Modeling Technique to System Prognostics Based on A Health Index for Condition-Based Maintenance

A Bayesian inference-based approach for performance prognostics towards uncertainty quantification and its applications on the marine diesel engine