Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Fault-Tolerant Temperature Control of HRSG Superheaters: Stability Analysis Under Valve Leakage Using Physics-Informed Neural Networks

Mojtaba Fanoodi, Farzaneh Abdollahi, Mahdi Aliyari Shoorehdeli, Mohsen Maboodi

TL;DR

To address disturbances from valve leakage in HRSG temperature control, the paper proposes a fault-tolerant framework that integrates PI plus feedforward control with a Physics-Informed Neural Network to adapt gains in real time. A NARX-enhanced physics-based model captures nonlinear desuperheater dynamics; a Lyapunov-based analysis proves asymptotic convergence under bounded leakage. Simulations and field tests at the Pareh-Sar plant show substantial improvements in response time, reduced temperature deviations, and robust fault tolerance, achieving smoother gain transitions and potential increases in power output. The approach blends physics-based modeling with data-driven adaptation to enable reliable, autonomous HRSG operation under faults.

Abstract

Faults and operational disturbances in Heat Recovery Steam Generators (HRSGs), such as valve leakage, present significant challenges, disrupting steam temperature regulation and potentially causing efficiency losses, safety risks, and unit shutdowns. Traditional PI controllers often struggle due to inherent system delays, nonlinear dynamics, and static gain limitations. This paper introduces a fault-tolerant temperature control framework by integrating a PI plus feedforward control strategy with Physics-Informed Neural Networks (PINNs). The feedforward component anticipates disturbances, preemptively adjusting control actions, while the PINN adaptively tunes control gains in real-time, embedding thermodynamic constraints to manage varying operating conditions and valve leakage faults. A Lyapunov-based stability analysis confirms the asymptotic convergence of temperature tracking errors under bounded leakage conditions. Simulation results using operational data from the Pareh-Sar combined cycle power plant demonstrate significantly improved response times, reduced temperature deviations, enhanced fault resilience, and smooth gain adjustments. The proposed adaptive, data-driven methodology shows strong potential for industrial deployment, ensuring reliable operation, autonomous fault recovery, and enhanced performance in HRSG systems.

Fault-Tolerant Temperature Control of HRSG Superheaters: Stability Analysis Under Valve Leakage Using Physics-Informed Neural Networks

TL;DR

To address disturbances from valve leakage in HRSG temperature control, the paper proposes a fault-tolerant framework that integrates PI plus feedforward control with a Physics-Informed Neural Network to adapt gains in real time. A NARX-enhanced physics-based model captures nonlinear desuperheater dynamics; a Lyapunov-based analysis proves asymptotic convergence under bounded leakage. Simulations and field tests at the Pareh-Sar plant show substantial improvements in response time, reduced temperature deviations, and robust fault tolerance, achieving smoother gain transitions and potential increases in power output. The approach blends physics-based modeling with data-driven adaptation to enable reliable, autonomous HRSG operation under faults.

Abstract

Faults and operational disturbances in Heat Recovery Steam Generators (HRSGs), such as valve leakage, present significant challenges, disrupting steam temperature regulation and potentially causing efficiency losses, safety risks, and unit shutdowns. Traditional PI controllers often struggle due to inherent system delays, nonlinear dynamics, and static gain limitations. This paper introduces a fault-tolerant temperature control framework by integrating a PI plus feedforward control strategy with Physics-Informed Neural Networks (PINNs). The feedforward component anticipates disturbances, preemptively adjusting control actions, while the PINN adaptively tunes control gains in real-time, embedding thermodynamic constraints to manage varying operating conditions and valve leakage faults. A Lyapunov-based stability analysis confirms the asymptotic convergence of temperature tracking errors under bounded leakage conditions. Simulation results using operational data from the Pareh-Sar combined cycle power plant demonstrate significantly improved response times, reduced temperature deviations, enhanced fault resilience, and smooth gain adjustments. The proposed adaptive, data-driven methodology shows strong potential for industrial deployment, ensuring reliable operation, autonomous fault recovery, and enhanced performance in HRSG systems.

Paper Structure

This paper contains 25 sections, 1 theorem, 33 equations, 16 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Superheater and Desuperheater Model Development
  3. Superheater
  4. De-superheater
  5. NARX Model Structure
  6. Control Strategy
  7. Extended Model with Valve Leakage Fault
  8. Time Variant Gains
  9. Development of a Physics-Informed Neural Network (PINN)
  10. Introduction to Physics-Informed Neural Networks
  11. Design and Architecture of the PINN
  12. Physics-Informed Neural Network (PINN) Training
  13. Stability Analysis
  14. Simulation and Real Time Site Results
  15. Introduction
  16. ...and 10 more sections

Key Result

Theorem 1

Consider the HRSG temperature control system described by eq:statespace-fault. For outlet temperature regulation under valve leakage fault $f \leq f_{\text{max}}$: Under Assumption 1 (Lipschitz continuity) and Assumption 2 (weight error sensitivity), with $\Gamma^{-1} \succ \frac{\beta^2}{4L} I$, all closed-loop signals are uniformly ultimately bounded.

Figures (16)

  • Figure 1: Valve leakage in nozzles of superheaters and its effect in Pareh-Sar Power Plant, located in northern Iran. a: DSH nozzle taken down for maintenance. b: Cracks inside the nozzle.
  • Figure 2: DSH of the HRSG in Pareh-Sar Powerplant
  • Figure 3: Schematic illustration of the HRSG temperature-control loop highlighting the feedforward signal and location of the leakage fault in the spray-water valve.
  • Figure 4: The proposed control structure using PINN
  • Figure 5: The PINN and its Structure
  • ...and 11 more figures

Theorems & Definitions (3)

  • Remark 1
  • Theorem 1
  • proof