PEM Fuel Cell Temperature Control Under Sensor Faults

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing temperature control methods for proton exchange membrane fuel cells lack robustness and accuracy due to the failure to account for system faults, leading to performance issues and inefficiencies.

Innovation Solution

An active fault-tolerant temperature control method using sliding-mode-based control and Dulmage-Mendelsohn decomposition to diagnose sensor faults and maintain robust temperature control, incorporating a structural analysis and semi-empirical models for the fuel cell and auxiliary systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If model-based fault diagnosis is used to improve robustness, then reliability is improved, but device complexity increases due to requiring accurate mathematical models

Engineering Contradiction:
ImproverobustnessVSAvoidmodel complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system is divided into two independent diagnosis modules: data-based fault diagnosis using neural networks for initial fault detection, and model-based fault diagnosis using mathematical models for precise fault isolation. This segmentation allows each module to specialize in specific tasks, improving overall reliability while managing complexity through functional decomposition.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The data-based fault diagnosis module performs preliminary fault detection before the model-based module conducts detailed analysis. This preliminary action filters obvious faults early in the process, reducing the computational burden on the more complex model-based module and improving overall system efficiency.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If data-based fault diagnosis is used to simplify implementation, then ease of operation is improved, but robustness deteriorates due to lack of robustness

Engineering Contradiction:
Improveimplementation easeVSAvoidrobustness
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent merges data-based fault diagnosis (using neural networks for pattern recognition) with model-based fault diagnosis (using mathematical models for physical constraint validation). This combination leverages the implementation ease of data-based methods while compensating for their lack of robustness through the added reliability of model-based validation.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system implements a feedback mechanism where the results from data-based diagnosis are fed into the model-based diagnosis module for verification. This feedback loop ensures that initial diagnoses are validated against physical models, improving robustness while maintaining the ease of operation provided by the data-driven approach.

Inventive Principle:
Principle #23Feedback

3Device complexity

If existing temperature control methods are used to maintain simplicity, then device complexity is reduced, but control precision deteriorates due to failure to account for system faults

Engineering Contradiction:
Improvecontrol system complexityVSAvoidcontrol precision
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The control system dynamically adjusts its behavior based on fault status. When faults are detected by the dual diagnosis modules, the controller automatically modifies control parameters and strategies to compensate for the faults. This dynamic adaptation maintains control precision without requiring a completely complex redesign of the control architecture.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes control parameters dynamically based on detected faults. Different fault conditions trigger different parameter adjustments in the temperature control algorithm, allowing the system to maintain precision under various fault conditions while keeping the base control structure relatively simple.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11916270B2Active fault-tolerant temperature control method for proton exchange membrane fuel cell system
Publication Date: 2024.02.27 ZHEJIANG UNIV
  • US11916270B2 patent drawing
  • US11916270B2 patent drawing
  • US11916270B2 patent drawing

AI summary

An active fault-tolerant temperature control method for a proton exchange membrane fuel cell system is disclosed. Firstly, a model of a proton exchange membrane fuel cell temperature control system is established, and a system structure matrix is established according to the model by structural analysis. The system structure matrix is decomposed by using a Dulmage-Mendelsohn method, a redundant part of the model is obtained, and a system residual is constructed to reflect faults of the temperature control system. On the basis of fault identification, sliding-mode-based active fault-tolerant control is designed to accurately control an outlet temperature of a stack. The new method solves the problem of sensor failure of a temperature control model of the proton exchange membrane fuel cell system during operation, and applies model-based fault-tolerant control to temperature control, so that the reliability and durability of the fuel cell system can be effectively improved.