Fuel Cell Fault Detection via Voltage PSD Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting faults and degradation in fuel cell assemblies are complex, costly, and not entirely reliable, often requiring active interaction with the system and external signal injection.
Innovation Solution
A method that analyzes the power spectrum density (PSD) of the voltage signal output by fuel cell assemblies during steady-state operations, using de-trended voltage segments and test statistics to identify faults or degradation without the need for external signals, thereby reducing system complexity and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If active signal injection methods are used for fault detection, then detection reliability is improved, but device complexity and manufacturing costs increase
Solution Approach 1:
The fuel cell assembly itself serves as the signal source by utilizing its natural voltage fluctuations during steady-state operation. The method extracts fault detection information from the assembly's own operational characteristics without requiring external signal injection or additional active components, thereby maintaining high detection reliability while reducing system complexity and manufacturing costs
2Measurement precision
If external signal injection is used for diagnosis, then detection accuracy is improved, but manufacturing costs increase
Solution Approach 1:
The method utilizes the fuel cell assembly's inherent voltage signal during normal operation as the diagnostic source. By applying spectral analysis techniques to this naturally occurring signal, the system achieves accurate fault detection without requiring external signal generators, additional sensors, or complex diagnostic equipment, thereby reducing manufacturing costs while maintaining detection accuracy
Solution Approach 2:
The method replaces physical signal injection hardware with computational signal processing. Instead of using external electrical signals to probe the system, the approach uses mathematical spectral analysis (Fast Fourier Transform) to extract fault information from the existing voltage signal, substituting mechanical/electrical diagnostic equipment with software-based analysis
3Reliability
If complex diagnostic systems are implemented, then detection reliability is improved, but maintenance costs increase
Solution Approach 1:
The diagnostic system monitors the fuel cell assembly's own operational signal without requiring external diagnostic equipment. The voltage signal naturally produced during steady-state operation contains fault information that can be extracted through spectral analysis, eliminating the need for complex external monitoring systems and reducing ongoing maintenance requirements
Solution Approach 2:
The method creates a spectral representation (power spectrum density) of the voltage signal that serves as a diagnostic copy. This spectral copy contains fault information in a more analyzable form, allowing reliable fault detection through comparison with reference spectra without requiring complex physical diagnostic equipment that would need maintenance
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method provides a reliable, efficient, and inexpensive means to monitor fuel cell assemblies, enabling effective diagnostic capabilities through passive spectral analysis, reducing manufacturing and maintenance costs while improving detection accuracy.
Implementation Method 1
one way of estimating the PSD of a signal is to simply find the Fast Fourier Transform, FFT, of the (equispaced) signal data and to appropriately scale the squared magnitude of the result
Data Source
Figure 1
Figure 2a~2b
Figure 3a~3d
AI summary
The present invention relates to a method for detect faults and/or degradation in fuel cell assemblies, analysing the power spectrum density (PSD) of the voltage signal output by a fuel cell assembly during steady-state operations. The method according to the present invention identifies specific faults and/or degradation signatures as different patterns attained by the PSD of the measured voltage signal.