Fuel Cell Airflow Fault Detection for Failed MAF Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mass air flow (MAF) sensors in fuel cells are unreliable due to susceptibility to contaminants, leading to incorrect air flow and pressure readings, which can cause interruptions in power generation or permanent damage to the fuel cell, posing safety risks in aircraft.
Innovation Solution
Detect MAF sensor failure by analyzing signals from non-MAF sensors, such as compressor RPM, pressure, and temperature, using a compressor map to estimate mass air flow and initiate a safe system response, such as limp mode, to maintain safe operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If MAF sensors are used to monitor air flow and pressure in fuel cells, then the fuel cell operation can be controlled, but the sensors are susceptible to contaminants causing clogging and performance degradation
Solution Approach 1:
The patent uses an intermediary approach by introducing a machine learning model that indirectly estimates MAF sensor readings through alternative measurements from other sensors (temperature, pressure, flow sensors). This intermediary estimation system bypasses the contaminated MAF sensor while still providing the necessary air flow data for fuel cell control.
Solution Approach 2:
The patent creates a virtual copy of the MAF sensor functionality by training a machine learning model to replicate the MAF sensor's measurement capabilities using data from other sensors. This digital twin approach allows the system to continue operating with accurate air flow estimates even when the physical MAF sensor is contaminated or failed.
2Ease of operation
If MAF sensor readings are used to control IGV positions, then fuel cell reactions can be regulated, but incorrect readings lead to incorrect IGV commands and potential fuel cell damage
Solution Approach 1:
The patent implements a feedback mechanism where the machine learning model continuously monitors alternative sensor readings and adjusts the estimated MAF values in real-time. This feedback loop ensures that even when the physical MAF sensor provides incorrect readings, the control system receives accurate estimated values for proper IGV positioning and fuel cell operation.
Solution Approach 2:
The patent replaces the mechanical MAF sensor measurement system with a computational machine learning model. Instead of relying on the physical sensor to provide accurate readings, the system uses algorithms to calculate and estimate the air flow parameters, substituting mechanical measurement with intelligent computation.
3Measurement precision
If multiple sensors are used to detect MAF failure, then detection accuracy improves, but system complexity increases
Solution Approach 1:
The patent makes the existing sensors serve multiple functions: they not only monitor their primary parameters (temperature, pressure, flow) but also collectively contribute to estimating MAF sensor readings and detecting MAF failures. This multi-functionality approach allows accurate failure detection without adding dedicated MAF failure detection sensors, thereby avoiding increased system complexity.
Data Source
AI summary
A method and system of detecting mass air flow (MAF) sensor failure on an aircraft includes at least one signal from a non-MAF sensor received by a controller of a fuel cell system having at least one MAF sensor. The signal received by the controller is analyzed relative to a compressor map to estimate mass air flow. A MAF sensor failure is detected based on the estimated mass air flow. When a MAF sensor failure is detected, a safe operating mode of the fuel cell system may be activated to provide adequate power for operation of the aircraft to a safe landing while minimizing risk of damage to the fuel cell system.