Gas Turbine Fuel Leak Detection via Node Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Gas turbine engines face challenges in detecting and isolating fuel leakage conditions during extended operations, which can lead to fuel loss and operational inefficiencies, as existing systems struggle to accurately diagnose and respond to abnormalities in the fuel system.
Innovation Solution
A propulsion system with an estimation module that calculates expected conditions based on control signals and sensor data, a comparison module that identifies abnormalities, and a detection module that isolates issues by comparing observed conditions with expected models, allowing for alerts and adjustments to flight plans or engine operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fuel leakage detection systems are implemented in gas turbine engines, then reliability is improved, but device complexity increases
Solution Approach 1:
The fuel system is divided into multiple localized nodes (fuel pump, actuator, metering valve, distribution valve, fuel injectors, combustor section) with separate sensors and detection zones. This segmentation allows the system to isolate and identify specific leakage locations without requiring a completely complex system-wide monitoring approach, thereby improving reliability while managing complexity through modular detection zones.
Solution Approach 2:
The system continuously monitors fuel flow parameters at multiple nodes and compares observed values against expected conditions. When discrepancies indicate leakage, the system provides feedback signals to alert operators and can automatically adjust fuel flow commands. This feedback mechanism improves detection reliability while using computational algorithms rather than additional physical components to manage system complexity.
2Measurement precision
If multiple localized nodes with sensors are deployed, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The same type of sensor (fuel flow sensor) is used at all localized nodes, and the same detection algorithm is applied to analyze data from any node. This universality simplifies the sensor network by using standardized components and processing logic, while still achieving high measurement precision through multi-point monitoring. The system can identify leakage location and magnitude accurately without requiring complex or specialized sensors at each node.
3Productivity
If real-time monitoring and analysis of fuel conditions is implemented, then productivity is improved, but use of energy increases
Solution Approach 1:
The system performs real-time monitoring and analysis only when necessary - specifically when fuel flow parameters deviate from expected conditions or when leakage is detected. Rather than continuously processing all data streams at full computational capacity, the system uses partial action by triggering intensive analysis only upon detecting anomalies, thereby improving productivity through timely detection while reducing overall energy consumption compared to continuous full-capacity monitoring.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A fuel system for a gas turbine engine includes a plurality of components defining a plurality of localized nodes at distinct locations relative to a fuel flow path, each of the plurality of localized nodes characterized by a distinct set of failure parameters. One or more fuel sensors are configured to measure at least one fuel condition relating to flow through the fuel flow path. A fuel observation assembly is coupled to one or more engine sensors configured to measure at least one engine condition. The fuel observation assembly comprises an estimation module operable to calculate an expected condition of each of the plurality of localized nodes based upon the set of failure parameters, an observation module operable to calculate an observed condition of each of the plurality of localized nodes, the observed condition based on present values of the at least one fuel condition and the at least one engine condition, and a comparison module operable to determine an abnormality corresponding to a first one of the plurality of localized nodes based upon a comparison of the expected condition and the observed condition for each of the plurality of localized nodes.