Sophisticated mechanical systems, including aerospace engines, possess intricate structures and operate in harsh environments with multisource disturbances, leading to rapid fault propagation, frequent false alarms, and distorted health assessments. Liquid aerospace engines, in particular, have limited operating margins and are therefore highly susceptible to failure in launch vehicles.
To address these challenges, researchers propose, in an article in IEEE Transactions on Instrumentation and Measurement, a multiparameter adaptive threshold method for real-time monitoring and robust health evaluation in the presence of intense interference.
The main contributions of this article are as follows:
- A multiparameter adaptive threshold approach that can monitor multiple engine failures simultaneously, as well as meet the real-time requirements of liquid aerospace engines.
- Alarm and output strategies for unbalanced faults, misalignment faults, and vibration-exceedance faults in liquid aerospace engines.
- The proposed method can cover the entire operating stage of liquid aerospace engines without being affected by working conditions or parameters.
The authors first introduce the frequency-domain characteristics of liquid aerospace engines and the feature extraction method, then describe in detail the proposed method, the fault alarm strategy, and the output strategy.
Multiparameter Adaptive Threshold Method
The authors present a multiparameter adaptive threshold method for real-time health monitoring of aerospace engines in the presence of heavy noise interference. By combining rapid-vibration feature extraction, an intelligent fault-alarm strategy, and fault-severity quantification, the approach enables accurate, millisecond-level health assessment.
Rotational frequency trends reveal unbalanced faults, as engine malfunctions cause rapid energy release and sharp increases in vibration amplitude. Excessive vibration is identified by amplitudes at key frequencies, including rotating, double-rotating, blade-passage, and double-blade-passage.
According to the authors, the proposed method sets three thresholds to determine the degree of malfunction. When the monitoring indicator exceeds, the proposed method outputs light faults and outputs abnormal. This algorithm simultaneously monitors multiple features and outputs the most severe fault severity. In addition, according to statistical results from historical engine test data, vibration-exceedance fault is the primary fault mode. Unbalance and misalignment faults have been eliminated in the manufacturing and assembly processes.
Algorithm flowchart
Experimental Cases Verification
Case studies on liquid aerospace engines confirm their effectiveness in identifying faults in real time under multisource noise. To assess the availability of the proposed approach, the researchers used low-temperature tests of liquid aerospace engines to validate the method. The proposed method focuses on computational efficiency and enables online health assessments, which are crucial for future reusable space engines.
Liquid aerospace engines low-temperature test bench.
This study adopted a new vibration monitoring method: using vibration signals and their characteristics instead of pulsation signals, which has two advantages: first, it can effectively characterize the health status of the turbopump; second, it reduces and avoids the complexity of the hardware system and the decrease in fault diagnosis speed caused by excessive irrelevant data under online conditions, which affects the emergency control of engine faults.
Conclusion and Future Research
In this article, researchers outline a multiparameter adaptive-threshold real-time monitoring method that enhances safety by detecting faults earlier and reducing false alarms, thereby giving the control system more time to respond effectively.
Beyond health assessment, life prediction technology can also be applied to aerospace engines to enable engine reuse. For instance, developing physics-informed data-driven methods improves prediction efficiency while reducing dependence on high-quality data. Furthermore, decision frameworks such as accelerated degradation testing and predictive maintenance can be applied to aerospace engines.
Interested in learning more about Aerospace Engines? We have thousands of articles related to these industries and more! Also learn more about our eBooks and eLearning collections on cutting edge technologies.
Interested in acquiring full-text access to this collection for your entire organization? Request a free demo and trial subscription for your organization.
