Bauman Moscow State Technical University
Essay
Research of contactless methods of diagnostics of jet engines
Post-graduate
/Zaitsev/
English teacher
/Loseva /
Moscow 2023
Table of contents
Introduction ............................................................................................................. 3
Theoretical part ....................................................................................................... 4
Practical part ........................................................................................................... 9
Conclusions ............................................................................................................ 13
Introduction
One of the main directions in the development of promising rocket and jet
propulsion systems (PS) in products of rocket and space technology is associated
with the development of reliable and highly efficient methods and diagnostic tools
necessary to implement the best technical characteristics of propulsion systems at
the lowest economic costs. Traditional methods for studying the characteristics of
the working process in combustion chambers (CC) of PS are not always effective,
especially if it is necessary to diagnose physical and chemical processes in the
volume of the combustion chamber, and are practically not suitable for creating
control and emergency protection systems that instantly respond to similar changes
in the characteristics of the working process. In order to eliminate these
shortcomings, non-traditional methods for diagnosing remote control have recently
been developed, based, for example, on recording the electrophysical and
electromagnetic characteristics of the working process. A prerequisite for this
diagnostic method is that the combustion processes of the majority of fuel
compositions encountered in practice proceed at a sufficiently high temperature
(2000–4000 K). In this regard, ionization processes take place in the CS, and ions
and electrons are present in the combustion products (CP), which determines the
physical nature of the appearance of the intrinsic electromagnetic field of the CP
during combustion and outflow from jet engine nozzles. The electrons and ions
generated in the CS are involved in the turbulent motion of the main gas flow. In
this case, the separation of charges, caused by a much higher mobility of electrons
compared to ions, ensures the presence of an excess charge in the gas flow, the
magnitude and distribution of which depends on the operating parameters of the PS.
The object of the study is a model RAMJET on powdered aluminum. The
subject of the study is the electrophysical properties of the high-enthalpy flow of
aluminum combustion products in the air.
The purpose of the work is to develop non-contact electrophysical methods
for the study and diagnosis of high-enthalpy flow of combustion products and the
working process of RAMJET on powdered aluminum.
Theoretical part
Health monitoring and fault diagnosis of liquid rocket engine (LRE), which
can effectively improve the reliability and safety of launch vehicle power system,
and relevant technologies have important engineering application value. Anomaly
detection can find the defects of LRE in development in time, which provides
support for designers to locate and analyze the faults. After the flight, researchers
evaluated the state of LRE based on the anomaly detection results, and provided
direction for engine optimization and improvement in order to improve the safety
and reliability of the launch system. At the same time, it is beneficial for researchers
to improve the engine structure and performance, and further reduce the cost of
rocket launch and operation. The history of fault detection on LRE can be traced
back to Apollo Moon-landing Project in 1967. To reduce engine failures, the U.S.
Naval Research Laboratory began to research health monitoring technology of LRE.
Since the 1970s, the United States has developed various LRE fault detection and
diagnosis systems, which are widely used in the Space Shuttle Main Engine (SSME).
This mainly includes Redline System, System for Anomaly and Failure Detection
(SAFD), Health Management System for Rocket Engine (HMSRE), Integrated
Vehicle Health Management System (IVHMS), Real-time Turbine Engine
Diagnostic System (RTTEDS), Real-time Vibration Monitoring System (RTVMS),
Post-test Diagnostic System (PTDS), Health Monitoring System (HMS), Intelligent
Integrated Vehicle Management System (IIVMS), etc. In the 21st century, Stennis
Space Center, Ames Research Center, and Rocketdyne proposed Integrated System
Health Management (ISHM) for the A-1 test bed and J-2X engine. Moonis
developed an Expert System for Fault Diagnosis. Gupta designed a real-time fault
diagnosis expert system named LEADER for SSME. The fault detection approach
based on an expert system is subject to expert knowledge and is more dependent on
past experience, so it is difficult to find more faults. It is generally used for offline
analysis of engine working status.
The structure of LRE is extremely complicated, with many uncertainties in
the fault detection. On the other hand, some failures occur only once in a long period
of time, and are unlikely to reappear. At the same time, the LRE fault samples are
reduced, and those that can cover the sample model are also reduced. The research
of fault detection is faced with the problem of fewer samples.
In 1995, Vapnik proposed Support Vector Machine (SVM) based on statistical
learning theory, which is a machine learning approach specifically for finite sample
processing that transforms the original problem into a quadratic global optimization
problem with strong generalizability. The fault detection using SVM has been
widely studied in the field of fault diagnosis. Fault detection using the approach of
neural network tend to fall into local optimal values, and SVM avoids this problem
to a certain extent, so it has received certain attention and research. Aiswarya
researched the SVM fault detection algorithm based on time domain and frequency
domain features. Tao proposed an adaptive fault detection algorithm based on
wavelet transform and SVM for real-time fault detection of LRE turbopump. Tao
proposed a fault detection algorithm based on root mean square of protrusion
frequency component and SVM for real-time fault detection of LRE turbopump.
Yuan applied SVM based on boundary samples to real-time fault detection of LRE
turbopump. Heng researched fault prediction of aircraft engine based on SVM. Hu
used One-Class Support Vector Machine (OCSVM) to analyze the historical
vibration signals of turbopump, and proposed an approach that was used for
detecting and diagnosing faults online, separating sensor faults from actual faults of
the turbopump. Zhu and Zhang established LRE fault prediction model based on
SVM. Although the development of LRE fault detection using artificial intelligence
has made a lot of research achievements so far, it is still mainly affected by the scale
and quality of sensor data. The data samples of rocket engines, especially new
engines, are few, and relevant fault characteristics cannot be fully grasped. In
addition, the collection, classification, and processing of engine sensor data are not
standardized and unified, which limits the training of fault detection model.
From the development of LRE health monitoring system, it was seen that the
approach of rocket engine fault detection has shifted from single algorithm detection
to multi-algorithm fusion detection, and from traditional sensor-based diagnosis to
fault prediction based on intelligent approaches. However, most of the relevant
research is based on the classification for fault diagnosis of LRE. For a certain type
of hydrogen-oxygen rocket engine, its fault type and fault characteristics have not
been completely understood, and the fault detection approach based on classification
may not be able to detect the fault correctly. Therefore, this research proposes a
genetic algorithm-based least squares support vector regression (LSSVR) algorithm
for real-time fault detection of hydrogen-oxygen rocket engine, which provides
technical support for the engineering application of online fault detection of rocket
engine [1].
Also in this work, the effects of the electric fields on the flame propagation
and combustion characteristics of lean premixed methane–air mixtures were
experimentally investigated in a constant volume chamber. Results show that the
flame front is remarkably stretched by the applied electric field, the stretched flame
propagation velocity and the average flame propagation velocity are all accelerated
significantly as the input voltage increases. This indicates that the applied electric
field can augment the stretch in flame, and the result is more obvious for leaner
mixture. According to the analyses of the combustion pressure variation and the heat
release rate, the peak combustion pressure Pmax increases and its appearance time tp
is advanced with the increase of the input voltage. For the mixture of λ = 1.6 at the
input voltage of −12 kV, Pmax increases by almost 12.3%, and tp is advanced by
almost 31.4%, compared to the case of without electric fields. In addition, the
normalized mass burning rate and the accumulated mass fraction burned are all
enhanced substantially, and the flame development duration and the rapid burning
duration are remarkably reduced with the increase of the input voltage, and again,
the influence of electric field is more profound for leaner mixtures [2].
The experimental system is sketched in Figure 1. It consists of a constant
volume combustion chamber, the ignition control, fuel supply, electric field
generation, data acquisition and a high-speed schlieren photography system. The
combustion chamber is a cylinder type with an inner diameter of 130 mm and a
length of 130 mm. An insulating bush made of the polytetrafluoroethylene (PTFE)
with an inner diameter of 115 mm, a thickness of 7.5 mm and a length of 130 mm is
installed inside the combustion chamber. Two quartz windows with a 145 mm
diameter are mounted on two sides of the bomb to allow optical accessibility. The
ignition electrodes are located in the vertical direction along the bomb centerline and
are surrounded by the PTFE to insulate the electrics. The high-voltage electrodes
applying an electric field inside the combustion chamber, are a pair of stainless steel
straight needles with a diameter of 4 mm, and are opposing located in the horizontal
direction along the bomb centerline, 35 mm away from the bomb center,
respectively. The outside surface of the needle is very smooth, and there are not
sharp edges at the needle tip to avoid discharge at high input voltage. The power
supply (DEL30N45, Weisiman company, Xianyang, China) connected to the highvoltage electrode has the output range between 0 and 30 kV. In the experiment, a
high-speed digital camera (HG-100K) operating at 10,000 frames per second is used
to take photos of the flames during the flame propagation. The piezo-electric Kistler
absolute pressure transducer (Kistler Instruments Gmbh, Ostfildern, Germany) with
an accuracy of 0.01 kPa is applied to record the combustion pressure. Methane and
air are supplied into the chamber sequentially and their partial pressures are
determined by the desired excess air ratio λ and the total initial pressure. Five
minutes are waited before starting the ignition to ensure the homogeneity of the
methane–air mixtures. The excess air ratio is defined as λ = mair/mstoic, where mair,
and mstoic are the air masses needed in burn 1 kg fuel actually and theoretically,
respectively. For fuel-rich mixtures, λ < 1, and for fuel-lean mixture, λ > 1. In the
case of a stoichiometric mixture, λ equals unity.
Figure 2 shows the schematic diagram of the constant volume combustion
chamber arrangements with the high-voltage electrodes and the ignition electrodes.
In the experiment, four electric voltages (0, −5, −10, −12 kV) are applied to produce
the electric fields across the flames. The methane–air mixture is prepared at three
excess air radios of 1.2, 1.4 and 1.6, and it is charged in the combustion chamber at
room temperature and atmospheric pressure. For the experimental system, the input
voltage for breakdown of the flame gas between the high-voltage electrode and the
ignition electrode is about −18 kV. While the maximum voltage applied in the
experiments is −12 kV, and the electric power does not exceed 0.1% of the thermal
power. Thus, the gas discharges are not likely to occur under these conditions. Each
experiment is repeated at least three times under the same conditions and excellent
repeatability is achieved.
Practical part
After the suitable parameters of LSSVR were obtained by genetic algorithm,
a fault detection model of LRE was established based on a genetic algorithm to
optimize LSSVR (GA-LSSVR). The principle for sensor information input will be
an LRE real-time measure trained GA-LSSVR model based on historical test data,
with the engine of a sensor forecasting the output. This will get the output of the
actual output of the engine which, at the moment, is poor, residual value, and then
the residual values and setting threshold comparison, to judge the current working
state of the engine. Figure 3 shows the LRE fault detection process based on GALSSVR.
Figure 3. LRE fault detection process based on GA-LSSVR.
The computational environment for the validation simulation is MATLAB
R2020b. The computer’s processor and memory used for the computations are,
respectively, i7-10700 2.90 GHz and 64 GB. The test data of a certain type of
hydrogen-oxygen rocket engine are analyzed by using the trained model. Figure 4
shows the fitting of the GA-LSSVR fault detection model to the data collected in a
normal test cycle. It can be seen from Figure 4 that the GA-LSSVR fault detection
model can fit the normal test state well.
Figure 4. GA-LSSVR fault detection: normal test data.
It can be seen that for a GA-LSSVR based fitting of normal engine test status,
the threshold setting of 0.35 has a margin and is generally appropriate. The fitting
output of GA-LSSVR for failure test data is shown in Figure 5.
Figure 5. GA-LSSVR fault detection: fault test data.
In the experiment, the spatial electric field distributions at various input
voltages are simulated using ANSYS 13.0. For simplicity, Figure 6 only shows the
electric field distributions in the vertical cross section along the bomb centerline at
the input voltage of −12 kV. It can be seen in Figure 6 that the electric fields
produced by the high-voltage electrodes are typically non-uniform but their
distributions are all symmetrical. The electric field strengths in the horizontal
direction are more intense than that in the vertical direction. Meanwhile, the
direction of the electric field line mostly points horizontally to the high-voltage
electrodes from the vertical bomb centerline and the bomb circumference.
Figure 6. Electric field distributions at input voltage of −12 kV.
Figure 7 shows the value of electric field strength along the bomb horizontal
centerline between the high-voltage electrodes at the input voltage of −12 kV. The
electric field strength of each position can be obtained from Figure 4, and the mean
value of the electric field strength is defined as the ratio of the sum of the electric
field strength of each position to the number of the electric field strength position. It
can be seen that the value of the electric field strength almost monotonous increases
from 0.14 × 105 V/m (located in the ignition electrodes) to 2.95 × 106 V/m (located
in the high-voltage electrodes). While the electric fields in the region within 27 mm
away from the center are relatively uniform, its mean value being about 1.45 × 105
V/m, However, in the region close to the needle tip, the electric field intensity
increases significantly, reaching about 2.95 × 106 V/m from 2.7 × 105 V/m in an
interval distance of about 8 mm. The changing trends of the electric field strength at
the input voltage of −5 kV and −10 kV are in consistent with that at −12 kV due to
the electric field strength proportional to the applied voltages.
Figure 7. The electric field strength versus the horizontal distance for the case at
−12 kV.
Conclusions
1)
LRE fault detection based on genetic algorithm optimized LSSVR is
established for a certain type of hydrogen-oxygen rocket engine. To solve the
problem of LSSVR’s difficulty in finding suitable parameters, the grid search
method is used to search parameters first, and then the genetic algorithm is used to
search for optimization. After obtaining the appropriate parameters, the LRE fault
detection model was established based on genetic algorithm to optimize LSSVR,
and the model was used for fault detection of a certain type of hydrogen-oxygen
rocket engine. The simulation results show that the LRE fault detection based on
genetic algorithm optimized LSSVR can better identify the working state of the
hydrogen-oxygen rocket engine and accurately warn the fault state. The proposed
approach has engineering application value. It is worth noting that the existing fault
detection methods of rocket engine still have a lot of room for improvement due to
various limitations. In the future research, fusion detection of multiple fault detection
methods based on deep learning should be an important direction of its development.
2)
The electric field distributions created by the high-voltage electrodes
are symmetrical and their directions point to the high voltage electrodes from the
bomb circumference and the vertical bomb centerline. The electric field strengths in
the horizontal direction are more intense than that in the vertical direction. The
horizontal electric fields in the region within 27 mm away from the center are
relatively uniform, while its value vicinal the needle tip is relatively large.
3)
When the electric field is applied, the flame front is stretched
remarkably in the direction of the electric field, the stretched flame propagation
velocity and the average flame propagation velocity are accelerated significantly as
the input voltage increases. For various excess air ratios, the increase of the flame
speed for λ = 1.6 is the largest, followed by that for λ = 1.4 and 1.2. This indicates
the electric field can augment the stretch in flame, and the behavior is more
pronounced when the mixture becomes leaner.
4)
When the electric field is applied, the combustion peak pressure is
increased and its appearance time is advanced compared with the case of no electric
fields. For the heat release rate, the normalized mass burning rate and the
accumulated mass fraction burned are all improved substantially, the flame
development duration and the rapid burning duration are remarkably reduced with
the increase of the input voltage. Moreover, the electric field shows larger influence
on the combustion characteristics for the leaner mixtures.
References
1. Huang, P.; Yu, H.; Wang, T. A Study Using Optimized LSSVR for RealTime Fault Detection of Liquid Rocket Engine. Processes 2022, 10, 1643.
2. Fang, J.; Wu, X.; Duan, H.; Li, C.; Gao, Z. Effects of Electric Fields on the
Combustion Characteristics of Lean Burn Methane-Air Mixtures. Energies
2015, 8, 2587-2605.