ArticlePDF Available

Fault diagnosis for the Space Shuttle main engine

Authors:
  • Artesis Technology Systems

Abstract

A conceptual design of a model-based fault detection and diagnosis system is developed for the Space Shuttle main engine. The design approach consists of process modeling, residual generation, and fault detection and diagnosis. The engine is modeled using a discrete time, quasilinear state-space representation. Model parameters are determined by identification. Residuals generated from the model are used by a neural network to detect and diagnose engine component faults. Fault diagnosis is accomplished by training the neural network to recognize the pattern of the respective fault signatures. Preliminary results for a failed valve, generated using a full, nonlinear simulation of the engine, are presented. These results indicate that the developed approach can be used for fault detection and diagnosis. The results also show that the developed model is an accurate and reliable predictor of the highly nonlinear and very complex engine.
JOURNAL
OF
GUIDANCE, CONTROL,
AND
DYNAMICS
Vol.
15, No. 2,
March-April
1992
Fault
Diagnosis
for the
Space Shuttle
Main
Engine
Ahmet Duyar*
Florida
Atlantic
University,
Boca
Raton,
Florida 33431
and
Walter
Merrillt
NASA
Lewis
Research
Center,
Cleveland,
Ohio
44135
A
conceptual design
of a
model-based fault detection
and
diagnosis system
is
developed
for the
Space Shuttle
main
engine.
The
design approach consists
of
process modeling, residual generation,
and
fault
detection
and
diagnosis.
The
engine
is
modeled using
a
discrete
time,
quasilinear state-space representation. Model parameters
are
determined
by
identification. Residuals generated from
the
model
are
used
by a
neural network
to
detect
and
diagnose engine component faults. Fault diagnosis
is
accomplished
by
training
the
neural network
to
recognize
the
pattern
of the
respective
fault
signatures. Preliminary results
for a
failed valve, generated using
a
full,
nonlinear
simulation
of the
engine,
are
presented. These results indicate
that
the
developed approach
can be
used
for
fault detection
and
diagnosis.
The
results also show that
the
developed model
is an
accurate
and
reliable
predictor
of the
highly nonlinear
and
very complex engine.
Introduction
T
HIS
paper describes
a
model-based fault diagnosis system
based
on a
neural network classifier
for the
Space
Shuttle
main engine (SSME).
The
system
may be
used
to
monitor
the
life
cycle
of
engine components
and for the
early detection,
isolation,
and the
diagnosis
of
engine faults.
As
such,
the
proposed system
will
be one
part
of a
larger, engine health
monitoring
system.1
The
health monitoring system
will
allow
for
accommodation
of
faults, reduce maintenance
cost,
in-
crease engine availability,
and be one
part
of an
integrated,
intelligent
control
system2
for the
SSME.
A
description
of
SSME dynamics
and its
modeling
is
given
in a
study
by
Duyar
et
al.3
A
summary
of the
major failures
of the
SSME
that
have
occurred
are
outlined
by
Cikanek.4 Several
authors5'8
survey
the
available methods
and
approaches
for
fault detection
and
diagnosis.
In
particular,
the
survey
by
Isermann6 gives several
examples
of the use of
identification techniques
for
process
fault
detection.
A
fault
is the
abnormal behavior
of a
component
due to
physical
change
in the
component.
A
fault event often impairs
or
deteriorates
the
system's
ability
to
perform
its
specified
tasks
or
mission.
The
detection task
is
defined
as the act of
identifying
the
presence
of an
unspecified fault. After
a
fault
is
detected, then
the
fault must
be
isolated
to the
component
that
has
failed. During
the
process
of
isolation,
the
magnitude
of
the
fault
may be
estimated. Fault diagnosis
is the
isolation
and
estimation
of a
fault
mode.
Once
a
fault
is
detected
and
diagnosed,
the
fault
can be
accommodated through reconfigu-
ration
of the
system. Reconfiguration includes
both
hardware
actions (e.g., activating backup systems)
and
software tasks
(e.g., adjusting
the
feedback control gains).
The
detection
and
diagnostic tasks
may be
accomplished
by an
onboard
proces-
sor,
on
line
and in
real time
for
fault accommodation,
as
well
as by an
off-line
processor that analyzes recorded
data
for
life-cycle
analysis
and
preventive maintenance.
Received July
7,
1990; revision received Oct.
26,
1990;
accepted
for
publication Feb.
21,
1991.
Copyright
©
1991
by the
American
Insti-
tute
of
Aeronautics
and
Astronautics, Inc.
No
copyright
is
asserted
in
the
United States under
Title
17,
U.S.
Code.
The
U.S. Government
has a
royalty-free license
to
exercise
all
rights under
the
copyright
claimed herein
for
Governmental purposes.
All
other
rights
are re-
served
by the
copyright owner.
*Professor, Mechanical Engineering Department.
fDeputy
Branch Chief, Advanced
Controls
Technology Branch,
Mail
Stop
77-1,
21000
Brookpark
Road.
Senior Member
AIAA.
384
Initially,
a
brief description
of the
conceptual design
of the
model-based fault detection
and
diagnostic system (FDDS)
is
given.
This
is
followed
by a
description
of the
process model-
ing,
the
residual generation method,
and the
detection
and
diagnostic system design. Finally, results
of the
application
of
the
FDDS
to the
detection
of a
stuck valve fault using simu-
lated
data
are
presented.
Conceptual Design
Model-based
fault
detection methods rely
on the
determina-
tion
of
changes appearing
in the
system
due to the
existence
of
a
fault,
in
comparison with
the
normal status
of the
system.
For
example,
in
aerospace applications control actuator faults
may
be
represented
as
shifts
in the
parameters
of the
control
gain matrix. Sensor faults
may be
represented
as
abrupt
changes
in the
parameters
of the
output matrix
or
increases
in
measurement
noise. These changes
are
determined
by
compar-
ing
the
parameters
of the
observed process with
the
parame-
ters
obtained
from
the
model
of the
normal process.
The
differences
between these parameters
are
called
residuals.
The
residuals
and
their patterns
are
analyzed
for
fault detection
and
diagnosis
by
comparing them with
the
known fault signa-
tures
of the
process.
Fault signatures, which show
the
effect
of a
fault
on the
parameters,
are
generated
by
inducing faults
in the
nonlinear
dynamic
simulation
of the
process. Fault diagnosis
is
accom-
plished
by
training
a
neural network classifier
to
recognize
the
pattern
of the
respective fault signatures.
The
design
of the
FDDS
is
accomplished
in
three stages:
process modeling, residual generation
and
fault
detection,
and
diagnostic
classifier design.
In the
following sections
the
meth-
ods
used
in
these stages
are
briefly
explained.
Process Modeling
A
complete nonlinear dynamic simulation
of
SSME perfor-
mance
was
developed
by
Rocketdyne Division
of
Rockwell
International
Corporation.9
In
this study, this nonlinear
model
is
considered
as the
unknown process.
It is
used
for the
generation
of
fault
signatures
by
modifying
the
actuator mod-
els
to
include
a
fault
model.
The
input-output
data
generated
from
this simulation
are
also used
to
identify
the
parameters
of
the
engine.
Due to its
size
and
complexity
(40 min of CPU
time
for 20 s of
real-time operation with
a VAX
8800), this
nonlinear
simulation cannot
be
used
to
generate data
in
real
time
to
describe
the
normal mode
of
operation.
... It was initially used for the monitoring and detection of damages on NASA's Space Shuttle main engine. [6]. The use of statistical approach methods for the determination of the maintenance time was one of the basic research methodologies. ...
... The use of predictive maintenance is becoming increasingly popular. It was initially used for monitoring and detecting engine failures at NASA's Space Shuttle [6]. The use of statistical approaches for the determination of the correct maintenance time has been a research method that turned out to be less successful than expected and therefore quickly sidelined. ...
Article
Full-text available
Nowadays, frequent maintenance and repair of mechanical equipment whose goal is to deter the suspension time of railway infrastructures are proven to be ineffectual. It also results in loss of reliability as well as consuming unnecessary means and costs, since at least half of the precautionary maintenance activities are considered redundant. Despite this spending, operators are struggling to adequately maintain their assets—resulting in unacceptably frequent delays and cancellations and low levels of satisfaction among rail users. Thanks to the increasing availability and sophistication of advanced analytics, operators have a significant opportunity to create solutions to long-standing maintenance challenges. The role of predictive maintenance and especially that of Design-Out Maintenance constitute the necessary procedure that can predict in time any hardware failures, while reducing the damage or wearing down of the overall operational equipment. This can increase the effectiveness of the railways, while significantly reducing the overall expenditure needed for the repair and maintenance of the industry’s infrastructure. This paper proposes a predictive railways maintenance strategy based on deep learning techniques. Specifically, and in order to achieve the exact remaining useful life of the railway equipment, a hybrid neural architecture of long short-term memory autoencoder network is used. The purpose of this suggested architecture is the automatic feature extraction of dynamic time series and their utilization on a prediction model, which can predict with high accuracy the remaining useful life of the railway’s mechanical equipment.
... Currently, the predictive care maintenance is ignored although it provides many benefits and prevent failure before it occurs [10]. The analysis of data from the malfunctions can be used in high-capacity use of engines and in avoiding shutdowns [11]. The systems used to determine the malfunction of products such as electric motors and generators commonly measure voltage and current signals to determine the malfunction status [12,13,14]. ...
... This parameter is essential for the correct operation of the injection system and of the control unit. In the mathematical model (11), the movement of the shutter is modeled, consequently it is possible to derive the anchor movement [24]. ...
... With the development of control theory and computer technology, control technology has been continuously improved. At the same time, due to the increasingly high requirements for improving the reliability and performance of the system, many linear models such as: Kalman filter [1], parameter estimation method [2] and feature structure allocation technology [3] are applied to the field of automotive engine fault diagnosis expert system. Because the linear model can not reasonably and accurately simulate the fault impact, Passino et al. developed a fault diagnosis system based on a nonlinear multi-model fault diagnosis isolation method, and studied the robustness and fault sensitivity of the method. ...
... II) Adaptability function The fitness function is represented by the dependency in (2)(3)(4)(5)(6)(7). The greater the dependency between the conditional attribute and the decision attribute, the more important the conditional attribute is, that is, the greater the adaptability of the individual. ...
Article
Full-text available
Under the high load, high frequency and high strength operating environment, the frequent occurrence of vehicle fault gradually attracts the attention of the society. The real-time monitoring and data recording function of vehicle-mounted equipment provides data support for vehicle status assessment and fault warning. In this paper, the real-time data collected by CAN-BUS system of Beijing Bus Group are preprocessed and discretized. On the basis of the traditional rough set theory, a new coding method is set up, and the dependency between conditional attributes and decision attributes is set as an adaptive function, which is reduced by genetic algorithm and cellular genetic algorithm respectively. The calculation results show that the key fault information of public transport vehicles is instrumental speed, oil pressure, percentage of torque, timing engine speed, and coolant temperature. By comparing the results of reduction, it is found that the cellular genetic algorithm has higher applicability than the genetic algorithm in terms of algorithm efficiency, stability, and convergence quality. Although the genetic algorithm attribute reduction is slightly better than the cellular genetic algorithm attribute reduction in the rule matching, the cellular genetic algorithm has a better ability to excavate information within the acceptable compatibility range. Finally, the selected key factors will be deployed on the Beijing Bus Group’s big data platform and displayed in real time. The conclusion of this paper enriches the theory of bus engine fault warning and establishes an engine failure warning system, which can effectively reduce the failure rate of bus vehicles and reduce the maintenance cost expenditure. It has certain guiding significance for the bus operation work of Beijing Bus Group.
... Although the benefits of predictive care are known, it has been determined that the proportion of firms that determine their malfunctions by applying fractional care in the world is 0.04% [13]. e systems used to determine the malfunction of products such as electric motors, generators, and transformers commonly used in the industry measure voltage and current signals to determine the malfunction status of related products [14,15]. Depending on the operating conditions and the characteristics of the points to be measured using various parameters that characterize the behaviour of systems at runtime to monitor the operating conditions and performances of the machine, the working performance of the machine under various physical quantities can be observed at certain intervals [16]. ...
Article
Full-text available
Vibration and temperature data of a servo motor are analyzed with PLC which is widely used in the industry. With this system, power supply can be detected on the servo motors. In this way, undesirable situations such as disruptions in production and productivity loss can be prevented from occurring. It is an important problem for businesses to detect malfunctions that may occur in servo motor dysfunction. Previously, methods such as ultrasonic sound measurements, thermal cameras, endoscopy equipment, and energy analysis have been used and discussed in the literature. Our study offers a PLC-based vibration and temperature measurement system designed as a solution of this problem. In this system, vibration and temperature measurements were made while the servo motor was kept running. These measurements were measured with or without load, considering the operating ranges of the servo motor, and the compatibility of the data was evaluated.
... Monitoring engines and test benches is a major challenge in the develop-45 ment and integration of new propulsion systems for rockets, including reusable ones [1]. The increase of launchers operation safety, economical efficiency and launcher's engines reliability [2,3] are major problematic [4], [5], [6]. For those reasons Health Management Systems (HMS) have to be improved to perform technical and economical optimization of launchers systems, and in particular 50 of engine systems. ...
Article
In this paper a method is proposed to design a fault detection and isolation scheme based on quantitative physics-based models, as well as fault-tolerant control strategy to improve the reliability of a cryogenic combustion bench operation. The detection and isolation scheme is composed of an extended observer, a cumulative sum algorithm and an exponentially weighted moving average chart. In the case of interdependent parts, a dynamic parity space approach is proposed to isolate one or two simultaneous faults with constraints based on the mass flow rate continuity and the energy conservation for the overall system. The method allows settling adaptive thresholds that avoid pessimistic decision about the continuation of tests while detecting and isolating faults in the system. Then a fault-tolerant system reconfiguration mechanism is provided with a control law which compensates for an estimated actuator additive fault to maintain the overall system stability or allows converging to a reference state in order to overcome instabilities and take into account actuator saturations. For that purpose, the fault-tolerant control algorithm comprises a linear quadratic regulator, an unknown input observer to estimate the fault and an anti-windup scheme. The model and the estimation part were validated on real data from the ONERA/CNES MASCOTTE test bench, and the reconfiguration control law was validated in realistic simulations of the same system.
... In [3], Duyar and Merril generated linear-point models offline with an identification algorithm to develop an HMS for the SSME, see Table 2.2. An HMS was proposed to enhance the monitoring of SSME and consists in three detection algorithms, ARMA, RESID and Cluster used in the first level to process sensor data in parallel. ...
Thesis
Monitoring and improving the operating modes of launcher propulsion systems are major challenges in the aerospace industry. A failure or malfunction of the propulsion system can have a significant impact for institutional or private customers and results in environmental or human catastrophes. Health Management Systems (HMS) for liquid propellant rocket engines (LPREs), have been developed to take into account the current challenges by addressing safety and reliability issues. Their objective was initially to detect failures or malfunctions, isolate them and take a decision using Redlines and Expert Systems. However, those methods can induce false alarms or undetected failures that can be critical for the operation safety and reliability. Hence, current works aim at eliminating some catastrophic failures but also to mitigate benign shutdowns to non-shutdown actions. Since databases are not always sufficient to use efficiently data-based analysis methods, model-based methods are essentially used. The first task is to detect component and / or instrument failures with Fault Detection and Isolation (FDI) approaches. If the failure is minor, non-shutdown actions must be defined to maintain the overall system current performances close to the desirable ones and preserve stability conditions. For this reason, it is required to perform a robust (uncertainties, unknown disturbances) reconfiguration of the engine. Input saturation should also be considered in the control law design since unlimited control signals are not available due to physical actuators characteristics or performances. The three objectives of this thesis are therefore: the modeling of the different main subsystems of a LPRE, the development of FDI algorithms from the previously developed models and the definition of a real-time engine reconfiguration system to compensate for certain types of failures. The developed FDI and Reconfiguration (FDIR) scheme based on those three objectives has then been validated with the help of simulations with CARINS (CNES) and the MASCOTTE test bench (CNES/ONERA).
... For instance, the steady-state performance of the motor can be estimated once a dynamical model of the motor is available. This follows from the fact that the dynamical behavior of the motor is completely determined by its equations of motion, such as those given in Equation (1). For instance, the torque produced by the motor at steady-state is given by ...
Article
Full-text available
A model based approach to fault detection is described, Its practical application has resulted in a breakthrough technology for the evaluation of electrical motors. The Motor quality Monitor (MQM) is a system for the detection of motor faults on the production line or at an incoming quality control station. The technique can detect a wide range of electrical and mechanical problems, including noise and vibration. In addition, the physical model allows rapid estimation of motor performance characteristics.
Thesis
Full-text available
Conveyor belt systems (CBS) are a fundamental class of asset throughout bulk handling industries, which provide materials transfer functionality within processes spanning all stages of a material’s life, from initial extraction through to end-of-life disposal or reuse. The generic function a CBS affords can result in a single system being used in a diverse range of applications throughout its service life, including exposure to different materials, environments and throughputs. However, regardless of the specific application a CBS will be expected to realise very high levels of availability, such that the occurrence of costly process downtime is minimised. To support the realisation of such levels of availability, increasingly across industry continuous monitoring (CM) of systems is being implemented, to provide greater insight into the operation of systems. Development within the area of CM has historically been driven by the needs of high value industries such as aerospace and renewables, thus, it cannot be assumed that existing solutions represent the most appropriate form of CM for CBS applications. Furthermore, the interrogation of raw CM parameters requires effort and expertise from practitioners in order to elicit actionable insights, which cannot be assumed available across operations. Accordingly, this thesis reports the development of a practical method for observing the operation of a CBS in-service, using continuous monitoring techniques. Initially, current practice and challenges faced by industrial practitioners are investigated through interactions with a manufacturer and an operator of bulk handling CBSs. From these interactions an opportunity to realise improvements across the design, operation and maintenance of CBSs through the adoption of CM is established. Next, through a series of laboratory and industrial trials a range of CBS parameters are investigated and evaluated within the context of continuous monitoring, from which motor electrical power consumption (MEPC) is identified as most appropriate given the specific characteristics of bulk handling applications. Finally, a data processing method for translating raw MEPC parameters into actionable insights is developed, comprising the production of three data descriptors, each of which provides a quantification of a specific aspect of CBS operation, related to its usage or health. Finally, the application and potential utility therein of the method is demonstrated through further industrial trials.
Thesis
The main objective of this work is to demonstrate and analyze the potential benefits of advanced real time algorithms for rocket engines monitoring and diagnosis. In the last two decades in Europe many research efforts have been devoted to the development of specific diagnostic technics such as neural networks, vibration analysis or parameter identifcation but few results are available concerning algorithms comparison and diagnosis performances analysis. Another major objective of this work has been the improvement of the monitoring system of the Mascotte test bench (ONERA/CNES).This is a cryogenic test facility based in ONERA Palaiseau used to perform analysis of cryogenic combustion and nozzle expansion behavior representative of real rocket engine operations. The first step of the work was the selection of a critical system of the bench, the water cooling circuit, and then the analysis of the possible model based technics for diagnostic such as parameter identification and Kalman filters. Three new algorithms were developed, after a preliminary validation based on real test data, they were thoroughly analyzed via a functional benchmark with representative failure cases.The last part of the work consisted in the integration of the diagnosis algorithms on the bench computer environment in order to prepare a set-up for a future real time application. A simple closed loop architecture based on the new diagnostic tools has been studied in order to assess the potential of the new methods for future application in the context of intelligent bench control strategies.
Article
Based on fuzzy rule sets match method which is a series of fuzzy neural networks, a system framework used for the fault diagnosis is proposed. This fault diagnosis system consists of five parts, including the extraction of fuzzy rules, fuzzy reference rule sets, the fuzzy rule scheduled to detect, the fuzzy match module and the diagnosis logic module. The extraction of fuzzy rules involves two steps: step one adaptively divides the whole space of the trained data into the subspaces in the form of hypersphere, which is expected efficiently to work out the recognition questions in the high dimension space; step two generates a fuzzy rule in each sample subspace and calculates the membership degree of each fuzzy rule. Many fuzzy reference rule sets are produced by the extraction module of fuzzy rules for the offline learning, and a fuzzy rule set to be detected is online formed while the monitoring process is happening. Beliefs estimated from the fuzzy match process of fuzzy rule sets, which indicate the existence of the working classes in the plant, the diagnosis logic module can export fault detection time, fault isolation time, fault type and fault degree. The simulation researches of the fault diagnosis in space propulsion system demonstrate the superior qualities of the fault diagnosis method on the basis of the fuzzy match of the fuzzy rule sets.
Article
Full-text available
A simplified model is presented of the space shuttle main engine (SSME) dynamics valid within the range of operation of the engine. This model is obtained by linking the linearized point models obtained at 25 different operating points of SSME. The simplified model was developed for use with a model-based diagnostic scheme for failure detection and diagnostics studies, as well as control design purposes.
Article
Full-text available
A parameter and structure estimation technique for multivariable systems is used to obtain state space representation of open loop dynamics of the Space Shuttle Main Engine (SSME). The parameterization being used is both minimal and unique. The simplified linear models may be used for fault detection studies and control system design and development.
Article
Full-text available
System identification techniques are used to represent the dynamic behavior of the SSME (space shuttle main engine). The comparison of the responses of a nonlinear simulation with the responses of an identified model indicates very good agreement. The identified model can be used for control design purposes. The identified model does not include valve linkage backlash and valve stiction nonlinearities. These nonlinearities should be added to the identified model, and the validity of the model should be checked by comparing it with the full nonlinear simulation. The identified model is valid for a limited response region at about the 100% power level operating condition.< >
Article
Full-text available
Techniques to detect and isolate failures in complex technological systems, such as sensor biases, actuator malfunctions, leaks, and equipment deterioration are surveyed. The methods are based on analytical redundancy afforded by a mathematical model of the system. The main components of such techniques are residual generation using the model, signature generation by statistical testing, and signature analysis. Model-structural conditions for failure isolation are introduced together with transformation methods to implement them. Sensitivity and robustness considerations are presented, and a design framework based on model redundancy is proposed.< >
Article
In this paper we survey a number of methods for the detection of abrupt changes (such as failures) in stochastic dynamical systems. We concentrate on the class of linear systems, but the basic concepts, if not the detailed analyses, carry over to other classes of systems. The methods surveyed range from the design of specific failure-sensitive filters, to the use of statistical tests on filter innovations, to the development of jump process formulations. Tradeoffs in complexity vs performance are discussed.
Article
The paper reviews the state of the art of fault detection and isolation in automatic processes using analytical redundancy, and presents some new results. It outlines the principles and most important techniques of model-based residual generation using parameter identification and state estimation methods with emphasis upon the latest attempts to achieve robustness with respect to modelling errors. A solution to the fundamental problem of robust fault detection, providing the maximum achievable robustness by decoupling the effects of faults from each other and from the effects of modelling errors, is given. This approach not only completes the theory but is also of great importance for practical applications. For the case where the prerequisites for complete decoupling are not given, two approximate solutions—one in the time domain and one in the frequency domain—are presented, and the crossconnections to earlier approaches are evidenced. The resulting observer schemes for robust instrument fault detection, component fault detection, and actuator fault detection are briefly discussed. Finally, the basic scheme of fault diagnosis using a combination of analytical and knowledge-based redundancy is outlined.
Article
The supervision of technical processes is the subject of increased development because of the increasing demands on reliability and safety. The use of process computers and microcomputers permits the application of methods which result in an earlier detection of process faults than is possible by conventional limit and trend checks. With the aid of process models, estimation and decision methods it is possible to also monitor nonmeasurable variables like process states, process parameters and characteristic quantities. This contribution presents a brief summary of some basic fault detection methods. This is followed by a description of suitable parameter estimation methods for continuous-time models. Then two examples are considered, the fault detection of an electrical driven centrifugal pump by parameter monitoring and the leak detection for pipelines by a special correlation method.
Article
The problem of how to improve recognition of anomalous behavior in the Space Shuttle Main Engine is currently under study. The effort is planned to lead to an advanced real-time failure detection system for test stand application. This paper addresses motivation for the study, engine characteristics, failure detection problems, and the technical issues that are involved.