Fault Diagnosis In Dynamic Systems Research Articles

Analysis of Numerical Integration in RNN-Based Residuals for Fault Diagnosis of Dynamic Systems

Data-driven modeling and machine learning are widely used to model the behavior of dynamic systems. One application is the residual evaluation of technical systems where model predictions are compared with measurement data to create residuals for fault diagnosis applications. While recurrent neural network models have been shown capable of modeling complex non-linear dynamic systems, they are limited to fixed steps discrete-time simulation. Modeling using neural ordinary differential equations, however, make it possible to evaluate the state variables at specific times, compute gradients when training the model and use standard numerical solvers to explicitly model the underlying dynamic of the time-series data. Here, the effect of solver selection on the performance of neural ordinary differential equation residuals during training and evaluation is investigated. The paper includes a case study of a heavy-duty truck's after-treatment system to highlight the potential of these techniques for improving fault diagnosis performance.

An evolving framework for fault diagnosis of dynamic systems

<h2>Abstract</h2> Currently, many applications work with data streams, such as financial market analysis, detection of attacks on computer network systems, fraud detection, detection and fault identification, social media analysis, e-commerce analysis, and marketing campaign assistance. In this context, approaches based on evolving systems have shown to be quite adequate for analyzing and processing this kind of data. AutoCloud is an evolving algorithm specially designed for dynamic clustering of data stream with online learning. Here, the AutoCloud is applied to fault detection and diagnosis of dynamic systems using online and unsupervised learning.

A flexi-pipe model for residual-based engine fault diagnosis to handle incomplete data and class overlapping

Data-driven fault diagnosis of dynamic systems is complicated by incomplete training data, unknown faults, and overlapping classes. Many existing machine learning models and data-driven classifiers are not expected to perform well if training data is not representative of all relevant fault realizations. In this work, a data-driven model, called a flexi-pipe model, is proposed to capture the variability of data in residual space from a few realizations of each fault class. A diagnosis system is developed as an open set classification algorithm that can handle both incomplete training data and overlapping fault classes. Data from different fault scenarios in an engine test bench is used to evaluate the performance of the proposed methods. Results show that the proposed fault class models generalize to new fault realizations when training data only contains a few realizations of each fault class.

A Data-Driven Clustering Algorithm for Residual Data Using Fault Signatures and Expectation Maximization

Clustering is an important tool in data-driven fault diagnosis to make use of unlabeled data. Collecting representative data for fault diagnosis is a difficult task since faults are rare events. In addition, using data collected from the field, e.g., logged operational data and data from different workshops about replaced components, can result in labelling uncertainties. A common approach for fault diagnosis of dynamic systems is to use residual-based features that filter out system dynamics while being sensitive to faults. The use of conventional clustering algorithms is complicated by that the distribution of residual data from one fault class varies for different realizations and system operating conditions. In this work, a clustering algorithm is proposed for residual data that clusters data by estimating fault signatures in residual space. The proposed clustering algorithm can be used on time-series data by clustering batches of data from the same fault scenario instead of clustering data sample-by-sample. The usefulness of the proposed clustering algorithm is illustrated using residual data from different fault scenarios collected from an internal combustion engine test bench.

Data-driven fault diagnosis analysis and open-set classification of time-series data

Fault diagnosis of dynamic systems is done by detecting changes in time-series data, for example residuals, caused by system degradation and faulty components. The use of general-purpose multi-class classification methods for fault diagnosis is complicated by imbalanced training data and unknown fault classes. Another complicating factor is that different fault classes can result in similar residual outputs, especially for small faults, which causes classification ambiguities. In this work, a framework for data-driven analysis and open-set classification is developed for fault diagnosis applications using the Kullback–Leibler divergence. A data-driven fault classification algorithm is proposed which can handle imbalanced datasets, class overlapping, and unknown faults. In addition, an algorithm is proposed to estimate the size of the fault when training data contains information from known fault realizations. An advantage of the proposed framework is that it can also be used for quantitative analysis of fault diagnosis performance, for example, to analyze how easy it is to classify faults of different magnitudes. To evaluate the usefulness of the proposed methods, multiple datasets from different fault scenarios have been collected from an internal combustion engine test bench to illustrate the design process of a data-driven diagnosis system, including quantitative fault diagnosis analysis and evaluation of the developed open set fault classification algorithm.

An evolving approach for fault diagnosis of dynamic systems

This work proposes a methodology for fault identification of dynamic systems using an online evolving approach. The proposed methodology is divided into three stages: pre-processing, processing, and post-processing. The central part of our approach concerns the processing itself, in which we use an online learning evolving algorithm, named AutoCloud, for clustering the different types of faults. The proposal has been validated using data from a real-level control process on a pilot scale. The obtained results indicate that our proposal is adequate for fault identification of dynamic systems.

Fault Diagnosis in Dynamic Systems Using Recurrent Neural Networks

The aim of this paper is to discuss the methodology of Fault Detection and Diagnosis (FDD) in dynamic systems. Fault diagnosis is characterized as a control system that controls the ability to adapt system component faults automatically. Fault detection is an implementation of using the error signals, where when error signal is zero or approximately zero. A coupled water tank system was used as a study case model for implementing and testing the proposed methodology. The developed system should generate a set of signals to notify the process operator about the faults that are occurring, enabling changes in control strategy or control parameters. Due to the damage risks involved with sensors, actuators and structural faults of the real plant, the data set of the faults are computationally generated and the results will be collected from numerical simulations of the process model. A Recurrent Neural Networks (RNN) is employed in this paper for modeling the used system. This paper shows how to determine the structure and how to estimate the results using the gradient-based algorithm which it is allowed to obtain a Neural Network with relatively small modeling uncertainly. It describes how to increase the stability of dynamic systems using FDD based on RNN. Faults in the study case are faults in the sensors, faults in the actuators and the structural faults. The faults which are presented in the controlled system are reduced to 98% from the default value.

Nonparametric methods for fault diagnosis in dynamic systems

SummaryThe problem of fault diagnosis in engineering systems with uncertainties described by nonlinear discrete‐time models is studied within the scope of analytical redundancy concept. Solution of the problem assumes the checking redundancy relations existing among system inputs and outputs measured over a finite time window. The nonparametric method is used to construct redundancy relations involving transformation of the initial system model into canonical form with special properties. The obtained results are illustrated by the general electric servoactuator of manipulation robots.

On Teaching Model-Based Fault Diagnosis in Engineering Curricula [Lecture Notes

Control and systems theory is a common subject in many engineering curricula. Students learn from the very basics of system modeling to the most recent advances in control algorithms. However, although fault diagnosis of dynamic systems is already a mature and an important field, it is not usually included in most engineering curricula.

State Tracking and Fault Diagnosis for Dynamic Systems Using Labeled Uncertainty Graph.

Cyber-physical systems such as autonomous spacecraft, power plants and automotive systems become more vulnerable to unanticipated failures as their complexity increases. Accurate tracking of system dynamics and fault diagnosis are essential. This paper presents an efficient state estimation method for dynamic systems modeled as concurrent probabilistic automata. First, the Labeled Uncertainty Graph (LUG) method in the planning domain is introduced to describe the state tracking and fault diagnosis processes. Because the system model is probabilistic, the Monte Carlo technique is employed to sample the probability distribution of belief states. In addition, to address the sample impoverishment problem, an innovative look-ahead technique is proposed to recursively generate most likely belief states without exhaustively checking all possible successor modes. The overall algorithms incorporate two major steps: a roll-forward process that estimates system state and identifies faults, and a roll-backward process that analyzes possible system trajectories once the faults have been detected. We demonstrate the effectiveness of this approach by applying it to a real world domain: the power supply control unit of a spacecraft.

Fault diagnosis of dynamical systems using recurrent fuzzy systems with application to an electrohydraulic servo axis

This paper presents a novel approach for fault isolation based on discrete-time recurrent fuzzy systems (DTRFS) extending the well-known static fuzzy systems for fault isolation. The DTRFS additionally allows for the isolation of faults causing dynamic residuals and the isolation of multiple subsequently occurring faults. Due to the flexibility of the DTRFS, multiplicative as well as additive faults can be handled. The approach is applied to an electrohydraulic servo axis with a duplex valve system. Experimental results at the testbed underline the effectiveness and applicability of the approach and show the enhanced performance compared to static fuzzy systems and state automata.

Diagnosis and Diagnosability of Discrete Event Systems using Petri Nets

Fault diagnosis in dynamic systems is a subject that has received a lot of attention in the past decades: the objective is that of understanding if a system is affected by a fault, and if so to identify the fault. In the context of discrete event systems faults are modeled by unobservable (or undistinguishable) events whose firing cannot be directly detected but must be reconstructed on the basis of the observed evolution. This implies that the approaches for diagnosis are strictly related to the problems of designing an observer for state and event sequence estimation.Diagnosability analysis aims to determine if a system is diagnosable, i.e., if the occurrence of fault can be identified in a finite number of steps. Both diagnosis and diagnosability have been mostly studied in the context of finite state automata, and the complexity of solving these problems is often unmanageable due to the state space explosion.This talk focus on a particular class of discrete event models, Petri nets. The goal is to show that the observer design can be efficiently addressed using Petri net models and can naturally be extended to diagnosis and diagnosability.

Fault Diagnosis in Dynamic Systems Using Fuzzy Interacting Observers

A method of fault diagnosis in dynamic systems based on a fuzzy approach is proposed. The new method possesses two basic specific features which distinguish it from the other known fuzzy methods based on the application of fuzzy logic and a bank of state observers. First, this method uses a bank of interacting observers instead of traditional independent observers. The second specific feature of the proposed method is the assumption that there is no strict boundary between the serviceable and disabled technical states of the system, which makes it possible to specify a decision making rule for fault diagnosis.

Adaptive fault detection and diagnosis using an evolving fuzzy classifier

This paper suggests an approach for adaptive fault detection and diagnosis. The proposed approach detects new operation modes of a process such as operation point changes and faults, and incorporates information about operation modes in an evolving fuzzy classifier used for diagnosis. The approach relies upon an incremental clustering procedure to generate fuzzy rules describing new operational states detected. The classifier performs diagnostic adaptively and, since every new operation mode detected is learnt and incorporated into the classifier, it is capable of identifying the same operation mode the next time it occurs. The efficiency of the approach is verified in fault detection and diagnosis of an industrial actuator. Experimental results suggest that the approach is a promising alternative for fault diagnosis of dynamic systems when there is no a priori information about all failure modes, and as an alternative to incremental learning of diagnosis systems using data streams.

A Bayesian solution to the multiple composite hypothesis testing for fault diagnosis in dynamic systems

This paper is concerned with model-based isolation and estimation of additive faults in discrete-time linear Gaussian systems. The isolation problem is stated as a multiple composite hypothesis testing on the innovation sequence of the Kalman filter (KF) that considers the system operating under fault-free conditions. Fault estimation is carried out, after isolating a fault mode, by using the Maximum a Posteriori (MAP) criterion. An explicit solution is presented for both fault isolation and estimation when the parameters of the fault modes are assumed to be realizations of specific random variables (RV).

Fault detection in nonlinear continuous‐time systems with uncertain parameters

AbstractIn model‐based fault diagnosis for dynamic systems with uncertain parameters, an envelope of all fault‐free behaviors can be determined from the model and used as a reference for detecting faults. We demonstrate here a method for generating an envelope that is rigorously guaranteed to be complete, but without significant overestimation. The method is based on an interval approach, but uses Taylor models to reduce the overestimation often associated with interval methods. To speed fault detection, a method that uses bounded‐error measurement data and a constraint propagation procedure is proposed for shrinking the envelope. Several fault detection scenarios involving nonlinear, continuous‐time systems are used to evaluate this approach. © 2008 American Institute of Chemical Engineers AIChE J, 2008

Monte Carlo simulation for model-based fault diagnosis in dynamic systems

Fault diagnosis requires the accurate estimation of the dynamic state of the system in real time. This can be pursued starting from a model of the system dynamics and on measurements related to the state of the system. In real applications, the nonlinearity of the model and non-Gaussianity of the noise typically affecting the measurement challenge the classical approximate approaches, e.g. the extended-Kalman, Gaussian-sum and grid-based filters, which often turn out to be inaccurate and/or too computationally expensive for real-time applications. On the contrary, Monte Carlo estimation methods, also called particle filters, can be very effective. Based on sequential importance sampling and on a Bayesian formulation of the estimation problem, these methods recursively approximate the relevant probability distributions of the system state by random measures composed of particles (sampled values of the unknown state variables) and associated weights. The present paper aims at demonstrating the power of particle filtering for fault diagnosis. This is done by applying an estimation procedure called sampling importance resampling (SIR) to a case study of literature.

Fault Diagnosis of Dynamical Systems Based on State-Set Observers

This paper shows how state-set observation can be used to detect and identify faults in a dynamical systems with bounded measurement uncertainties. A state-set observer uses the process model and the measured input and output sequence to determine the set of states the process can be in. For diagnosis, a bank of observers is used based on the process models for every fault case. IT an observer arrives at an empty set of states, the measured behaviour is inconsistent with the model of the corresponding fault and this fault is known to be not present. The remaining faults form the set of fault candidates. To further distinguish these, fault probabilities are calculated. A method for determining conditional probabilities from the observation sets is developed. This allows to combine the diagnostic strength of state-set observation and stochastic observation

DSD: a generic software package for model-based fault diagnosis in dynamic systems

Abstract A user-friendly, interactive software package is described that can be used for fault diagnosis in dynamic systems. The methodology is based on the representation of system evolution in time as probability of transitions between sets of magnitude intervals in the state/parameter space. The software is developed in C++ for Windows NT platform. The display capabilities of the software and its implementation are illustrated using a second order system.

Soft Computing Approaches to Fault Diagnosis for Dynamic Systems

Recent approaches to fault detection and isolation for dynamic systems using methods of integrating quantitative and qualitative model information, based upon soft computing (SC) methods are surveyed and studied in some detail. SC methods are considered an important extension to the quantitative model-based approach for residual generation in fault detection and isolation (FDI). When quantitative models are not readily available, a correctly trained neural network (NN) can be used as a non-linear dynamic model of the system. The paper describes some powerful NN methods, taking into account the dynamic as well as non-linear system behaviour. Sometimes, further insight is required as to the explicit behaviour of the model-involved and it is then that fuzzy and even neurofuzzy methods come to their own in data-driven FDI applications. The paper also discusses the use of evolutionary programming tools for observer and NN design. The paper provides many powerful examples of the use of SC methods for achieving good detection and isolation of faults in the presence of uncertain plant behaviour, together with their practical value for fault diagnosis of real process systems.