Spelling suggestions: "subject:"[een] FAULT DETECTION"" "subject:"[enn] FAULT DETECTION""
61 |
Model based fault detection for two-dimensional systemsWang, Zhenheng 05 May 2014 (has links)
Fault detection and isolation (FDI) are essential in ensuring safe and reliable operations in industrial
systems. Extensive research has been carried out on FDI for one dimensional (1-D)
systems, where variables vary only with time. The existing FDI strategies are mainly focussed
on 1-D systems and can generally be classified as model based and process history data based
methods. In many industrial systems, the state variables change with space and time (e.g., sheet
forming, fixed bed reactors, and furnaces). These systems are termed as distributed parameter
systems (DPS) or two dimensional (2-D) systems. 2-D systems have been commonly represented
by the Roesser Model and the F-M model. Fault detection and isolation for 2-D systems
represent a great challenge in both theoretical development and applications and only limited
research results are available.
In this thesis, model based fault detection strategies for 2-D systems have been investigated
based on the F-M and the Roesser models. A dead-beat observer based fault detection has been
available for the F-M model. In this work, an observer based fault detection strategy is investigated
for systems modelled by the Roesser model. Using the 2-D polynomial matrix technique,
a dead-beat observer is developed and the state estimate from the observer is then input to a
residual generator to monitor occurrence of faults. An enhanced realization technique is combined
to achieve efficient fault detection with reduced computations. Simulation results indicate
that the proposed method is effective in detecting faults for systems without disturbances as well
as those affected by unknown disturbances.The dead-beat observer based fault detection has been shown to be effective for 2-D systems
but strict conditions are required in order for an observer and a residual generator to exist. These
strict conditions may not be satisfied for some systems. The effect of process noises are also not
considered in the observer based fault detection approaches for 2-D systems. To overcome the
disadvantages, 2-D Kalman filter based fault detection algorithms are proposed in the thesis. A recursive 2-D Kalman filter is applied to obtain state estimate minimizing the estimation
error variances. Based on the state estimate from the Kalman filter, a residual is generated
reflecting fault information. A model is formulated for the relation of the residual with faults
over a moving evaluation window. Simulations are performed on two F-M models and results
indicate that faults can be detected effectively and efficiently using the Kalman filter based fault
detection.
In the observer based and Kalman filter based fault detection approaches, the residual signals
are used to determine whether a fault occurs. For systems with complicated fault information
and/or noises, it is necessary to evaluate the residual signals using statistical techniques. Fault
detection of 2-D systems is proposed with the residuals evaluated using dynamic principal component
analysis (DPCA). Based on historical data, the reference residuals are first generated using
either the observer or the Kalman filter based approach. Based on the residual time-lagged
data matrices for the reference data, the principal components are calculated and the threshold
value obtained. In online applications, the T2 value of the residual signals are compared with
the threshold value to determine fault occurrence. Simulation results show that applying DPCA
to evaluation of 2-D residuals is effective.
|
62 |
Soft Sensors for Process Monitoring of Complex ProcessesSerpas, Mitchell Roy 2012 August 1900 (has links)
Soft sensors are an essential component of process systems engineering schemes. While soft sensor design research is important, investigation into the relationships between soft sensors and other areas of advanced monitoring and control is crucial as well. This dissertation presents two new techniques that enhance the performance of fault detection and sensor network design by integration with soft sensor technology. In addition, a chapter is devoted to the investigation of the proper implementation of one of the most often used soft sensors. The performance advantages of these techniques are illustrated with several cases studies.
First, a new approach for fault detection which involves soft sensors for process monitoring is developed. The methodology presented here deals directly with the state estimates that need to be monitored. The advantage of such an approach is that the nonlinear effect of abnormal process conditions on the state variables can be directly observed. The presented technique involves a general framework for using soft sensor design and computation of the statistics that represent normal operating conditions.
Second, a method for determining the optimal placement of multiple sensors for processes described by a class of nonlinear dynamic systems is described. This approach is based upon maximizing a criterion, i.e., the determinant, applied to the empirical observability gramian in order to optimize certain properties of the process state estimates. The determinant directly accounts for redundancy of information, however, the resulting optimization problem is nontrivial to solve as it is a mixed integer nonlinear programming problem. This paper also presents a decomposition of the optimization problem such that the formulated sensor placement problem can be solved quickly and accurately on a desktop PC.
Many comparative studies, often based upon simulation results, between Extended Kalman filters (EKF) and other estimation methodologies such as Moving Horizon Estimation or Unscented Kalman Filter have been published over the last few years. However, the results returned by the EKF are affected by the algorithm used for its implementation and some implementations may lead to inaccurate results. In order to address this point, this work provides a comparison of several different algorithms for implementation.
|
63 |
Impact of data quality on photovoltaic (PV) performance assessmentKoubli, Eleni January 2017 (has links)
In this work, data quality control and mitigation tools have been developed for improving the accuracy of photovoltaic (PV) system performance assessment. These tools allow to demonstrate the impact of ignoring erroneous or lost data on performance evaluation and fault detection. The work mainly focuses on residential PV systems where monitoring is limited to recording total generation and the lack of meteorological data makes quality control in that area truly challenging. Main quality issues addressed in this work are with regards to wrong system description and missing electrical and/or meteorological data in monitoring. An automatic detection of wrong input information such as system nominal capacity and azimuth is developed, based on statistical distributions of annual figures of PV system performance ratio (PR) and final yield. This approach is specifically useful in carrying out PV fleet analyses where only monthly or annual energy outputs are available. The evaluation is carried out based on synthetic weather data which is obtained by interpolating from a network of about 80 meteorological monitoring stations operated by the UK Meteorological Office. The procedures are used on a large PV domestic dataset, obtained by a social housing organisation, where a significant number of cases with wrong input information are found. Data interruption is identified as another challenge in PV monitoring data, although the effect of this is particularly under-researched in the area of PV. Disregarding missing energy generation data leads to falsely estimated performance figures, which consequently may lead to false alarms on performance and/or the lack of necessary requirements for the financial revenue of a domestic system through the feed-in-tariff scheme. In this work, the effect of missing data is mitigated by applying novel data inference methods based on empirical and artificial neural network approaches, training algorithms and remotely inferred weather data. Various cases of data loss are considered and case studies from the CREST monitoring system and the domestic dataset are used as test cases. When using back-filled energy output, monthly PR estimation yields more accurate results than when including prolonged data gaps in the analysis. Finally, to further discriminate more obscure data from system faults when higher temporal resolution data is available, a remote modelling and failure detection framework is ii developed based on a physical electrical model, remote input weather data and system description extracted from PV module and inverter manufacturer datasheets. The failure detection is based on the analysis of daily profiles and long-term PR comparison of neighbouring PV systems. By employing this tool on various case studies it is seen that undetected wrong data may severely obscure fault detection, affecting PV system s lifetime. Based on the results and conclusions of this work on the employed residential dataset, essential data requirements for domestic PV monitoring are introduced as a potential contribution to existing lessons learnt in PV monitoring.
|
64 |
Distributed sensor fault detection and isolation over wireless sensor networkJingjing, Hao 07 July 2017 (has links)
Wireless sensor networks (WSNs) can provide new methods for information gathering for a variety of applications. In order to ensure the network quality of service, the quality of the measurements has to be guaranteed. Distributed fault detection and isolation schemes are preferred to centralized solutions to diagnose faulty sensors in WSNs. Indeed the first approach avoids the need for a central node that collects information from every sensor node, and hence it limits complexity and energy cost while improving reliability.In the case of state estimation over distributed architectures, the sensor faults can be propagated in the network during the information exchanging process. To build a reliable state estimate one has to make sure that the measurements issued by the different sensors are fault free. That is one of the motivations to build a distributed fault detection and isolation (FDI) system that generates an alarm as soon as a measurement is subject to a fault (has drift, cdots ). In order to diagnose faults with small magnitude in wireless sensor networks, a systematic methodology to design and implement a distributed FDI system is proposed. It resorts to distinguishability measures to indicate the performance of the FDI system and to select the most suitable node(s) for information exchange in the network with a view to FDI. It allows one to determine the minimum amount of data to be exchanged between the different nodes for a given FDI performance. In this way, the specifications for FDI can be achieved while the communication and computation cost are kept as small as possible. The distributed FDI systems are designed both in deterministic and stochastic frameworks. They are based on the parity space approach that exploits spacial redundancy as well as temporal redundancy in the context of distributed schemes. The decision systems with the deterministic method and the stochastic method are designed not only to detect a fault but also to distinguish which fault is occurring in the network. A case study with a WSN is conducted to verify the proposed method. The network is used to monitor the temperature and humidity in a computer room. The distributed FDI system is validated both with simulated data and recorded data. / Doctorat en Sciences de l'ingénieur et technologie / info:eu-repo/semantics/nonPublished
|
65 |
Advanced process monitoring using wavelets and non-linear principal component analysisFourie, Steven 12 January 2007 (has links)
The aim of this study was to propose a nonlinear multiscale principal component analysis (NLMSPCA) methodology for process monitoring and fault detection based upon multilevel wavelet decomposition and nonlinear principal component analysis via an input-training neural network. Prior to assessing the capabilities of the monitoring scheme on a nonlinear industrial process, the data is first pre-processed to remove heavy noise and significant spikes through wavelet thresholding. The thresholded wavelet coefficients are used to reconstruct the thresholded details and approximations. The significant details and approximations are used as the inputs for the linear and nonlinear PCA algorithms in order to construct detail and approximation conformance models. At the same time non-thresholded details and approximations are reconstructed and combined which are used in a similar way as that of the thresholded details and approximations to construct a combined conformance model to take account of noise and outliers. Performance monitoring charts with non-parametric control limits are then applied to identify the occurrence of non-conforming operation prior to interrogating differential contribution plots to help identify the potential source of the fault. A novel summary display is used to present the information contained in bivariate graphs in order to facilitate global visualization. Positive results were achieved. / Dissertation (M Eng (Control Engineering))--University of Pretoria, 2007. / Chemical Engineering / unrestricted
|
66 |
Fault monitoring in hydraulic systems using unscented Kalman filterSepasi, Mohammad 05 1900 (has links)
Condition monitoring of hydraulic systems is an area that has grown
substantially in the last few decades. This thesis presents a scheme that
automatically generates the fault symptoms by on-line processing of raw sensor data
from a real test rig. The main purposes of implementing condition monitoring in
hydraulic systems are to increase productivity, decrease maintenance costs and
increase safety. Since such systems are widely used in industry and becoming more
complex in function, reliability of the systems must be supported by an efficient
monitoring and maintenance scheme.
This work proposes an accurate state space model together with a novel
model-based fault diagnosis methodology. The test rig has been fabricated in the
Process Automation and Robotics Laboratory at UBC. First, a state space model of
the system is derived. The parameters of the model are obtained through either
experiments or direct measurements and manufacturer specifications. To validate the
model, the simulated and measured states are compared. The results show that under
normal operating conditions the simulation program and real system produce similar
state trajectories.
For the validated model, a condition monitoring scheme based on the
Unscented Kalman Filter (UKF) is developed. In simulations, both measurement and
process noises are considered. The results show that the algorithm estimates the
iii
system states with acceptable residual errors. Therefore, the structure is verified to
be employed as the fault diagnosis scheme.
Five types of faults are investigated in this thesis: loss of load, dynamic
friction load, the internal leakage between the two hydraulic cylinder chambers, and
the external leakage at either side of the actuator. Also, for each leakage scenario,
three levels of leakage are investigated in the tests. The developed UKF-based fault
monitoring scheme is tested on the practical system while different fault scenarios
are singly introduced to the system. A sinusoidal reference signal is used for the
actuator displacement. To diagnose the occurred fault in real time, three criteria,
namely residual moving average of the errors, chamber pressures, and actuator
characteristics, are considered. Based on the presented experimental results and
discussions, the proposed scheme can accurately diagnose the occurred faults. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate
|
67 |
Automatic clustering with application to time dependent fault detection in chemical processesLabuschagne, Petrus Jacobus 06 July 2009 (has links)
Fault detection and diagnosis presents a big challenge within the petrochemical industry. The annual economic impact of unexpected shutdowns is estimated to be $20 billion. Assistive technologies will help with the effective detection and classification of the faults causing these shutdowns. Clustering analysis presents a form of unsupervised learning which identifies data with similar properties. Various algorithms were used and included hard-partitioning algorithms (K-means and K-medoid) and fuzzy algorithms (Fuzzy C-means, Gustafson-Kessel and Gath-Geva). A novel approach to the clustering problem of time-series data is proposed. It exploits the time dependency of variables (time delays) within a process engineering environment. Before clustering, process lags are identified via signal cross-correlations. From this, a least-squares optimal signal time shift is calculated. Dimensional reduction techniques are used to visualise the data. Various nonlinear dimensional reduction techniques have been proposed in recent years. These techniques have been shown to outperform their linear counterparts on various artificial data sets including the Swiss roll and helix data sets but have not been widely implemented in a process engineering environment. The algorithms that were used included linear PCA and standard Sammon and fuzzy Sammon mappings. Time shifting resulted in better clustering accuracy on a synthetic data set based on than traditional clustering techniques based on quantitative criteria (including Partition Coefficient, Classification Entropy, Partition Index, Separation Index, Dunn’s Index and Alternative Dunn Index). However, the time shifted clustering results of the Tennessee Eastman process were not as good as the non-shifted data. Copyright / Dissertation (MEng)--University of Pretoria, 2009. / Chemical Engineering / unrestricted
|
68 |
DETECTION AND EXCLUSION OF FAULTY GNSS MEASUREMENTS: A PARAMETERIZED QUADRATIC PROGRAMMING APPROACH AND ITS INTEGRITYTeng-yao Yang (8742285) 23 April 2020 (has links)
<div>This research investigates the detection and exclusion of faulty global navigation satellite system (GNSS) measurements using a parameterized quadratic programming formulation (PQP) approach. Furthermore, the PQP approach is integrated with the integrity risk and continuity risk bounds of the Chi-squared advanced receiver autonomous integrity monitoring (ARAIM). The integration allows for performance evaluation of the PQP approach in terms of accuracy, integrity, continuity, and availability, which is necessary for the PQP approach to be applied to the vertical navigation in the performance-based navigation (PBN). In the case of detection, the PQP approach can also be integrated with the vertical protection level and the associated lower and upper bounds derived for the solution separation ARAIM. While there are other computationally efficient and less computationally efficient fault detection and</div><div>exclusion methods to detect and exclude faulty GNSS measurements, the strength of the PQP approach can summarized from two different perspectives. Firstly, the PQP</div><div>approach belongs to the group of the computationally efficient methods, which makes the PQP approach more favorable when it comes to detect and exclude multiple simultaneous faulty GNSS measurements. Secondly, because of the integration of the PQP approach with the integrity risk and continuity risk bounds of the Chi-squared</div><div>ARAIM, the PQP approach is among the first computationally efficient fault detection and exclusion methods to incorporate the concept of integrity, which lies in</div><div>the foundation of PBN. Despite the PQP approach not being a practical integrity monitoring method in its current form because of the combinatorial natural of the integrity risk bound calculation and the rather conservative integrity performance, further research can be pursued to improve the PQP approach. Any improvement on the integrity risk bound calculation for the Chi-squared ARAIM can readily be</div><div>applied to the integrity risk bound calculation for the PQP approach. Also, the connection between the PQP approach and the support vector machines, the application of the extreme value theory to obtain a conservative tail probability may shed light upon the parameter tuning of the PQP approach, which in turn will result in tight integrity risk bound.</div>
|
69 |
Identifying symptoms of fault in District Heating Substations : An investigation in how a predictive heat load software can help with fault detectionBergentz, Tobias January 2020 (has links)
District heating delivers more than 70% of the energy used for heating and domestichot water in Swedish buildings. To stay competitive, district heating needs toreduce its losses and increase capabilities to utilise low grade heat. Finding faultysubstations is one way to allow reductions in supply temperatures in district heatingnetworks, which in turn can help reduce the losses. In this work three suggestedsymptoms of faults: abnormal quantization, drifting and anomalous values, are investigatedwith the help of hourly meter data of: heat load, volume flow, supplyand return temperatures from district heating substations. To identify abnormalquantization, a method is proposed based on Shannon’s entropy, where lower entropysuggests higher risk of abnormal quantization. The majority of the substationsidentified as having abnormal quantization with the proposed method has a meterresolution lower than the majority of the substations in the investigated districtheating network. This lower resolution is likely responsible for identifying thesesubstation, suggesting the method is limited by the meter resolution of the availabledata. To improve result from the method higher resolution and sampling frequencyis likely needed.For identifying drift and anomalous values two methods are proposed, one for eachsymptom. Both methods utilize a software for predicting hourly heat load, volumeflow, supply and return temperatures in individual district heating substations.The method suggested for identifying drift uses the mean value of each predictedand measured quantity during the investigated period. The mean of the prediction iscompared to the mean of the measured values and a large difference would suggestrisk of drift. However this method has not been evaluated due to difficulties infinding a suitable validation method.The proposed method for detecting anomalous values is based on finding anomalousresiduals when comparing the prediction from the prediction software to themeasured values. To find the anomalous residuals the method uses an anomalydetection algorithm called IsolationForest. The method produces rankable lists inwhich substations with risk of anomalies are ranked higher in the lists. Four differentlists where evaluated by an experts. For the two best preforming lists approximatelyhalf of the top 15 substations where classified to contain anomalies by the expertgroup. The proposed method for detecting anomalous values shows promising resultespecially considering how easily the method could be added to a district heatingnetwork. Future work will focus on reducing the number of false positives. Suggestionsfor lowering the false positive rate include, alternations or checks on theprediction models used.
|
70 |
Detection of Cardiovascular Anomalies: An Observer-Based ApproachLedezma, Fernando 07 1900 (has links)
In this thesis, a methodology for the detection of anomalies in the cardiovascular
system is presented. The cardiovascular system is one of the most fascinating and
complex physiological systems. Nowadays, cardiovascular diseases constitute one of
the most important causes of mortality in the world. For instance, an estimate of 17.3
million people died in 2008 from cardiovascular diseases. Therefore, many studies have
been devoted to modeling the cardiovascular system in order to better understand its
behavior and find new reliable diagnosis techniques.
The lumped parameter model of the cardiovascular system proposed in [1] is restructured
using a hybrid systems approach in order to include a discrete input vector
that represents the influence of the mitral and aortic valves in the different phases of
the cardiac cycle. Parting from this model, a Taylor expansion around the nominal
values of a vector of parameters is conducted. This expansion serves as the foundation
for a component fault detection process to detect changes in the physiological
parameters of the cardiovascular system which could be associated with cardiovascular
anomalies such as atherosclerosis, aneurysm, high blood pressure, etc. An Extended Kalman Filter is used in order to achieve a joint estimation of the state vector and
the changes in the considered parameters. Finally, a bank of filters is, as in [2], used
in order to detect the appearance of heart valve diseases, particularly stenosis and
regurgitation. The first numerical results obtained are presented.
|
Page generated in 0.0292 seconds