Mehrdad Saif | University of Windsor (original) (raw)

Papers by Mehrdad Saif

Research paper thumbnail of Ensemble-Based Fault Detection and Isolation of an Industrial Gas Turbine

2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2020

Research paper thumbnail of Adaptive Fault Detection for a Class of Nonlinear Systems Based on Output Estimator Design

IFAC Proceedings Volumes, 2008

Research paper thumbnail of Actuator Fault Diagnosis Using High-Order Sliding Mode Differentiator (HOSMD) and Its Application to a Laboratory 3D Crane

IFAC Proceedings Volumes, 2008

Research paper thumbnail of Robust Fault Detection and Isolation in Constrained Nonlinear Systems via a Second Order Sliding Mode Observer

IFAC Proceedings Volumes, 2002

Research paper thumbnail of An actuator fault isolation strategy for linear and nonlinear systems

Proceedings of the 2005, American Control Conference, 2005.

Research paper thumbnail of A New Compressor Failure Prognostic Method Using Nonlinear Observers and a Bayesian Algorithm for Heavy-Duty Gas Turbines

IEEE Sensors Journal

Failure prognostic predicts the remaining useful life (RUL) of machine/ components, which will al... more Failure prognostic predicts the remaining useful life (RUL) of machine/ components, which will allow timely maintenance and repair leading to continuous reliable and safe operating conditions. In this article, a novel hybrid RUL prediction approach is proposed for heavy-duty gas turbines. Two common failures, namely the fouling in the gas turbine compressor and filter defect, are investigated.First, a discrete wavelet transform(DWT) is applied to real-time measurements to reduce the effect of noise. A parallel structure consisting of a Laguerre filter and neuro-fuzzy is then constructed to identify nonlinear failure dynamics and generate residuals. These residuals are then utilized to estimate the failure severity. Following that, Bayesian theory is employed to predict the RUL. A novel feature of the approach is that the Laguerre filter is designed by using orthogonal basis functions (OBFs), which deliver precise estimates. Another benefit is that the proposed parallel configuration accurately identifies failure dynamics and boosts the RUL prediction performance. Experimental test studies on heavy-duty gas turbines indicate the high efficiency of the proposed RUL estimation in comparison to other failure prognostic strategies.

Research paper thumbnail of SMS–A Security Management System for Steam Turbines Using a Multisensor Array

IEEE Systems Journal, 2020

Cyber-physical systems, such as large power plants, apply open networks for monitoring and contro... more Cyber-physical systems, such as large power plants, apply open networks for monitoring and control purposes. This may increase the risk of cyberattacks to these infrastructures. Cybersecurity methods have been employed as promising techniques to deal with cyber threats and isolate possible cyberattacks. This article introduces a new security management system (SMS) for an industrial steam turbine. As such, the most probable threats, such as denial-of-service (DoS) attack, deception attack, and replay attack in various sensors and actuators of the steam turbine system, are considered. Then, a new SMS system consisting of an attack detection unit and an attack isolation unit is designed. The attack detection unit utilizes a dynamic neural network to detect any potential attack in the system using the concept of residual generation. The attack isolation unit identifies the type of attacks using an integrated feature selection strategy and support vector machine classifier through multisensor information. Several case studies are investigated to evaluate the proposed SMS. The test results show the effectiveness of the proposed SMS with multisensor information when compared to SMS without multisensor array.

Research paper thumbnail of An Adaptive Passive Fault Tolerant Control System for a Steam Turbine Using a PCA Based Inverse Neural Network Control Strategy

2018 World Automation Congress (WAC), 2018

Fault tolerant control (FTC) becomes an effective way to defectively control a plant and ensure r... more Fault tolerant control (FTC) becomes an effective way to defectively control a plant and ensure reliability and safety in the system. This paper presents a new adaptive passive fault tolerant control (FTC) methodology based on inverse control strategy. An adaptive principal component analysis (PCA) algorithm is incorporated as a pretreatment data processing to recursively capture inherent time-varying information embedded in the plant time-series measurements. A multi-layered perceptron (MLP) neural network is then trained online with the reduced PCA extracted features to emulate an adaptive inverse controller based on actual post-fault plant dynamic model. The adaptive MLP-based controller will be able to minimize induced tracking error using an error back propagation (BP) learning algorithm without a priori knowledge of the occurred faults on the basis of the PCA-uncorrelated measurement data. This enhances the generalization capability of the realized controller due to distinctiveness of the PCA-based data representation. An extensive set of test scenarios has been considered to explore effectiveness of the proposed FTC scheme against three major faults in an industrial steam turbine benchmark. The results demonstrate promising capability of the proposed FTC to automatically maintain the steam turbine availability with efficient fault accommodation.

Research paper thumbnail of A new clustering method using wavelet based probability density functions for identifying patterns in time-series data

2016 IEEE EMBS International Student Conference (ISC), 2016

Clustering is a prominent method to identify similar patterns in large groups of data and can be ... more Clustering is a prominent method to identify similar patterns in large groups of data and can be beneficial in the bioinformatics studies due to this property. Classical methods such as k-means and maximum likelihood consider a mixture of Gaussian probability density function (PDF) of data and find clusters based on maximizing the PDF. However, correlation among different groups of data and existence of noise on the data make it difficult to correctly detect the correct number of clusters. Furthermore, the assumption of the Gaussian distance for the PDF is not necessarily true in real applications. This paper presents a new clustering method via wavelet-based probability density functions. For this purpose, first, a mixture of PDFs is estimated by the wavelet for each feature. After this, a multilevel thresholding method is implemented on the mixture of PDFs of each feature to obtain the clusters. Finally, a forward feature selection with memory is used to cluster the dataset based on combinations of the features. The profile alignment and agglomerative clustering (PAAC) index is applied for evaluating the number of clusters and features. Transcript expression throughout the various stages of prostate cancer is considered as a case study to identify patterns. The experimental results show the ability of the proposed method in detecting patterns of similar transcripts throughout disease progression. The results are promising in comparison with the other methods.

Research paper thumbnail of Multiple Fault Detection and Isolation in DC-DC Converters

Research paper thumbnail of Failure Diagnosis of Wind Turbine Bearing Using Feature Extraction and a Neuro-Fuzzy Inference System (ANFIS)

Advances in Computational Intelligence, 2019

Bearing failures are the most common type of malfunction in wind turbines. As such, isolating the... more Bearing failures are the most common type of malfunction in wind turbines. As such, isolating these defects enables maintenance scheduling in advance; hence, preventing further damage to turbines. This paper introduces a new fault detection and diagnosis (FDD) method to isolate two types of bearing failures in Wind turbines (WTs). The proposed FDD method consists of a feature extraction/feature selection and an adaptive neuro-fuzzy inference system (ANFIS) method. The feature extraction and selection phase identifies proper features to capture the nonlinear dynamics of the failure. Then, the ANFIS classifier diagnoses the failure type using the extracted features. Several experimental test studies with the historical data of wind farms in South-western Ontario are performed to evaluate the performance of the FDD system. Test results indicate that the proposed monitoring system is accurate and effective.

Research paper thumbnail of A Control Oriented Cyber-Secure Strategy Based on Multiple Sensor Fusion

2019 IEEE International Conference on Systems, Man and Cybernetics (SMC), 2019

This paper introduces a cyber-secure strategy for radar tracking systems. Two common cyber attack... more This paper introduces a cyber-secure strategy for radar tracking systems. Two common cyber attacks including denial-of-service (DoS) and false data injection (deception) attacks are investigated. The proposed secure control strategy consists of two subsystems: 1) an attack detection and isolation (ADI) subsystem, and 2) a resilient observer (RO) subsystem. The ADI subsystem is used to observe the state of the system using a bank of Kalman Filters and multi-sensor measurements. Then, residuals generated by local Kalman filters are used to isolate the cyber attacks. Afterward, ordered weighted averaging (OWA) operator is utilized to drive a resilient observer to estimate the real correct value of variables such as position under cyber attacks. Weighting factors of the OWA operator are derived using the covariance matrix, and proof of convergence is provided. Simulation studies on a radar tracking system show that the proposed secure control strategy using multi-sensor fusion enhances the performance of the system and results in a more resilient control system against cyber attacks.

Research paper thumbnail of Data fusion for fault diagnosis in smart grid power systems

2017 IEEE 30th Canadian Conference on Electrical and Computer Engineering (CCECE), 2017

In smart grid power systems, fast and accurate fault detection and diagnosis (FDD) is vital for i... more In smart grid power systems, fast and accurate fault detection and diagnosis (FDD) is vital for isolating faulty components and avoiding further complications. This paper introduces a new data fusion method based on ordered weighted averaging (OWA) operator for power smart grids. For this purpose, the discrete time data from circuit breakers (CB) is combined with continuous time data of recorders to enhance the reliability of the fault diagnosis approach. Radial basis functions (RBF) artificial neural network and wavelet transform (WT) are individually employed to identify the location of the fault from the continuous voltage of the buses. Then, a combination of these two methods along with the information from CBs are utilized into a unique framework by OWA operator to diagnose the faults at an early stage. IEEE standard 14 bus system is used to illustrate and validate the proposed method. Several phase to ground faults are injected into the simulation model to validate the diagnostic capability of the FDD system. Simulation results show a better performance of the fusion FDD system in comparison with three other methods.

Research paper thumbnail of Planetary Gear Faults Detection in Wind Turbine Gearbox Based on a Ten Years Historical Data From Three Wind Farms

Research paper thumbnail of A New Hybrid Fault Prognosis Method for MFS Systems Based on Distributed Neural Networks and Recursive Bayesian Algorithm

IEEE Systems Journal, 2020

This article introduces a new hybrid prognosis method to predict a remaining useful lifetime (RUL... more This article introduces a new hybrid prognosis method to predict a remaining useful lifetime (RUL) of multi-functional spoiler (MFS) systems. The MFS is vital to the healthy operation of aircraft spoiler control systems, and any fault or failure in these systems could compromise the safe operation of the aircraft. The proposed prognosis methodology is a hybrid framework composed of a failure parameter estimation unit and an RUL unit. The failure parameter estimation unit observes the failure parameters using distributed neural networks via available measurements of the MFS system. Simultaneously, the remaining useful life is anticipated by the RUL unit employing the estimated failure parameter with a recursive Bayesian algorithm. Moreover, a relative accuracy (RA) measure is invoked to validate the effectiveness of the proposed method. Simulink model of the MFS system is verified by experimental data of the LJ200 series aircraft under fight condition. Furthermore, simulation test results indicate a high accuracy of the distributed structure compared to a centralized network.

Research paper thumbnail of Improved Estimation for Well-Logging Problems Based on Fusion of Four Types of Kalman Filters

IEEE Transactions on Geoscience and Remote Sensing, 2018

Research paper thumbnail of A New Fusion Estimation Method for Multi-Rate Multi-Sensor Systems With Missing Measurements

Research paper thumbnail of A New Hybrid Fault Detection Method for Wind Turbine Blades Using Recursive PCA and Wavelet-Based PDF

IEEE Sensors Journal, 2019

Research paper thumbnail of Novel Multiagent Model-Predictive Control Performance Indices for Monitoring of a Large-Scale Distributed Water System

IEEE Systems Journal, 2017

Research paper thumbnail of A non-iterative LMI based PID power system stabilizer

2016 World Automation Congress (WAC), 2016

This paper presents a new method of designing a robust PID power system stabilizer. This methodol... more This paper presents a new method of designing a robust PID power system stabilizer. This methodology provides an exact way to tune PID parameters and find an optimal controller using non-iterative Linear Matrix Inequality (LMI) approach. The uncertainties inherent in the system model is also taken into account in the design process to increase the robustness of the proposed controller. For this purpose, H∞ control theory is employed in a LMI framework to design a PID controller that damp oscillations in power system and makes it robust against uncertainty. The obtained matrix inequality is nonlinear which is converted to a LMI in the proposed framework. The Simulation results show the superior performance of the controller compared to conventional method in tuning of PID controller.

Research paper thumbnail of Ensemble-Based Fault Detection and Isolation of an Industrial Gas Turbine

2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2020

Research paper thumbnail of Adaptive Fault Detection for a Class of Nonlinear Systems Based on Output Estimator Design

IFAC Proceedings Volumes, 2008

Research paper thumbnail of Actuator Fault Diagnosis Using High-Order Sliding Mode Differentiator (HOSMD) and Its Application to a Laboratory 3D Crane

IFAC Proceedings Volumes, 2008

Research paper thumbnail of Robust Fault Detection and Isolation in Constrained Nonlinear Systems via a Second Order Sliding Mode Observer

IFAC Proceedings Volumes, 2002

Research paper thumbnail of An actuator fault isolation strategy for linear and nonlinear systems

Proceedings of the 2005, American Control Conference, 2005.

Research paper thumbnail of A New Compressor Failure Prognostic Method Using Nonlinear Observers and a Bayesian Algorithm for Heavy-Duty Gas Turbines

IEEE Sensors Journal

Failure prognostic predicts the remaining useful life (RUL) of machine/ components, which will al... more Failure prognostic predicts the remaining useful life (RUL) of machine/ components, which will allow timely maintenance and repair leading to continuous reliable and safe operating conditions. In this article, a novel hybrid RUL prediction approach is proposed for heavy-duty gas turbines. Two common failures, namely the fouling in the gas turbine compressor and filter defect, are investigated.First, a discrete wavelet transform(DWT) is applied to real-time measurements to reduce the effect of noise. A parallel structure consisting of a Laguerre filter and neuro-fuzzy is then constructed to identify nonlinear failure dynamics and generate residuals. These residuals are then utilized to estimate the failure severity. Following that, Bayesian theory is employed to predict the RUL. A novel feature of the approach is that the Laguerre filter is designed by using orthogonal basis functions (OBFs), which deliver precise estimates. Another benefit is that the proposed parallel configuration accurately identifies failure dynamics and boosts the RUL prediction performance. Experimental test studies on heavy-duty gas turbines indicate the high efficiency of the proposed RUL estimation in comparison to other failure prognostic strategies.

Research paper thumbnail of SMS–A Security Management System for Steam Turbines Using a Multisensor Array

IEEE Systems Journal, 2020

Cyber-physical systems, such as large power plants, apply open networks for monitoring and contro... more Cyber-physical systems, such as large power plants, apply open networks for monitoring and control purposes. This may increase the risk of cyberattacks to these infrastructures. Cybersecurity methods have been employed as promising techniques to deal with cyber threats and isolate possible cyberattacks. This article introduces a new security management system (SMS) for an industrial steam turbine. As such, the most probable threats, such as denial-of-service (DoS) attack, deception attack, and replay attack in various sensors and actuators of the steam turbine system, are considered. Then, a new SMS system consisting of an attack detection unit and an attack isolation unit is designed. The attack detection unit utilizes a dynamic neural network to detect any potential attack in the system using the concept of residual generation. The attack isolation unit identifies the type of attacks using an integrated feature selection strategy and support vector machine classifier through multisensor information. Several case studies are investigated to evaluate the proposed SMS. The test results show the effectiveness of the proposed SMS with multisensor information when compared to SMS without multisensor array.

Research paper thumbnail of An Adaptive Passive Fault Tolerant Control System for a Steam Turbine Using a PCA Based Inverse Neural Network Control Strategy

2018 World Automation Congress (WAC), 2018

Fault tolerant control (FTC) becomes an effective way to defectively control a plant and ensure r... more Fault tolerant control (FTC) becomes an effective way to defectively control a plant and ensure reliability and safety in the system. This paper presents a new adaptive passive fault tolerant control (FTC) methodology based on inverse control strategy. An adaptive principal component analysis (PCA) algorithm is incorporated as a pretreatment data processing to recursively capture inherent time-varying information embedded in the plant time-series measurements. A multi-layered perceptron (MLP) neural network is then trained online with the reduced PCA extracted features to emulate an adaptive inverse controller based on actual post-fault plant dynamic model. The adaptive MLP-based controller will be able to minimize induced tracking error using an error back propagation (BP) learning algorithm without a priori knowledge of the occurred faults on the basis of the PCA-uncorrelated measurement data. This enhances the generalization capability of the realized controller due to distinctiveness of the PCA-based data representation. An extensive set of test scenarios has been considered to explore effectiveness of the proposed FTC scheme against three major faults in an industrial steam turbine benchmark. The results demonstrate promising capability of the proposed FTC to automatically maintain the steam turbine availability with efficient fault accommodation.

Research paper thumbnail of A new clustering method using wavelet based probability density functions for identifying patterns in time-series data

2016 IEEE EMBS International Student Conference (ISC), 2016

Clustering is a prominent method to identify similar patterns in large groups of data and can be ... more Clustering is a prominent method to identify similar patterns in large groups of data and can be beneficial in the bioinformatics studies due to this property. Classical methods such as k-means and maximum likelihood consider a mixture of Gaussian probability density function (PDF) of data and find clusters based on maximizing the PDF. However, correlation among different groups of data and existence of noise on the data make it difficult to correctly detect the correct number of clusters. Furthermore, the assumption of the Gaussian distance for the PDF is not necessarily true in real applications. This paper presents a new clustering method via wavelet-based probability density functions. For this purpose, first, a mixture of PDFs is estimated by the wavelet for each feature. After this, a multilevel thresholding method is implemented on the mixture of PDFs of each feature to obtain the clusters. Finally, a forward feature selection with memory is used to cluster the dataset based on combinations of the features. The profile alignment and agglomerative clustering (PAAC) index is applied for evaluating the number of clusters and features. Transcript expression throughout the various stages of prostate cancer is considered as a case study to identify patterns. The experimental results show the ability of the proposed method in detecting patterns of similar transcripts throughout disease progression. The results are promising in comparison with the other methods.

Research paper thumbnail of Multiple Fault Detection and Isolation in DC-DC Converters

Research paper thumbnail of Failure Diagnosis of Wind Turbine Bearing Using Feature Extraction and a Neuro-Fuzzy Inference System (ANFIS)

Advances in Computational Intelligence, 2019

Bearing failures are the most common type of malfunction in wind turbines. As such, isolating the... more Bearing failures are the most common type of malfunction in wind turbines. As such, isolating these defects enables maintenance scheduling in advance; hence, preventing further damage to turbines. This paper introduces a new fault detection and diagnosis (FDD) method to isolate two types of bearing failures in Wind turbines (WTs). The proposed FDD method consists of a feature extraction/feature selection and an adaptive neuro-fuzzy inference system (ANFIS) method. The feature extraction and selection phase identifies proper features to capture the nonlinear dynamics of the failure. Then, the ANFIS classifier diagnoses the failure type using the extracted features. Several experimental test studies with the historical data of wind farms in South-western Ontario are performed to evaluate the performance of the FDD system. Test results indicate that the proposed monitoring system is accurate and effective.

Research paper thumbnail of A Control Oriented Cyber-Secure Strategy Based on Multiple Sensor Fusion

2019 IEEE International Conference on Systems, Man and Cybernetics (SMC), 2019

This paper introduces a cyber-secure strategy for radar tracking systems. Two common cyber attack... more This paper introduces a cyber-secure strategy for radar tracking systems. Two common cyber attacks including denial-of-service (DoS) and false data injection (deception) attacks are investigated. The proposed secure control strategy consists of two subsystems: 1) an attack detection and isolation (ADI) subsystem, and 2) a resilient observer (RO) subsystem. The ADI subsystem is used to observe the state of the system using a bank of Kalman Filters and multi-sensor measurements. Then, residuals generated by local Kalman filters are used to isolate the cyber attacks. Afterward, ordered weighted averaging (OWA) operator is utilized to drive a resilient observer to estimate the real correct value of variables such as position under cyber attacks. Weighting factors of the OWA operator are derived using the covariance matrix, and proof of convergence is provided. Simulation studies on a radar tracking system show that the proposed secure control strategy using multi-sensor fusion enhances the performance of the system and results in a more resilient control system against cyber attacks.

Research paper thumbnail of Data fusion for fault diagnosis in smart grid power systems

2017 IEEE 30th Canadian Conference on Electrical and Computer Engineering (CCECE), 2017

In smart grid power systems, fast and accurate fault detection and diagnosis (FDD) is vital for i... more In smart grid power systems, fast and accurate fault detection and diagnosis (FDD) is vital for isolating faulty components and avoiding further complications. This paper introduces a new data fusion method based on ordered weighted averaging (OWA) operator for power smart grids. For this purpose, the discrete time data from circuit breakers (CB) is combined with continuous time data of recorders to enhance the reliability of the fault diagnosis approach. Radial basis functions (RBF) artificial neural network and wavelet transform (WT) are individually employed to identify the location of the fault from the continuous voltage of the buses. Then, a combination of these two methods along with the information from CBs are utilized into a unique framework by OWA operator to diagnose the faults at an early stage. IEEE standard 14 bus system is used to illustrate and validate the proposed method. Several phase to ground faults are injected into the simulation model to validate the diagnostic capability of the FDD system. Simulation results show a better performance of the fusion FDD system in comparison with three other methods.

Research paper thumbnail of Planetary Gear Faults Detection in Wind Turbine Gearbox Based on a Ten Years Historical Data From Three Wind Farms

Research paper thumbnail of A New Hybrid Fault Prognosis Method for MFS Systems Based on Distributed Neural Networks and Recursive Bayesian Algorithm

IEEE Systems Journal, 2020

This article introduces a new hybrid prognosis method to predict a remaining useful lifetime (RUL... more This article introduces a new hybrid prognosis method to predict a remaining useful lifetime (RUL) of multi-functional spoiler (MFS) systems. The MFS is vital to the healthy operation of aircraft spoiler control systems, and any fault or failure in these systems could compromise the safe operation of the aircraft. The proposed prognosis methodology is a hybrid framework composed of a failure parameter estimation unit and an RUL unit. The failure parameter estimation unit observes the failure parameters using distributed neural networks via available measurements of the MFS system. Simultaneously, the remaining useful life is anticipated by the RUL unit employing the estimated failure parameter with a recursive Bayesian algorithm. Moreover, a relative accuracy (RA) measure is invoked to validate the effectiveness of the proposed method. Simulink model of the MFS system is verified by experimental data of the LJ200 series aircraft under fight condition. Furthermore, simulation test results indicate a high accuracy of the distributed structure compared to a centralized network.

Research paper thumbnail of Improved Estimation for Well-Logging Problems Based on Fusion of Four Types of Kalman Filters

IEEE Transactions on Geoscience and Remote Sensing, 2018

Research paper thumbnail of A New Fusion Estimation Method for Multi-Rate Multi-Sensor Systems With Missing Measurements

Research paper thumbnail of A New Hybrid Fault Detection Method for Wind Turbine Blades Using Recursive PCA and Wavelet-Based PDF

IEEE Sensors Journal, 2019

Research paper thumbnail of Novel Multiagent Model-Predictive Control Performance Indices for Monitoring of a Large-Scale Distributed Water System

IEEE Systems Journal, 2017

Research paper thumbnail of A non-iterative LMI based PID power system stabilizer

2016 World Automation Congress (WAC), 2016

This paper presents a new method of designing a robust PID power system stabilizer. This methodol... more This paper presents a new method of designing a robust PID power system stabilizer. This methodology provides an exact way to tune PID parameters and find an optimal controller using non-iterative Linear Matrix Inequality (LMI) approach. The uncertainties inherent in the system model is also taken into account in the design process to increase the robustness of the proposed controller. For this purpose, H∞ control theory is employed in a LMI framework to design a PID controller that damp oscillations in power system and makes it robust against uncertainty. The obtained matrix inequality is nonlinear which is converted to a LMI in the proposed framework. The Simulation results show the superior performance of the controller compared to conventional method in tuning of PID controller.

Research paper thumbnail of A Control Oriented Cyber-Secure Strategy Based on Multiple Sensor Fusion

SMC Conference 2019, 2019

This paper introduces a cyber-secure strategy for radar tracking systems. Two common cyber attack... more This paper introduces a cyber-secure strategy for radar tracking systems. Two common cyber attacks including denial-of-service (DoS) and false data injection (deception) attacks are investigated. The proposed secure control strategy consists of two subsystems: 1) an attack detection and isolation (ADI) subsystem, and 2) a resilient observer (RO) subsystem. The ADI subsystem is used to observe the state of the system using a bank of Kalman Filters and multi-sensor measurements. Then, residuals generated by local Kalman filters are used to isolate the cyber attacks. Afterward, ordered weighted averaging
(OWA) operator is utilized to drive a resilient observer to estimate the real correct value of variables such as position under cyber attacks. Weighting factors of the OWA operator are derived using the covariance matrix, and proof of convergence is provided. Simulation studies on a radar tracking system show that the proposed secure control strategy using multi-sensor fusion enhances the performance of the system and results in a more resilient control system against cyber attacks.