Marzieh kamali | Isfahan University of Technology (original) (raw)
Papers by Marzieh kamali
International Journal of Automation and Computing, Dec 5, 2016
This paper presents a robust adaptive state feedback control scheme for a class of parametric-str... more This paper presents a robust adaptive state feedback control scheme for a class of parametric-strict-feedback nonlinear systems in the presence of time varying actuator failures. The designed adaptive controller compensates a general class of actuator failures without any need for explicit fault detection. The parameters, times, and patterns of the considered failures are completely unknown. The proposed controller is constructed based on a backstepping design method. The global boundedness of all the closed-loop signals is guaranteed and the tracking error is proved to converge to a small neighborhood of the origin. The proposed approach is employed for a two-axis positioning stage system as well as an aircraft wing system. The simulation results show the correctness and effectiveness of the proposed robust adaptive actuator failure compensation approach.
2022 30th International Conference on Electrical Engineering (ICEE)
Additional file 1: Interaction list for drug–target. The raw data used in the drug–target is incl... more Additional file 1: Interaction list for drug–target. The raw data used in the drug–target is included in this file. Drug–target interactions were extracted from DrugBank database
Additional file 6: Impact of the encoder. Removing the encoder significantly declines the perform... more Additional file 6: Impact of the encoder. Removing the encoder significantly declines the performance of DIDL
Additional file 5: Evaluation of DIDL with Hetionet network. Performance of DIDL for prediction o... more Additional file 5: Evaluation of DIDL with Hetionet network. Performance of DIDL for prediction of interactions between different layers of Hetionet network is assessed using 10 fold cross-validation
Additional file 3: Interaction list for TF–DNA. The raw data used in the TF–DNA is included in th... more Additional file 3: Interaction list for TF–DNA. The raw data used in the TF–DNA is included in this file. Interactions were extracted from the Enrichr database using ChEA 2016
2017 Iranian Conference on Electrical Engineering (ICEE), 2017
This paper offers a robust adaptive control based dynamic surface control (DSC) for uncertain non... more This paper offers a robust adaptive control based dynamic surface control (DSC) for uncertain nonlinear systems in which the unknown nonlinearities are not linearly parameterized with respect to uncertain parameters. Therefore, radial basis function (RBF) neural networks (NNs) are used for approximating uncertain nonlinearities. The proposed controller guarantees uniformly ultimately boundedness of the closed loop system and guarantees much higher tracking accuracy compared with the previous works in backstepping and DSC methods. Simulation results are presented to show the effectiveness of the proposed approach.
The main purpose of this paper is to present an adaptive-neural controller for strictfeedback non... more The main purpose of this paper is to present an adaptive-neural controller for strictfeedback nonlinear systems with unknown time delays and in the presence of external disturbances and actuator failure. The proposed adaptive-neural controller is constructed based on DSC design technique. Radial Basis Functions (RBF) networks are utilized to approximate unknown nonlinear functions. Adaptive rules are obtained based on Lyapunov design for updating the parameters of neural networks. Disturbances are unknown functions which their bounds are partially known. Therefore, continuous robust terms are applied in order to minimize their effects. Furthermore, due to the existence of unknown time delays in the system, Lyapunov– Krasovskii functionals are utilized in the process of designing the controller and proofing the stability of the system. In addition, the controller is designed so that it can compensate its effect if the considered actuator failure happens. For the designed controller, ...
IEEE Transactions on Smart Grid, 2021
State estimation has a special role in the real-time control and monitoring of smart distribution... more State estimation has a special role in the real-time control and monitoring of smart distribution networks. State estimation process is typically based on network topology and measurements sent from meters. Employing an accurate state estimation algorithm as well as transferring high volumes of measurements are serious challenges in large scale grids. In this paper, compressive sensing is used to reduce the measurement data volume, before transmission, to alleviate problems such as lack of storage space, interference and delay. In this paper, a modified extended Kalman filter algorithm is proposed which estimates states from compressed data directly without applying the reconstruction procedure. The main differences between the proposed method and EKF are the network dynamic modeling approach and the states correction mechanism. The IEEE 33-node distribution network with two DGs is employed to illustrate the effective performance of the proposed method. Results show that the states of the test feeder are accurately estimated even with only 50% compressed measurements.
2019 27th Iranian Conference on Electrical Engineering (ICEE), 2019
Recently, consensus tracking control of a higher order multi agent nonlinear system has become an... more Recently, consensus tracking control of a higher order multi agent nonlinear system has become an interesting problem. In order to solve the explosion of design complexity, in this paper the Dynamic Surface Control (DSC) approach is used to present a new consensus method for an unknown multi agent high order nonlinear systems with uncertain disturbances. Compared to previous methods, the computational burden of the distributed adaptive control scheme is significantly reduced and a simpler distributed adaptive control scheme is designed. Moreover, unknown multi agent nonlinear system is estimated without using fuzzy logic systems or neural networks, thus the number of tuned parameters is significantly reduced. Finally, the global uniform boundedness of all the closed-loop signals and the convergence of the distributed consensus tracking errors to a small boundedness of the origin are proved. Simulation results verified that the proposed controller is effective and appropriate.
2022 30th International Conference on Electrical Engineering (ICEE)
2019 6th International Conference on Control, Instrumentation and Automation (ICCIA)
This paper offers a robust adaptive control based dynamic surface control for uncertain nonlinear... more This paper offers a robust adaptive control based dynamic surface control for uncertain nonlinear systems in the presence of actuator saturation and external disturbance. Radial Basis Function Neural Networks are used for approximating uncertain nonlinearities. Using dynamic surface control technique, the problem of explosion of complexity caused by differentiating the virtual controllers during the recursive procedures in the conventional backstepping method is avoided. The proposed controller guarantees uniformly ultimately boundedness of the closed loop system. The effectiveness of the proposed method is illustrated with a simulation example.
Robotica, 2021
In this paper, new distributed adaptive methods are proposed for solving both leaderless and lead... more In this paper, new distributed adaptive methods are proposed for solving both leaderless and leader–follower consensus problems in networks of uncertain robot manipulators, by estimating only the gravitational torque forces. Comparing with the existing adaptive methods, which require the estimation of the whole dynamics, presented methods reduce the excitation levels required for efficient parameter search, the convergence time, and the complexity of the regressor. Additionally, proposed schemes eliminate the need for velocity information exchange between the agents. Global asymptotic synchronization is shown by introducing new Lyapunov functions. Simulation results are provided for a network of 10 4-DOF robot manipulators.
Consensus is the most basic synchronization behavior in multiagent systems. For networks of Euler... more Consensus is the most basic synchronization behavior in multiagent systems. For networks of Euler-Lagrange (EL) agents different controllers have been proposed to achieve consensus, requiring in all cases, either the cancellation or the estimation of the gravity forces. In the latter case, it is necessary to estimate, not just the gravity forces, but the parameters of the whole dynamics. This requires the computation of a complicated regressor matrix, that grows in complexity as the degrees-of-freedom of the EL-agents increase. In this paper, we propose an adaptive controllers to solve the leaderless consensus problem by only estimating the gravitational term of the agents and hence without requiring the complete regressor matrix. To the best of our knowledge, this is the first work that achieves such an objective. The controller is a simple Proportional plus damping (P+d) scheme that does not require to exchange velocity information between the agents. Simulation results demonstrat...
IET Control Theory & Applications, 2017
A tube-based robust model predictive control (MPC) is proposed to be applied in constrained linea... more A tube-based robust model predictive control (MPC) is proposed to be applied in constrained linear systems with parametric uncertainty. An estimation method is applied in this proposed technique to adapt the system model at each sampling time and to reduce the conservatism nature of the tube-based MPC as the system model approaches the real model as time passes. By updating the subject model online through this newly proposed approach the performance of the system is improved. Asymptotic stability of the closed-loop system is established. The simulation results of a DC motor are applied to illustrate the effectiveness of this proposed controller in dealing with one practical system. 2 Problem description Consider a linear system in standard state-space form x k + 1 = Ax k + Bu k (1)
Neurocomputing, 2016
This paper investigates the consensus tracking problem of nonlinear multi-agent systems with stat... more This paper investigates the consensus tracking problem of nonlinear multi-agent systems with state constraints and unknown disturbances. An observer is presented for the case that the states of each follower and its neighbors are unmeasurable. State constraint for multi-agent systems is a challenging problem. Barrier Lyapunov functions are applied in this paper to deal with this difficulty. Then based on adaptive back-stepping control approach and dynamic surface control technique, an adaptive fuzzy distributed controller is proposed to guarantee that the tracking errors between all followers and the leader converge to a small neighborhood of the origin. Moreover, it is proved that all the signals in the multi-agent systems are semi-globally uniformly ultimately bounded (SUUB). Finally, some numerical simulation results are presented to testify the effectiveness of the proposed algorithm.
Despite enormous achievements in production of high throughput datasets, constructing comprehensi... more Despite enormous achievements in production of high throughput datasets, constructing comprehensive maps of interactions remains a major challenge. The lack of sufficient experimental evidence on interactions is more significant for heterogeneous molecular types. Hence, developing strategies to predict inter-omics connections is essential to construct holistic maps of disease. Here, Data Integration with Deep Learning (DIDL), a novel nonlinear deep learning method is proposed to predict inter-omics interactions. It consists of an encoder that automatically extracts features for biomolecules according to existing interactions, and a decoder that predicts novel interactions. The applicability of DIDL is assessed with different networks namely drug-target protein, transcription factor-DNA element, and miRNA-mRNA. Also, the validity of novel predictions is assessed by literature surveys. Furthermore, DIDL outperformed state-of-the-art methods. Area under the curve, and area under the pr...
ABSTRACT In this paper a new design method is presented for reconfigurable fault tolerant control... more ABSTRACT In this paper a new design method is presented for reconfigurable fault tolerant control of multivariable nonlinear time delay systems. The control reconfiguration algorithm comprises two main parts: fault detection and diagnosis and model reference adaptive control. First, a combination of generic residual generation method and its Laguerre approximation is proposed due to the problem of residual similarity in fault diagnostic process. Second, a novel approach for adaptive reconfigurable control synthesis is proposed using a nonlinear system identification approach based on the approximation of systems via Laguerre series. The Laguerre-based reconfiguration method is implemented and tested on a time delay form of COSY benchmark model as a preliminary study of reconfigurable control applied to a nonlinear time-delayed model. Actuator faults have been implemented in the COSY benchmark and used to evaluate the control reconfiguration schemes. Simulation results showed acceptable level of fault tolerance.
International Journal of Adaptive Control and Signal Processing
2019 27th Iranian Conference on Electrical Engineering (ICEE)
International Journal of Automation and Computing, Dec 5, 2016
This paper presents a robust adaptive state feedback control scheme for a class of parametric-str... more This paper presents a robust adaptive state feedback control scheme for a class of parametric-strict-feedback nonlinear systems in the presence of time varying actuator failures. The designed adaptive controller compensates a general class of actuator failures without any need for explicit fault detection. The parameters, times, and patterns of the considered failures are completely unknown. The proposed controller is constructed based on a backstepping design method. The global boundedness of all the closed-loop signals is guaranteed and the tracking error is proved to converge to a small neighborhood of the origin. The proposed approach is employed for a two-axis positioning stage system as well as an aircraft wing system. The simulation results show the correctness and effectiveness of the proposed robust adaptive actuator failure compensation approach.
2022 30th International Conference on Electrical Engineering (ICEE)
Additional file 1: Interaction list for drug–target. The raw data used in the drug–target is incl... more Additional file 1: Interaction list for drug–target. The raw data used in the drug–target is included in this file. Drug–target interactions were extracted from DrugBank database
Additional file 6: Impact of the encoder. Removing the encoder significantly declines the perform... more Additional file 6: Impact of the encoder. Removing the encoder significantly declines the performance of DIDL
Additional file 5: Evaluation of DIDL with Hetionet network. Performance of DIDL for prediction o... more Additional file 5: Evaluation of DIDL with Hetionet network. Performance of DIDL for prediction of interactions between different layers of Hetionet network is assessed using 10 fold cross-validation
Additional file 3: Interaction list for TF–DNA. The raw data used in the TF–DNA is included in th... more Additional file 3: Interaction list for TF–DNA. The raw data used in the TF–DNA is included in this file. Interactions were extracted from the Enrichr database using ChEA 2016
2017 Iranian Conference on Electrical Engineering (ICEE), 2017
This paper offers a robust adaptive control based dynamic surface control (DSC) for uncertain non... more This paper offers a robust adaptive control based dynamic surface control (DSC) for uncertain nonlinear systems in which the unknown nonlinearities are not linearly parameterized with respect to uncertain parameters. Therefore, radial basis function (RBF) neural networks (NNs) are used for approximating uncertain nonlinearities. The proposed controller guarantees uniformly ultimately boundedness of the closed loop system and guarantees much higher tracking accuracy compared with the previous works in backstepping and DSC methods. Simulation results are presented to show the effectiveness of the proposed approach.
The main purpose of this paper is to present an adaptive-neural controller for strictfeedback non... more The main purpose of this paper is to present an adaptive-neural controller for strictfeedback nonlinear systems with unknown time delays and in the presence of external disturbances and actuator failure. The proposed adaptive-neural controller is constructed based on DSC design technique. Radial Basis Functions (RBF) networks are utilized to approximate unknown nonlinear functions. Adaptive rules are obtained based on Lyapunov design for updating the parameters of neural networks. Disturbances are unknown functions which their bounds are partially known. Therefore, continuous robust terms are applied in order to minimize their effects. Furthermore, due to the existence of unknown time delays in the system, Lyapunov– Krasovskii functionals are utilized in the process of designing the controller and proofing the stability of the system. In addition, the controller is designed so that it can compensate its effect if the considered actuator failure happens. For the designed controller, ...
IEEE Transactions on Smart Grid, 2021
State estimation has a special role in the real-time control and monitoring of smart distribution... more State estimation has a special role in the real-time control and monitoring of smart distribution networks. State estimation process is typically based on network topology and measurements sent from meters. Employing an accurate state estimation algorithm as well as transferring high volumes of measurements are serious challenges in large scale grids. In this paper, compressive sensing is used to reduce the measurement data volume, before transmission, to alleviate problems such as lack of storage space, interference and delay. In this paper, a modified extended Kalman filter algorithm is proposed which estimates states from compressed data directly without applying the reconstruction procedure. The main differences between the proposed method and EKF are the network dynamic modeling approach and the states correction mechanism. The IEEE 33-node distribution network with two DGs is employed to illustrate the effective performance of the proposed method. Results show that the states of the test feeder are accurately estimated even with only 50% compressed measurements.
2019 27th Iranian Conference on Electrical Engineering (ICEE), 2019
Recently, consensus tracking control of a higher order multi agent nonlinear system has become an... more Recently, consensus tracking control of a higher order multi agent nonlinear system has become an interesting problem. In order to solve the explosion of design complexity, in this paper the Dynamic Surface Control (DSC) approach is used to present a new consensus method for an unknown multi agent high order nonlinear systems with uncertain disturbances. Compared to previous methods, the computational burden of the distributed adaptive control scheme is significantly reduced and a simpler distributed adaptive control scheme is designed. Moreover, unknown multi agent nonlinear system is estimated without using fuzzy logic systems or neural networks, thus the number of tuned parameters is significantly reduced. Finally, the global uniform boundedness of all the closed-loop signals and the convergence of the distributed consensus tracking errors to a small boundedness of the origin are proved. Simulation results verified that the proposed controller is effective and appropriate.
2022 30th International Conference on Electrical Engineering (ICEE)
2019 6th International Conference on Control, Instrumentation and Automation (ICCIA)
This paper offers a robust adaptive control based dynamic surface control for uncertain nonlinear... more This paper offers a robust adaptive control based dynamic surface control for uncertain nonlinear systems in the presence of actuator saturation and external disturbance. Radial Basis Function Neural Networks are used for approximating uncertain nonlinearities. Using dynamic surface control technique, the problem of explosion of complexity caused by differentiating the virtual controllers during the recursive procedures in the conventional backstepping method is avoided. The proposed controller guarantees uniformly ultimately boundedness of the closed loop system. The effectiveness of the proposed method is illustrated with a simulation example.
Robotica, 2021
In this paper, new distributed adaptive methods are proposed for solving both leaderless and lead... more In this paper, new distributed adaptive methods are proposed for solving both leaderless and leader–follower consensus problems in networks of uncertain robot manipulators, by estimating only the gravitational torque forces. Comparing with the existing adaptive methods, which require the estimation of the whole dynamics, presented methods reduce the excitation levels required for efficient parameter search, the convergence time, and the complexity of the regressor. Additionally, proposed schemes eliminate the need for velocity information exchange between the agents. Global asymptotic synchronization is shown by introducing new Lyapunov functions. Simulation results are provided for a network of 10 4-DOF robot manipulators.
Consensus is the most basic synchronization behavior in multiagent systems. For networks of Euler... more Consensus is the most basic synchronization behavior in multiagent systems. For networks of Euler-Lagrange (EL) agents different controllers have been proposed to achieve consensus, requiring in all cases, either the cancellation or the estimation of the gravity forces. In the latter case, it is necessary to estimate, not just the gravity forces, but the parameters of the whole dynamics. This requires the computation of a complicated regressor matrix, that grows in complexity as the degrees-of-freedom of the EL-agents increase. In this paper, we propose an adaptive controllers to solve the leaderless consensus problem by only estimating the gravitational term of the agents and hence without requiring the complete regressor matrix. To the best of our knowledge, this is the first work that achieves such an objective. The controller is a simple Proportional plus damping (P+d) scheme that does not require to exchange velocity information between the agents. Simulation results demonstrat...
IET Control Theory & Applications, 2017
A tube-based robust model predictive control (MPC) is proposed to be applied in constrained linea... more A tube-based robust model predictive control (MPC) is proposed to be applied in constrained linear systems with parametric uncertainty. An estimation method is applied in this proposed technique to adapt the system model at each sampling time and to reduce the conservatism nature of the tube-based MPC as the system model approaches the real model as time passes. By updating the subject model online through this newly proposed approach the performance of the system is improved. Asymptotic stability of the closed-loop system is established. The simulation results of a DC motor are applied to illustrate the effectiveness of this proposed controller in dealing with one practical system. 2 Problem description Consider a linear system in standard state-space form x k + 1 = Ax k + Bu k (1)
Neurocomputing, 2016
This paper investigates the consensus tracking problem of nonlinear multi-agent systems with stat... more This paper investigates the consensus tracking problem of nonlinear multi-agent systems with state constraints and unknown disturbances. An observer is presented for the case that the states of each follower and its neighbors are unmeasurable. State constraint for multi-agent systems is a challenging problem. Barrier Lyapunov functions are applied in this paper to deal with this difficulty. Then based on adaptive back-stepping control approach and dynamic surface control technique, an adaptive fuzzy distributed controller is proposed to guarantee that the tracking errors between all followers and the leader converge to a small neighborhood of the origin. Moreover, it is proved that all the signals in the multi-agent systems are semi-globally uniformly ultimately bounded (SUUB). Finally, some numerical simulation results are presented to testify the effectiveness of the proposed algorithm.
Despite enormous achievements in production of high throughput datasets, constructing comprehensi... more Despite enormous achievements in production of high throughput datasets, constructing comprehensive maps of interactions remains a major challenge. The lack of sufficient experimental evidence on interactions is more significant for heterogeneous molecular types. Hence, developing strategies to predict inter-omics connections is essential to construct holistic maps of disease. Here, Data Integration with Deep Learning (DIDL), a novel nonlinear deep learning method is proposed to predict inter-omics interactions. It consists of an encoder that automatically extracts features for biomolecules according to existing interactions, and a decoder that predicts novel interactions. The applicability of DIDL is assessed with different networks namely drug-target protein, transcription factor-DNA element, and miRNA-mRNA. Also, the validity of novel predictions is assessed by literature surveys. Furthermore, DIDL outperformed state-of-the-art methods. Area under the curve, and area under the pr...
ABSTRACT In this paper a new design method is presented for reconfigurable fault tolerant control... more ABSTRACT In this paper a new design method is presented for reconfigurable fault tolerant control of multivariable nonlinear time delay systems. The control reconfiguration algorithm comprises two main parts: fault detection and diagnosis and model reference adaptive control. First, a combination of generic residual generation method and its Laguerre approximation is proposed due to the problem of residual similarity in fault diagnostic process. Second, a novel approach for adaptive reconfigurable control synthesis is proposed using a nonlinear system identification approach based on the approximation of systems via Laguerre series. The Laguerre-based reconfiguration method is implemented and tested on a time delay form of COSY benchmark model as a preliminary study of reconfigurable control applied to a nonlinear time-delayed model. Actuator faults have been implemented in the COSY benchmark and used to evaluate the control reconfiguration schemes. Simulation results showed acceptable level of fault tolerance.
International Journal of Adaptive Control and Signal Processing
2019 27th Iranian Conference on Electrical Engineering (ICEE)