Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945663
Deepak Kumar, Umair Zulfiqar, V. Sreeram, M. Imran, W. Muda, A. Jazlan, A. Wu
In this paper, a new frequency-weighted positive real balanced truncation technique is presented by using a combination of frequency-weighted model reduction with the positive real-truncated balanced realization (PR-TBR). The proposed technique yields passive reduced-order model for a given passive high order model with single-sided weighting only. The simulation results of the proposed technique are shown by using an example problem of RLC ladder network.
{"title":"Positive-Real Truncated Balanced Realization based Frequency-Weighted Model reduction","authors":"Deepak Kumar, Umair Zulfiqar, V. Sreeram, M. Imran, W. Muda, A. Jazlan, A. Wu","doi":"10.1109/ANZCC47194.2019.8945663","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945663","url":null,"abstract":"In this paper, a new frequency-weighted positive real balanced truncation technique is presented by using a combination of frequency-weighted model reduction with the positive real-truncated balanced realization (PR-TBR). The proposed technique yields passive reduced-order model for a given passive high order model with single-sided weighting only. The simulation results of the proposed technique are shown by using an example problem of RLC ladder network.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114759933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945668
A. Kaldmäe, J. Doná
This paper addresses the problem of characterizing analytically invariant sets for nonlinear discrete-time systems. Different notions of invariance are defined and the effect of state and input transformations on these invariant sets is studied. The main part of the paper considers finding robustly positively invariant sets for feedback linearizable (with respect to the disturbance input) systems. It is shown that taking the system into controller canonical form, simplifies the computations considerably. The ultimate goal is to find the minimal robustly invariant set, which is described through the notion of reachability. Finally, it is shown that convex invariant sets of discretized systems using the Euler forward discretization scheme are also invariant for the respective continuous-time system. The purpose of this article is to present some known as well as some new results, illustrated by simple examples, in a tutorial, self-contained form, invoking only basic set theoretic methods and coordinate transformations.
{"title":"On Robustly Positively Invariant Sets and Coordinate Transformations for Discrete-time Nonlinear Systems: a Tutorial","authors":"A. Kaldmäe, J. Doná","doi":"10.1109/ANZCC47194.2019.8945668","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945668","url":null,"abstract":"This paper addresses the problem of characterizing analytically invariant sets for nonlinear discrete-time systems. Different notions of invariance are defined and the effect of state and input transformations on these invariant sets is studied. The main part of the paper considers finding robustly positively invariant sets for feedback linearizable (with respect to the disturbance input) systems. It is shown that taking the system into controller canonical form, simplifies the computations considerably. The ultimate goal is to find the minimal robustly invariant set, which is described through the notion of reachability. Finally, it is shown that convex invariant sets of discretized systems using the Euler forward discretization scheme are also invariant for the respective continuous-time system. The purpose of this article is to present some known as well as some new results, illustrated by simple examples, in a tutorial, self-contained form, invoking only basic set theoretic methods and coordinate transformations.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125733118","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945515
Umair Zulfiqar, V. Sreeram, M. Imran
The frequency-limited model reduction techniques aim to construct a model of significantly lower order which approximates the original higher order model accurately within some specified frequency interval. In this paper, we propose a frequency-limited model reduction technique for multi-input and multi-output systems wherein a different desired frequency region can be specified for each input-output channel. The reduced order model thus obtained ensures superior accuracy within the frequency regions specified for each input-output channel. A numerical example is presented to demonstrate the applicability and the efficacy of the proposed algorithm.
{"title":"Channel-specific Frequency-limited Model Reduction","authors":"Umair Zulfiqar, V. Sreeram, M. Imran","doi":"10.1109/ANZCC47194.2019.8945515","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945515","url":null,"abstract":"The frequency-limited model reduction techniques aim to construct a model of significantly lower order which approximates the original higher order model accurately within some specified frequency interval. In this paper, we propose a frequency-limited model reduction technique for multi-input and multi-output systems wherein a different desired frequency region can be specified for each input-output channel. The reduced order model thus obtained ensures superior accuracy within the frequency regions specified for each input-output channel. A numerical example is presented to demonstrate the applicability and the efficacy of the proposed algorithm.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"347 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126677059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945712
Barbara Martinovic, J. Leth, T. Knudsen, T. B. Aradóttir, Henrik Bengtsson
This paper investigate the use of ARMAX models to describe the glucose-insulin system of type 2 diabetes patients. The parameters of the ARMAX models are estimated using clinical data of 416 type 2 diabetes patients. A variety of parameter combinations has been investigated and no significant change in fitness values and test statistics were found. Moreover, partially due to the delay parameter no conclusive models were found leading to the conclusion that other aspects (such as non-linearities) has to be taken into account when models of the glucose-insulin system are based on time series.
{"title":"Modelling the glucose-insulin system of type 2 diabetes patients using ARMAX models","authors":"Barbara Martinovic, J. Leth, T. Knudsen, T. B. Aradóttir, Henrik Bengtsson","doi":"10.1109/ANZCC47194.2019.8945712","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945712","url":null,"abstract":"This paper investigate the use of ARMAX models to describe the glucose-insulin system of type 2 diabetes patients. The parameters of the ARMAX models are estimated using clinical data of 416 type 2 diabetes patients. A variety of parameter combinations has been investigated and no significant change in fitness values and test statistics were found. Moreover, partially due to the delay parameter no conclusive models were found leading to the conclusion that other aspects (such as non-linearities) has to be taken into account when models of the glucose-insulin system are based on time series.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127467598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945696
Ayaz Ahmed Hoshu, A. Fisher, Liuping Wang
Multirotor unmanned aircraft systems (UAS) have enjoyed much popularity of late due to their excellent manoeuvrability, VTOL capability, precise hovering, simple deign and satisfactory speed. But they do lack in terms of energy efficiency and endurance when compared to helicopter and fixed wing counterparts. This paper explores a heterogeneous multirotor configuration which attains the benefits of both helicopter and multirotor. The design employs a single large central rotor for the lift and three tilted small boom rotors for control. Three single loop PI cascade control system is designed to control the attitude of the rotorcraft. Motor dynamics of each rotor are also considered in the multirotor model and a novel motor control loop is implemented for enhanced disturbance rejection in turblent environment. The presented configuration has been explored previously mainly focusing on energy efficiency but reported degraded attitude performance of the UAV. This work mainly focuses on the control system strategies to improve the robustness and stability of the system. We demonstrate the proposed control strategy with stable attitude performance, reference tracking and disturbance rejection by means of Simulink simulation. Deigning prototype is also in progress which will demonstrate its flight.
{"title":"Cascaded Attitude Control For Heterogeneous Multirotor UAS For Enhanced Disturbance Rejection","authors":"Ayaz Ahmed Hoshu, A. Fisher, Liuping Wang","doi":"10.1109/ANZCC47194.2019.8945696","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945696","url":null,"abstract":"Multirotor unmanned aircraft systems (UAS) have enjoyed much popularity of late due to their excellent manoeuvrability, VTOL capability, precise hovering, simple deign and satisfactory speed. But they do lack in terms of energy efficiency and endurance when compared to helicopter and fixed wing counterparts. This paper explores a heterogeneous multirotor configuration which attains the benefits of both helicopter and multirotor. The design employs a single large central rotor for the lift and three tilted small boom rotors for control. Three single loop PI cascade control system is designed to control the attitude of the rotorcraft. Motor dynamics of each rotor are also considered in the multirotor model and a novel motor control loop is implemented for enhanced disturbance rejection in turblent environment. The presented configuration has been explored previously mainly focusing on energy efficiency but reported degraded attitude performance of the UAV. This work mainly focuses on the control system strategies to improve the robustness and stability of the system. We demonstrate the proposed control strategy with stable attitude performance, reference tracking and disturbance rejection by means of Simulink simulation. Deigning prototype is also in progress which will demonstrate its flight.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129536707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945719
Ifrah Saeed, T. Alpcan, S. Erfani, M. Yilmaz
Coordinating two or more dynamic systems such as autonomous vehicles or satellites in a distributed manner poses an important research challenge. Multiple approaches to this problem have been proposed including Nonlinear Model Predictive Control (NMPC) and its model-free counterparts in reinforcement learning (RL) literature such as Deep QNetwork (DQN). This initial study aims to compare and contrast the optimal control technique, NMPC, where the model is known, with the popular model-free RL method, DQN. Simple distributed variants of these for the specific problem of balancing and synchronising two highly unstable cart-pole systems are investigated numerically. We found that both NMPC and trained DQN work optimally under ideal model and small communication delays. While NMPC performs sub-optimally under a model-mismatch scenario, DQN performance naturally does not suffer from this. Distributed DQN needs a lot of realworld experience to be trained but once it is trained, it does not have to spend its time finding the optimal action at every time-step like NMPC. This illustrative comparison lays a foundation for hybrid approaches, which can be applied to complex multi-agent scenarios.
{"title":"Distributed Nonlinear Model Predictive Control and Reinforcement Learning","authors":"Ifrah Saeed, T. Alpcan, S. Erfani, M. Yilmaz","doi":"10.1109/ANZCC47194.2019.8945719","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945719","url":null,"abstract":"Coordinating two or more dynamic systems such as autonomous vehicles or satellites in a distributed manner poses an important research challenge. Multiple approaches to this problem have been proposed including Nonlinear Model Predictive Control (NMPC) and its model-free counterparts in reinforcement learning (RL) literature such as Deep QNetwork (DQN). This initial study aims to compare and contrast the optimal control technique, NMPC, where the model is known, with the popular model-free RL method, DQN. Simple distributed variants of these for the specific problem of balancing and synchronising two highly unstable cart-pole systems are investigated numerically. We found that both NMPC and trained DQN work optimally under ideal model and small communication delays. While NMPC performs sub-optimally under a model-mismatch scenario, DQN performance naturally does not suffer from this. Distributed DQN needs a lot of realworld experience to be trained but once it is trained, it does not have to spend its time finding the optimal action at every time-step like NMPC. This illustrative comparison lays a foundation for hybrid approaches, which can be applied to complex multi-agent scenarios.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129782390","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945666
F. R. S. Sevilla, Jean Dobrowolski, A. Obushevs, P. Korba
In this work an algorithm for identification of power system frequency deviation is presented. The proposed approach can be used to monitor frequency measurements from syncrophasor measurement units (PMU) and to store data only for important events and save storage in the local server. The detection algorithm use a sliding window that rise a flag if the measured frequency deviates from a predefined set point. If the alarm flag is constant over several sliding windows, an event is captured and locally stored for further analysis. To demonstrate the effectiveness of the proposed approach, real PMU measurements from the Swiss power system are used as input.
{"title":"Detection of Frequency Deviations for Monitoring of Power Systems","authors":"F. R. S. Sevilla, Jean Dobrowolski, A. Obushevs, P. Korba","doi":"10.1109/ANZCC47194.2019.8945666","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945666","url":null,"abstract":"In this work an algorithm for identification of power system frequency deviation is presented. The proposed approach can be used to monitor frequency measurements from syncrophasor measurement units (PMU) and to store data only for important events and save storage in the local server. The detection algorithm use a sliding window that rise a flag if the measured frequency deviates from a predefined set point. If the alarm flag is constant over several sliding windows, an event is captured and locally stored for further analysis. To demonstrate the effectiveness of the proposed approach, real PMU measurements from the Swiss power system are used as input.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130074963","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945718
H. Nurdin
Conventional control of islanded microgrids employ power electronic inverters as voltage controllers. However, inverters can also be operated in current control mode with a number of desirable features. One proposed scheme is to use a proportional and resonant controller to regulate the inverter output current and use the phase estimate from a PLL as a nonlinear feedback signal. It was recently shown that for linear time-invariant loads this current control scheme has a family of sinusoidal orbits in its state space, producing a sinusoidal output at the inverter. However, it remained an open question as to whether this family of sinusoidal orbits is stable in the sense that it attracts nearby trajectories in the system’s state-space. In this work we establish stability theorems for the sinusoidal orbits, providing explicit conditions for their stability.
{"title":"Stability Analysis of the Sinusoidal Orbits of a Nonlinear Proportional and Resonant Current Regulator for Islanded Microgrids","authors":"H. Nurdin","doi":"10.1109/ANZCC47194.2019.8945718","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945718","url":null,"abstract":"Conventional control of islanded microgrids employ power electronic inverters as voltage controllers. However, inverters can also be operated in current control mode with a number of desirable features. One proposed scheme is to use a proportional and resonant controller to regulate the inverter output current and use the phase estimate from a PLL as a nonlinear feedback signal. It was recently shown that for linear time-invariant loads this current control scheme has a family of sinusoidal orbits in its state space, producing a sinusoidal output at the inverter. However, it remained an open question as to whether this family of sinusoidal orbits is stable in the sense that it attracts nearby trajectories in the system’s state-space. In this work we establish stability theorems for the sinusoidal orbits, providing explicit conditions for their stability.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"139 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133075401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945637
Jesse Cranney, J. Doná, F. Rigaut, V. Korkiakoski
In this paper, we implement and evaluate the performance of multi-rate predictive control for Adaptive Optics systems. The multi-rate strategy yields considerable performance improvements in the case of low-magnitude guide stars and in the presence of wave-front sensor read-out noise, since the wave-front sensor exposure time can be increased to provide useful signal-to-noise ratios without having to limit the temporal bandwidth of the deformable mirror (as is the case in a single-rate control approach). Simulations performed on an end-to-end numerical Adaptive Optics simulation package are presented to demonstrate the advantages of the approach.
{"title":"Evaluation of a Multi-rate Predictive Control Strategy for Adaptive Optics Systems","authors":"Jesse Cranney, J. Doná, F. Rigaut, V. Korkiakoski","doi":"10.1109/ANZCC47194.2019.8945637","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945637","url":null,"abstract":"In this paper, we implement and evaluate the performance of multi-rate predictive control for Adaptive Optics systems. The multi-rate strategy yields considerable performance improvements in the case of low-magnitude guide stars and in the presence of wave-front sensor read-out noise, since the wave-front sensor exposure time can be increased to provide useful signal-to-noise ratios without having to limit the temporal bandwidth of the deformable mirror (as is the case in a single-rate control approach). Simulations performed on an end-to-end numerical Adaptive Optics simulation package are presented to demonstrate the advantages of the approach.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"122 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124189353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-11-01DOI: 10.1109/ANZCC47194.2019.8945684
Midhun T. Augustine, N. Beena
In this paper, a feedback linearization based higher order sliding mode controller is proposed in which the switching gains are modified adaptively. The proposed method reduces the switching gain thereby reduces the control effort. Simulation of the proposed control law is carried out in DC motor speed tracking applications.
{"title":"Feedback Linearization based Adaptive Higher Order Sliding Mode Control","authors":"Midhun T. Augustine, N. Beena","doi":"10.1109/ANZCC47194.2019.8945684","DOIUrl":"https://doi.org/10.1109/ANZCC47194.2019.8945684","url":null,"abstract":"In this paper, a feedback linearization based higher order sliding mode controller is proposed in which the switching gains are modified adaptively. The proposed method reduces the switching gain thereby reduces the control effort. Simulation of the proposed control law is carried out in DC motor speed tracking applications.","PeriodicalId":322243,"journal":{"name":"2019 Australian & New Zealand Control Conference (ANZCC)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124546005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：