kth.sePublications
Change search
Link to record
Permanent link

Direct link
Feyzmahdavian, Hamid RezaORCID iD iconorcid.org/0000-0003-1149-4715
Publications (10 of 24) Show all publications
Åstrand, M., Johansson, M. & Feyzmahdavian, H. R. (2021). Short-Term Scheduling of Production Fleets in Underground Mines Using CP-Based LNS. In: Lecture Notes in Computer Science: . Paper presented at 18th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2021, Virtual, Online,5-8 July 2021 (pp. 365-382). Springer Nature
Open this publication in new window or tab >>Short-Term Scheduling of Production Fleets in Underground Mines Using CP-Based LNS
2021 (English)In: Lecture Notes in Computer Science, Springer Nature , 2021, p. 365-382Conference paper, Published paper (Refereed)
Abstract [en]

Coordinating the mobile production fleet in underground mines becomes increasingly important as the machines are more and more automated. We present a scheduling approach that applies to several of the most important production methods used in underground mines. Our algorithm combines constraint programming with a large neighborhood search strategy that dynamically adjusts the neighborhood size. The resulting algorithm is complete and able to rapidly improve constructed schedules in practice. In addition, it has important benefits when it comes to the acceptance of the approach in real-life operations. Our approach is evaluated on public and private industrial problem instances representing different mines and production methods. We find significant improvements over the current industrial practice.

Place, publisher, year, edition, pages
Springer Nature, 2021
Keywords
Constraint programming, Large neighborhood search, Scheduling, Underground mining, Computer programming, Constraint theory, Operations research, Optimization, Industrial practices, Industrial problem, Mobile productions, Neighborhood size, Production methods, Short-term scheduling, Artificial intelligence
National Category
Computer Sciences
Identifiers
urn:nbn:se:kth:diva-310725 (URN)10.1007/978-3-030-78230-6_23 (DOI)000885083100023 ()2-s2.0-85111409310 (Scopus ID)
Conference
18th International Conference on the Integration of Constraint Programming, Artificial Intelligence, and Operations Research, CPAIOR 2021, Virtual, Online,5-8 July 2021
Note

Part of proceedings ISBN: 978-3-030-78229-0

QC 20220413

Available from: 2022-04-13 Created: 2022-04-13 Last updated: 2022-12-16Bibliographically approved
Qazizadeh, A., Stichel, S. & Feyzmahdavian, H. R. (2018). Wheelset curving guidance using H∞ control. Vehicle System Dynamics, 56(3), 461-484
Open this publication in new window or tab >>Wheelset curving guidance using H∞ control
2018 (English)In: Vehicle System Dynamics, ISSN 0042-3114, E-ISSN 1744-5159, Vol. 56, no 3, p. 461-484Article in journal (Refereed) Published
Abstract [en]

This study shows how to design an active suspension system for guidance of a rail vehicle wheelset in curve. The main focus of the study is on designing the controller and afterwards studying its effect on the wheel wear behaviour. The controller is designed based on the closed-loop transfer function shaping method and (Formula presented.) control strategy. The study discusses designing of the controller for both nominal and uncertain plants and considers both stability and performance. The designed controllers in Simulink are then applied to the vehicle model in Simpack to study the wheel wear behaviour in curve. The vehicle type selected for this study is a two-axle rail vehicle. This is because this type of vehicle is known to have very poor curving performance and high wheel wear. On the other hand, the relative simpler structure of this type of vehicle compared to bogie vehicles make it a more economic choice. Hence, equipping this type of vehicle with the active wheelset steering is believed to show high enough benefit to cost ratio to remain attractive to rail vehicle manufacturers and operators. 

Place, publisher, year, edition, pages
Taylor & Francis, 2018
Keywords
Active suspension, actuated solid wheelset, control, shaping closed-loop transfer function, two-axle rail vehicle, wear, Active suspension systems, Automobile manufacture, Axles, Controllers, Cost benefit analysis, Magnetic levitation vehicles, Rail motor cars, Suspensions (components), Suspensions (fluids), Transfer functions, Vehicle wheels, Vehicles, Wear of materials, Wheels, Benefit to cost ratios, Closed loop transfer function, Control strategies, Curving performance, Rail vehicles, Uncertain plants, Wheelsets, Automobile suspensions
National Category
Mechanical Engineering
Identifiers
urn:nbn:se:kth:diva-223119 (URN)10.1080/00423114.2017.1391396 (DOI)000427768200007 ()2-s2.0-85033389907 (Scopus ID)
Note

QC 20201118

Available from: 2018-03-27 Created: 2018-03-27 Last updated: 2022-06-26Bibliographically approved
Feyzmahdavian, H. R., Aytekin, A. & Johansson, M. (2016). Asynchronous Mini-Batch Algorithm for Regularized Stochastic Optimization. IEEE Transactions on Automatic Control
Open this publication in new window or tab >>Asynchronous Mini-Batch Algorithm for Regularized Stochastic Optimization
2016 (English)In: IEEE Transactions on Automatic Control, ISSN 0018-9286, E-ISSN 1558-2523, ISSN 0018-9286Article in journal (Refereed) Published
Abstract [en]

Mini-batch optimization has proven to be a powerful paradigm for large-scale learning. However, the state of the art parallel mini–batch algorithms assume synchronous operation or cyclic update orders. When worker nodes are heterogeneous (due to different computational capabilities or different communication delays), synchronous and cyclic operations are inefficient since they will leave workers idle waiting for the slower nodes to complete their computations. In this paper, we propose an asynchronous mini-batch algorithm for regularized stochastic optimization problems with smooth loss functions that eliminates idle waiting and allows workers to run at their maximal update rates. We show that by suitably choosing the step-size values, the algorithm achieves a rate of the order O(1/ √ T) for general convex regularization functions, and the rate O(1/T ) for strongly convex regularization functions, where T is the number of iterations. In both cases, the impact of asynchrony on the convergence rate of our algorithm is asymptotically negligible, and a nearlinear speedup in the number of workers can be expected. Theoretical results are confirmed in real implementations on a distributed computing infrastructure.

Place, publisher, year, edition, pages
IEEE, 2016
Keywords
Optimization, Asynchronous, Delay, Convex, Nonsmooth
National Category
Engineering and Technology
Research subject
Electrical Engineering
Identifiers
urn:nbn:se:kth:diva-182598 (URN)10.1109/TAC.2016.2525015 (DOI)000389891100003 ()2-s2.0-85003890240 (Scopus ID)
Note

QC 20160311

Available from: 2016-02-20 Created: 2016-02-20 Last updated: 2024-03-15Bibliographically approved
Besselink, B., Feyzmahdavian, H. R., Sandberg, H. & Johansson, M. (2016). D-stability and delay-independent stability of monotone nonlinear systems with max-separable Lyapunov functions. In: 2016 IEEE 55th Conference on Decision and Control, CDC 2016: . Paper presented at 55th IEEE Conference on Decision and Control, CDC 2016, ARIA Resort and Casino, Las Vegas, United States, 12 December 2016 through 14 December 2016 (pp. 3172-3177). Institute of Electrical and Electronics Engineers (IEEE), Article ID 2-s2.0-85010722892.
Open this publication in new window or tab >>D-stability and delay-independent stability of monotone nonlinear systems with max-separable Lyapunov functions
2016 (English)In: 2016 IEEE 55th Conference on Decision and Control, CDC 2016, Institute of Electrical and Electronics Engineers (IEEE), 2016, p. 3172-3177, article id 2-s2.0-85010722892Conference paper, Published paper (Refereed)
Abstract [en]

Stability properties of monotone nonlinear systems with max-separable Lyapunov functions are considered in this paper, motivated by the following observations. First, recent results have shown that such Lyapunov functions are guaranteed to exist for asymptotically stable monotone systems on compact sets. Second, it is well-known that, for monotone linear systems, asymptotic stability implies the stronger properties of D-stability and robustness with respect to time-delays. This paper shows that similar properties hold for monotone nonlinear systems that admit max-separable Lyapunov functions. In particular, a notion of D-stability for monotone nonlinear systems and delay-independent stability will be discussed. The theoretical results are illustrated by means of examples.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2016
Series
IEEE Conference on Decision and Control, ISSN 0743-1546
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-208563 (URN)10.1109/CDC.2016.7798745 (DOI)000400048103058 ()2-s2.0-85010722892 (Scopus ID)978-1-5090-1837-6 (ISBN)
Conference
55th IEEE Conference on Decision and Control, CDC 2016, ARIA Resort and Casino, Las Vegas, United States, 12 December 2016 through 14 December 2016
Note

QC 20170615

Available from: 2017-06-15 Created: 2017-06-15 Last updated: 2024-03-15Bibliographically approved
Feyzmahdavian, H. R. (2016). Performance Analysis of Positive Systems and Optimization Algorithms with Time-delays. (Doctoral dissertation). Stockholm: KTH Royal Institute of Technology
Open this publication in new window or tab >>Performance Analysis of Positive Systems and Optimization Algorithms with Time-delays
2016 (English)Doctoral thesis, monograph (Other academic)
Abstract [en]

Time-delay dynamical systems are used to model many real-world engineering systems, where the future evolution of a system depends not only on current states but also on the history of states. For this reason, the study of stability and control of time-delay systems is of theoretical and practical importance. In this thesis, we develop several stability analysis frameworks for dynamical systems in the presence of communication and computation time-delays, and apply our results to different challenging engineering problems.

The thesis first considers delay-independent stability of positive monotone systems. We show that the asymptotic stability of positive monotone systems whose vector fields are homogeneous is independent of the magnitude and variation of time-varying delays. We present explicit expressions that allow us to give explicit estimates of the decay rate for various classes of time-varying delays. For positive linear systems, we demonstrate that the best decay rate that our results guarantee can be found via convex optimization. We also derive a set of necessary and sufficient conditions for asymptotic stability of general positive monotone (not necessarily homogeneous) systems with time-delays. As an application of our theoretical results, we discuss delay-independent stability of continuous-time power control algorithms in wireless networks.

The thesis continues by studying the convergence of asynchronous fixed-point iterations involving maximum norm pseudo-contractions. We present a powerful approach for characterizing the rate of convergence of totally asynchronous iterations, where both the update intervals and communication delays may grow unbounded. When specialized to partially asynchronous iterations (where the update intervals and communication delays have a fixed upper bound), or to particular classes of unbounded delays and update intervals, our approach allows to quantify how the degree of asynchronism affects the convergence rate. In addition, we use our results to analyze the impact of asynchrony on the convergence rate of discrete-time power control algorithms in wireless networks.

The thesis finally proposes an asynchronous parallel algorithm that exploits multiple processors to solve regularized stochastic optimization problems with smooth loss functions. The algorithm allows the processors to work at different rates, perform computations independently of each other, and update global decision variables using out-of-date gradients. We characterize the iteration complexity and the convergence rate of the proposed algorithm, and show that these compare favourably with the state of the art. Furthermore, we demonstrate that the impact of asynchrony on the convergence rate of the algorithm is asymptotically negligible, and a near-linear speedup in the number of processors can be expected.

Abstract [sv]

Tidsfördröjningar uppstår ofta i tekniska system: det tar tid för två ämnen attblandas, det tar tid för en vätska att rinna från ett kärl till ett annat, och det tar tid att överföra information mellan delsystem. Dessa tidsfördröjningar lederofta till försämrad systemprestanda och ibland även till instabilitet. Det är därförviktigt att utveckla teori och ingenjörsmetodik som gör det möjligt att bedöma hur tidsfördröjningar påverkar dynamiska system.

I den här avhandlingen presenteras flera bidrag till detta forskningsområde. Fokusligger på att karaktärisera hur tidsfördröjningar påverkar konvergenshastigheten hos olinjära dynamiska system. I kapitel 3 och 4 behandlar vi olinjära system varstillstånd alltid är positiva. Vi visar att stabiliteten av dessa positiva system är oberoende av tidsfördröjningar och karaktäriserar hur konvergenshastigheten hos olinjära positiva system beror på tidsfördröjningarnas storlek. I kapitel 5 betraktar vi iterationer som är kontraktionsavbildningar, och analyserar hur deras konvergens påverkas av begränsade och obegränsade tidsfördröjningar. I avhandlingens sistakapitel föreslår vi en asynkron algoritm för stokastisk optimering vars asymptotiska konvergenshastighet är oberoende av tidsfördröjningar i beräkningar och i kommunikation mellan beräkningselement.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2016. p. xv, 146
Series
TRITA-EE, ISSN 1653-5146 ; 2015:83
Keywords
Optimization, Delay, Positive system, Asynchronous
National Category
Engineering and Technology
Research subject
Electrical Engineering
Identifiers
urn:nbn:se:kth:diva-177651 (URN)978-91-7595-790-6 (ISBN)
Public defence
2016-01-15, Q2, Osquldas väg 10, KTH, Stockholm, 10:00 (English)
Opponent
Supervisors
Note

QC 20151204

Available from: 2015-12-04 Created: 2015-11-25 Last updated: 2022-09-06Bibliographically approved
Feyzmahdavian, H. R., Aytekin, A. & Johansson, M. (2015). An Asynchronous Mini-Batch Algorithm for Regularized Stochastic Optimization. In: 2015 54th IEEE Conference on Decision and Control (CDC): . Paper presented at CDC 2015 (pp. 1384-1389). IEEE
Open this publication in new window or tab >>An Asynchronous Mini-Batch Algorithm for Regularized Stochastic Optimization
2015 (English)In: 2015 54th IEEE Conference on Decision and Control (CDC), IEEE , 2015, p. 1384-1389Conference paper, Published paper (Refereed)
Abstract [en]

Mini-batch optimization has proven to be a powerful paradigm for large-scale learning. However, the state of the art mini-batch algorithms assume synchronous operation or cyclic update orders. When worker nodes are heterogeneous (due to different computational capabilities, or different communication delays), synchronous and cyclic operations are inefficient since they will leave workers idle waiting for the slower nodes to complete their work. We propose an asynchronous mini-batch algorithm for regularized stochastic optimization problems that eliminates idle waiting and allows workers to run at their maximal update rates. We show that the time necessary to compute an ϵ-optimal solution is asymptotically O(1/ϵ2), and the algorithm enjoys near-linear speedup if the number of workers is O(1/√ϵ). Theoretical results are confirmed in real implementations on a distributed computing infrastructure.

Place, publisher, year, edition, pages
IEEE, 2015
Keywords
Optimization, Convex, Asynchronous optimization, Delay, Stochastic optimization
National Category
Engineering and Technology
Identifiers
urn:nbn:se:kth:diva-183457 (URN)10.1109/CDC.2015.7402404 (DOI)000381554501089 ()2-s2.0-84962032572 (Scopus ID)978-1-4799-7884-7 (ISBN)
Conference
CDC 2015
Note

QC 20160314

Available from: 2016-03-12 Created: 2016-03-12 Last updated: 2024-03-15Bibliographically approved
Feyzmahdavian, H. R., Charalambous, T. & Johansson, M. (2015). Delay-independent Stability of Cone-invariant Monotone Systems. In: 54th IEEE Conference on Decision and Control (CDC): . Paper presented at CDC 2015 (pp. 6361-6366). IEEE
Open this publication in new window or tab >>Delay-independent Stability of Cone-invariant Monotone Systems
2015 (English)In: 54th IEEE Conference on Decision and Control (CDC), IEEE , 2015, p. 6361-6366Conference paper, Published paper (Refereed)
Abstract [en]

Recent results in the literature have shown that particular classes of positive systems are insensitive to time-varying delays, giving the impression that the delay-insensitivity property stems from the fact that the system is positive. Nonetheless, it has been lately shown that a linear cone-invariant system is insensitive to time-varying delays, asserting that the property of delay-independence may stem from the fact that the system is cone-invariant rather than positive. In this paper, we provide additional evidence for this claim by analyzing the stability of cone-invariant monotone systems with bounded time-varying delays. We present a set of sufficient conditions for delay independent stability of discrete- and continuous-time cone-invariant monotone systems. For linear cone-invariant systems, we show that thestability conditions we have derived are also necessary.

Place, publisher, year, edition, pages
IEEE, 2015
Keywords
Positive system, Stability, Cone, Delay
National Category
Engineering and Technology
Identifiers
urn:nbn:se:kth:diva-183456 (URN)10.1109/CDC.2015.7403221 (DOI)000381554506090 ()2-s2.0-84961990693 (Scopus ID)
Conference
CDC 2015
Note

QC 20160314

Available from: 2016-03-12 Created: 2016-03-12 Last updated: 2024-03-15Bibliographically approved
Lu, J., Feyzmahdavian, H. R. & Johansson, M. (2015). Dual coordinate descent algorithms for multi-agent optimization. In: European Control Conference (ECC15): . Paper presented at ECC 2015. IEEE conference proceedings
Open this publication in new window or tab >>Dual coordinate descent algorithms for multi-agent optimization
2015 (English)In: European Control Conference (ECC15), IEEE conference proceedings, 2015Conference paper, Published paper (Refereed)
Abstract [en]

Multi-agent optimization problems arise in a wide variety of networked systems, and are often required to be solved in an asynchronous and uncoordinated way. However, existing asynchronous algorithms for constrained multi-agent optimization do not have guaranteed convergence rates and, thus, lack performance guarantees in on-line applications. This paper addresses this shortcoming by developing randomized coordinate descent algorithms for solving the dual of a class of constrained multi-agent optimization problems. We show that the algorithms can be implemented asynchronously and distributively in multi-agent networks. Moreover, without relying on the standard assumption of boundedness of the dual optimal set, the proposed dual coordinate descent algorithms achieve sublinear convergence rates of both its primal and dual iterates in expectation. The competitive performance is demonstrated numerically on a constrained optimal rendezvous problem.

Place, publisher, year, edition, pages
IEEE conference proceedings, 2015
Keywords
Optimization, Coordinate descent, Convex, First order method
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-183461 (URN)10.1109/ECC.2015.7330626 (DOI)000380485400111 ()2-s2.0-84963852478 (Scopus ID)
Conference
ECC 2015
Note

QC 20160317

Available from: 2016-03-12 Created: 2016-03-12 Last updated: 2024-03-15Bibliographically approved
Ghadimi, E., Feyzmahdavian, H. R. & Johansson, M. (2015). Global Convergence of the Heavy-ball Method for Convex Optimization. In: European Control Conference (ECC15): . Paper presented at ECC 2015. IEEE
Open this publication in new window or tab >>Global Convergence of the Heavy-ball Method for Convex Optimization
2015 (English)In: European Control Conference (ECC15), IEEE , 2015Conference paper, Published paper (Refereed)
Abstract [en]

This paper establishes global convergence and provides global bounds of the rate of convergence for the Heavy-ball method for convex optimization. When the objective function has Lipschitz-continuous gradient, we show that the Cesáro average of the iterates converges to the optimum at a rate of O(1/k) where k is the number of iterations. When the objective function is also strongly convex, we prove that the Heavy-ball iterates converge linearly to the unique optimum. Numerical examples validate our theoretical findings.

Place, publisher, year, edition, pages
IEEE, 2015
Keywords
Optimization, Convex, Heavy ball, Gradient iteration
National Category
Engineering and Technology
Identifiers
urn:nbn:se:kth:diva-183460 (URN)10.1109/ECC.2015.7330562 (DOI)000380485400047 ()2-s2.0-84963894675 (Scopus ID)
Conference
ECC 2015
Note

QC 20160314

Available from: 2016-03-12 Created: 2016-03-12 Last updated: 2024-03-15Bibliographically approved
Demirel, B., Feyzmahdavian, H. R., Ghadimi, E. & Johansson, M. (2015). Stability Analysis of Discrete-Time Linear Systems with Unbounded Stochastic Delays. In: 5th IFAC Workshop on Distributed Estimation and Control of Networked Systems (NECSYS): . Paper presented at 5th IFAC Workshop on Distributed Estimation and Control in Networked Systems NecSys 2015 — Philadelphia, 10-11 September 2015. Elsevier, 48
Open this publication in new window or tab >>Stability Analysis of Discrete-Time Linear Systems with Unbounded Stochastic Delays
2015 (English)In: 5th IFAC Workshop on Distributed Estimation and Control of Networked Systems (NECSYS), Elsevier, 2015, Vol. 48Conference paper, Published paper (Refereed)
Abstract [en]

This paper investigates the stability of discrete-time linear systems with stochastic delays. We assume that delays are modeled as random variables, which take values in integers with a certain probability. For the scalar case, we provide an analytical bound on the probability to guarantee the stability of linear systems. In the vector case, we derive a linear matrix inequality condition to compute the probability for ensuring the stability of closed-loop systems. As a special case, we also determine the step size of gradient algorithms with stochastic delays in the unconstrained quadratic programming to guarantee convergence to the optimal solution. Numerical examples are provided to show the effectiveness of the proposed analysis techniques.

Place, publisher, year, edition, pages
Elsevier, 2015
Keywords
Delay, Ranodom delay, Stochastic system
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-183459 (URN)10.1016/ifacol.2015.10.312 (DOI)2-s2.0-84992490272 (Scopus ID)
Conference
5th IFAC Workshop on Distributed Estimation and Control in Networked Systems NecSys 2015 — Philadelphia, 10-11 September 2015
Note

QC 20160315

Available from: 2016-03-12 Created: 2016-03-12 Last updated: 2024-03-15Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0003-1149-4715

Search in DiVA

Show all publications