Optimal Sensors and Actuators Placement for Large-Scale Switched Systems

Masoud Seyed Sakha, Hamid Reza Shaker*, Maryamsadat Tahavori

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

The problem of sensor and actuator placement is computationally expensive in particular in dealing with large-scale systems. This problem is even more computationally intensive in the case of switched systems. In this paper, we propose a new numerical approach for sensor and actuator placement for large-scale switched systems. We first introduce restricted genetic algorithm (RGA). RGA is an evolutionary algorithm which is developed specifically for sensors and actuator placement. We then use RGA to reduce the computational burden of the sensor and actuator placement for switched system. The proposed method uses the generalized gramians for switched systems which are called nice gramians. The nice gramians quantify the level of observability and reachability of switched systems. To the best of our knowledge, this paper presents the first results on sensor and actuator placement of switched systems. We show the effectiveness of the approach with the help of numerical examples.

Original languageEnglish
JournalInternational Journal of Dynamics and Control
Volume7
Issue number1
Pages (from-to)147-156
ISSN2195-268X
DOIs
Publication statusPublished - Mar 2019

Fingerprint

Switched Systems
Large-scale Systems
Placement
Actuator
Actuators
Sensor
Sensors
Genetic algorithms
Genetic Algorithm
Observability
Evolutionary algorithms
Large scale systems
Reachability
Evolutionary Algorithms
Quantify
Numerical Examples

Keywords

  • Actuator placement
  • Restricted genetic algorithm
  • Sensor placement
  • Switched systems

Cite this

@article{4dc8b653668c46d2b7f9b0983adac4f4,
title = "Optimal Sensors and Actuators Placement for Large-Scale Switched Systems",
abstract = "The problem of sensor and actuator placement is computationally expensive in particular in dealing with large-scale systems. This problem is even more computationally intensive in the case of switched systems. In this paper, we propose a new numerical approach for sensor and actuator placement for large-scale switched systems. We first introduce restricted genetic algorithm (RGA). RGA is an evolutionary algorithm which is developed specifically for sensors and actuator placement. We then use RGA to reduce the computational burden of the sensor and actuator placement for switched system. The proposed method uses the generalized gramians for switched systems which are called nice gramians. The nice gramians quantify the level of observability and reachability of switched systems. To the best of our knowledge, this paper presents the first results on sensor and actuator placement of switched systems. We show the effectiveness of the approach with the help of numerical examples.",
keywords = "Actuator placement, Restricted genetic algorithm, Sensor placement, Switched systems",
author = "{Seyed Sakha}, Masoud and Shaker, {Hamid Reza} and Maryamsadat Tahavori",
year = "2019",
month = "3",
doi = "10.1007/s40435-018-0446-7",
language = "English",
volume = "7",
pages = "147--156",
journal = "International Journal of Dynamics and Control",
issn = "2195-268X",
publisher = "Heinemann",
number = "1",

}

Optimal Sensors and Actuators Placement for Large-Scale Switched Systems. / Seyed Sakha, Masoud; Shaker, Hamid Reza; Tahavori, Maryamsadat.

In: International Journal of Dynamics and Control, Vol. 7, No. 1, 03.2019, p. 147-156.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Optimal Sensors and Actuators Placement for Large-Scale Switched Systems

AU - Seyed Sakha, Masoud

AU - Shaker, Hamid Reza

AU - Tahavori, Maryamsadat

PY - 2019/3

Y1 - 2019/3

N2 - The problem of sensor and actuator placement is computationally expensive in particular in dealing with large-scale systems. This problem is even more computationally intensive in the case of switched systems. In this paper, we propose a new numerical approach for sensor and actuator placement for large-scale switched systems. We first introduce restricted genetic algorithm (RGA). RGA is an evolutionary algorithm which is developed specifically for sensors and actuator placement. We then use RGA to reduce the computational burden of the sensor and actuator placement for switched system. The proposed method uses the generalized gramians for switched systems which are called nice gramians. The nice gramians quantify the level of observability and reachability of switched systems. To the best of our knowledge, this paper presents the first results on sensor and actuator placement of switched systems. We show the effectiveness of the approach with the help of numerical examples.

AB - The problem of sensor and actuator placement is computationally expensive in particular in dealing with large-scale systems. This problem is even more computationally intensive in the case of switched systems. In this paper, we propose a new numerical approach for sensor and actuator placement for large-scale switched systems. We first introduce restricted genetic algorithm (RGA). RGA is an evolutionary algorithm which is developed specifically for sensors and actuator placement. We then use RGA to reduce the computational burden of the sensor and actuator placement for switched system. The proposed method uses the generalized gramians for switched systems which are called nice gramians. The nice gramians quantify the level of observability and reachability of switched systems. To the best of our knowledge, this paper presents the first results on sensor and actuator placement of switched systems. We show the effectiveness of the approach with the help of numerical examples.

KW - Actuator placement

KW - Restricted genetic algorithm

KW - Sensor placement

KW - Switched systems

U2 - 10.1007/s40435-018-0446-7

DO - 10.1007/s40435-018-0446-7

M3 - Journal article

VL - 7

SP - 147

EP - 156

JO - International Journal of Dynamics and Control

JF - International Journal of Dynamics and Control

SN - 2195-268X

IS - 1

ER -