Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems

Pontus Boström; Mikko Heikkilä; Mikko Huova; Marina Walden; Matti Linjama

doi:10.1007/978-3-319-22264-6_20

Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems

Pontus Boström, Mikko Heikkilä, Mikko Huova, Marina Walden, Matti Linjama

Information Technology

Research output: Chapter in Book/Conference proceeding › Conference contribution › Scientific › peer-review

Abstract

This paper presents a method for statistical analysis of hybrid systems affected by stochastic disturbances, such as random computation and communication delays. The method is applied to the analysis of a computer controlled digital hydraulic power management system, where such effects are present. Bayesian inference is used to perform parameter estimation and we use hypothesis testing based on Bayes factors to compare properties of different variants of the system to assess the impact of different random disturbances. The key idea is to use sequential sampling to generate only as many samples from the models as needed to achieve desired confidence in the result.

Original language	Undefined/Unknown
Title of host publication	Quantitative Evaluation of Systems - 12th International Conference, QEST 2015, Madrid, Spain, September 1-3, 2015, Proceedings
Editors	J Campos, BR Haverkort
Publisher	Springer
Pages	312–328
ISBN (Electronic)	978-3-319-22264-6
ISBN (Print)	978-3-319-22263-9
DOIs	https://doi.org/10.1007/978-3-319-22264-6_20
Publication status	Published - 2015
MoE publication type	A4 Article in a conference publication
Event	Quantitative Evaluation of Systems (QEST) - Quantitative Evaluation of Systems (QEST'15) Duration: 1 Sept 2015 → 3 Sept 2015

Conference

Conference	Quantitative Evaluation of Systems (QEST)
Period	01/09/15 → 03/09/15

Access to Document

10.1007/978-3-319-22264-6_20

MERGE: Merging digital hydraulic systems and supercomputing (Academy of Finland)
Walden, M., Westerholm, J., Boström, P., Ersfolk, J. & Wiik, J.
01/01/15 → 31/12/16
Project: Research Council of Finland/Other Research Councils

Cite this

Boström, P., Heikkilä, M., Huova, M., Walden, M., & Linjama, M. (2015). Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems. In J. Campos, & BR. Haverkort (Eds.), Quantitative Evaluation of Systems - 12th International Conference, QEST 2015, Madrid, Spain, September 1-3, 2015, Proceedings (pp. 312–328). Springer. https://doi.org/10.1007/978-3-319-22264-6_20

@inproceedings{ee7eb9af03d54515bdf91603a3a42bfb,

title = "Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems",

abstract = "This paper presents a method for statistical analysis of hybrid systems affected by stochastic disturbances, such as random computation and communication delays. The method is applied to the analysis of a computer controlled digital hydraulic power management system, where such effects are present. Bayesian inference is used to perform parameter estimation and we use hypothesis testing based on Bayes factors to compare properties of different variants of the system to assess the impact of different random disturbances. The key idea is to use sequential sampling to generate only as many samples from the models as needed to achieve desired confidence in the result.",

author = "Pontus Bostr{\"o}m and Mikko Heikkil{\"a} and Mikko Huova and Marina Walden and Matti Linjama",

year = "2015",

doi = "10.1007/978-3-319-22264-6_20",

language = "Odefinierat/ok{\"a}nt",

isbn = "978-3-319-22263-9",

pages = "312–328",

editor = "J Campos and BR Haverkort",

booktitle = "Quantitative Evaluation of Systems - 12th International Conference, QEST 2015, Madrid, Spain, September 1-3, 2015, Proceedings",

publisher = "Springer",

note = "Quantitative Evaluation of Systems (QEST) ; Conference date: 01-09-2015 Through 03-09-2015",

}

Boström, P, Heikkilä, M, Huova, M, Walden, M & Linjama, M 2015, Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems. in J Campos & BR Haverkort (eds), Quantitative Evaluation of Systems - 12th International Conference, QEST 2015, Madrid, Spain, September 1-3, 2015, Proceedings. Springer, pp. 312–328, Quantitative Evaluation of Systems (QEST), 01/09/15. https://doi.org/10.1007/978-3-319-22264-6_20

Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems. / Boström, Pontus; Heikkilä, Mikko; Huova, Mikko et al.
Quantitative Evaluation of Systems - 12th International Conference, QEST 2015, Madrid, Spain, September 1-3, 2015, Proceedings. ed. / J Campos; BR Haverkort. Springer, 2015. p. 312–328.

Research output: Chapter in Book/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems

AU - Boström, Pontus

AU - Heikkilä, Mikko

AU - Huova, Mikko

AU - Walden, Marina

AU - Linjama, Matti

PY - 2015

Y1 - 2015

N2 - This paper presents a method for statistical analysis of hybrid systems affected by stochastic disturbances, such as random computation and communication delays. The method is applied to the analysis of a computer controlled digital hydraulic power management system, where such effects are present. Bayesian inference is used to perform parameter estimation and we use hypothesis testing based on Bayes factors to compare properties of different variants of the system to assess the impact of different random disturbances. The key idea is to use sequential sampling to generate only as many samples from the models as needed to achieve desired confidence in the result.

AB - This paper presents a method for statistical analysis of hybrid systems affected by stochastic disturbances, such as random computation and communication delays. The method is applied to the analysis of a computer controlled digital hydraulic power management system, where such effects are present. Bayesian inference is used to perform parameter estimation and we use hypothesis testing based on Bayes factors to compare properties of different variants of the system to assess the impact of different random disturbances. The key idea is to use sequential sampling to generate only as many samples from the models as needed to achieve desired confidence in the result.

U2 - 10.1007/978-3-319-22264-6_20

DO - 10.1007/978-3-319-22264-6_20

M3 - Konferensbidrag

SN - 978-3-319-22263-9

SP - 312

EP - 328

BT - Quantitative Evaluation of Systems - 12th International Conference, QEST 2015, Madrid, Spain, September 1-3, 2015, Proceedings

A2 - Campos, J

A2 - Haverkort, BR

PB - Springer

T2 - Quantitative Evaluation of Systems (QEST)

Y2 - 1 September 2015 through 3 September 2015

ER -

Bayesian Statistical Analysis for Performance Evaluation in Real-Time Control Systems

Abstract

Conference

Access to Document

Projects

MERGE: Merging digital hydraulic systems and supercomputing (Academy of Finland)

Cite this