Accelerating LSTM-based High-Rate Dynamic System Models

In this paper, we evaluate the use of a trained Long Short-Term Memory (LSTM) network as a surrogate for a Euler-Bernoulli beam model, and then we describe and characterize an FPGA-based deployment of the model for use in real-time structural health monitoring applications. The focus of our efforts...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kabir, Ehsan, Coble, Daniel, Satme, Joud N, Downey, Austin R. J, Bakos, Jason D, Andrews, David, Huang, Miaoqing
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Kabir, Ehsan
Coble, Daniel
Satme, Joud N
Downey, Austin R. J
Bakos, Jason D
Andrews, David
Huang, Miaoqing
description In this paper, we evaluate the use of a trained Long Short-Term Memory (LSTM) network as a surrogate for a Euler-Bernoulli beam model, and then we describe and characterize an FPGA-based deployment of the model for use in real-time structural health monitoring applications. The focus of our efforts is the DROPBEAR (Dynamic Reproduction of Projectiles in Ballistic Environments for Advanced Research) dataset, which was generated as a benchmark for the study of real-time structural modeling applications. The purpose of DROPBEAR is to evaluate models that take vibration data as input and give the initial conditions of the cantilever beam on which the measurements were taken as output. DROPBEAR is meant to serve an exemplar for emerging high-rate "active structures" that can be actively controlled with feedback latencies of less than one microsecond. Although the Euler-Bernoulli beam model is a well-known solution to this modeling problem, its computational cost is prohibitive for the time scales of interest. It has been previously shown that a properly structured LSTM network can achieve comparable accuracy with less workload, but achieving sub-microsecond model latency remains a challenge. Our approach is to deploy the LSTM optimized specifically for latency on FPGA. We designed the model using both high-level synthesis (HLS) and hardware description language (HDL). The lowest latency of 1.42 $\mu$S and the highest throughput of 7.87 Gops/s were achieved on Alveo U55C platform for HDL design.
doi_str_mv 10.48550/arxiv.2309.00801
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2309_00801</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2309_00801</sourcerecordid><originalsourceid>FETCH-LOGICAL-a671-7f4d80bb78796f7196cad908150a544535ff767afbad65b49a7f93b88788021f3</originalsourceid><addsrcrecordid>eNotj0tOwzAUAL1hgQoHYIUv4PBcf569rMqnSKkq0eyj59gulpKCkgiR2yMKq9mNZhi7k1BpZww80Phdvqq1Al8BOJDXDDddl_o00lzOJ14fm70INKXId-X0Lt5oTvxxOdNQOn5cpjkNfP8RUz_dsKtM_ZRu_7lizfNTs92J-vDyut3UgixKgVlHByGgQ28zSm87ih6cNEBGa6NMzmiRcqBoTdCeMHsVnEPnYC2zWrH7P-2lvP0cy0Dj0v4etJcD9QOFqj79</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Accelerating LSTM-based High-Rate Dynamic System Models</title><source>arXiv.org</source><creator>Kabir, Ehsan ; Coble, Daniel ; Satme, Joud N ; Downey, Austin R. J ; Bakos, Jason D ; Andrews, David ; Huang, Miaoqing</creator><creatorcontrib>Kabir, Ehsan ; Coble, Daniel ; Satme, Joud N ; Downey, Austin R. J ; Bakos, Jason D ; Andrews, David ; Huang, Miaoqing</creatorcontrib><description>In this paper, we evaluate the use of a trained Long Short-Term Memory (LSTM) network as a surrogate for a Euler-Bernoulli beam model, and then we describe and characterize an FPGA-based deployment of the model for use in real-time structural health monitoring applications. The focus of our efforts is the DROPBEAR (Dynamic Reproduction of Projectiles in Ballistic Environments for Advanced Research) dataset, which was generated as a benchmark for the study of real-time structural modeling applications. The purpose of DROPBEAR is to evaluate models that take vibration data as input and give the initial conditions of the cantilever beam on which the measurements were taken as output. DROPBEAR is meant to serve an exemplar for emerging high-rate "active structures" that can be actively controlled with feedback latencies of less than one microsecond. Although the Euler-Bernoulli beam model is a well-known solution to this modeling problem, its computational cost is prohibitive for the time scales of interest. It has been previously shown that a properly structured LSTM network can achieve comparable accuracy with less workload, but achieving sub-microsecond model latency remains a challenge. Our approach is to deploy the LSTM optimized specifically for latency on FPGA. We designed the model using both high-level synthesis (HLS) and hardware description language (HDL). The lowest latency of 1.42 $\mu$S and the highest throughput of 7.87 Gops/s were achieved on Alveo U55C platform for HDL design.</description><identifier>DOI: 10.48550/arxiv.2309.00801</identifier><language>eng</language><subject>Computer Science - Hardware Architecture</subject><creationdate>2023-09</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2309.00801$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2309.00801$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Kabir, Ehsan</creatorcontrib><creatorcontrib>Coble, Daniel</creatorcontrib><creatorcontrib>Satme, Joud N</creatorcontrib><creatorcontrib>Downey, Austin R. J</creatorcontrib><creatorcontrib>Bakos, Jason D</creatorcontrib><creatorcontrib>Andrews, David</creatorcontrib><creatorcontrib>Huang, Miaoqing</creatorcontrib><title>Accelerating LSTM-based High-Rate Dynamic System Models</title><description>In this paper, we evaluate the use of a trained Long Short-Term Memory (LSTM) network as a surrogate for a Euler-Bernoulli beam model, and then we describe and characterize an FPGA-based deployment of the model for use in real-time structural health monitoring applications. The focus of our efforts is the DROPBEAR (Dynamic Reproduction of Projectiles in Ballistic Environments for Advanced Research) dataset, which was generated as a benchmark for the study of real-time structural modeling applications. The purpose of DROPBEAR is to evaluate models that take vibration data as input and give the initial conditions of the cantilever beam on which the measurements were taken as output. DROPBEAR is meant to serve an exemplar for emerging high-rate "active structures" that can be actively controlled with feedback latencies of less than one microsecond. Although the Euler-Bernoulli beam model is a well-known solution to this modeling problem, its computational cost is prohibitive for the time scales of interest. It has been previously shown that a properly structured LSTM network can achieve comparable accuracy with less workload, but achieving sub-microsecond model latency remains a challenge. Our approach is to deploy the LSTM optimized specifically for latency on FPGA. We designed the model using both high-level synthesis (HLS) and hardware description language (HDL). The lowest latency of 1.42 $\mu$S and the highest throughput of 7.87 Gops/s were achieved on Alveo U55C platform for HDL design.</description><subject>Computer Science - Hardware Architecture</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj0tOwzAUAL1hgQoHYIUv4PBcf569rMqnSKkq0eyj59gulpKCkgiR2yMKq9mNZhi7k1BpZww80Phdvqq1Al8BOJDXDDddl_o00lzOJ14fm70INKXId-X0Lt5oTvxxOdNQOn5cpjkNfP8RUz_dsKtM_ZRu_7lizfNTs92J-vDyut3UgixKgVlHByGgQ28zSm87ih6cNEBGa6NMzmiRcqBoTdCeMHsVnEPnYC2zWrH7P-2lvP0cy0Dj0v4etJcD9QOFqj79</recordid><startdate>20230901</startdate><enddate>20230901</enddate><creator>Kabir, Ehsan</creator><creator>Coble, Daniel</creator><creator>Satme, Joud N</creator><creator>Downey, Austin R. J</creator><creator>Bakos, Jason D</creator><creator>Andrews, David</creator><creator>Huang, Miaoqing</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230901</creationdate><title>Accelerating LSTM-based High-Rate Dynamic System Models</title><author>Kabir, Ehsan ; Coble, Daniel ; Satme, Joud N ; Downey, Austin R. J ; Bakos, Jason D ; Andrews, David ; Huang, Miaoqing</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a671-7f4d80bb78796f7196cad908150a544535ff767afbad65b49a7f93b88788021f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Hardware Architecture</topic><toplevel>online_resources</toplevel><creatorcontrib>Kabir, Ehsan</creatorcontrib><creatorcontrib>Coble, Daniel</creatorcontrib><creatorcontrib>Satme, Joud N</creatorcontrib><creatorcontrib>Downey, Austin R. J</creatorcontrib><creatorcontrib>Bakos, Jason D</creatorcontrib><creatorcontrib>Andrews, David</creatorcontrib><creatorcontrib>Huang, Miaoqing</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Kabir, Ehsan</au><au>Coble, Daniel</au><au>Satme, Joud N</au><au>Downey, Austin R. J</au><au>Bakos, Jason D</au><au>Andrews, David</au><au>Huang, Miaoqing</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Accelerating LSTM-based High-Rate Dynamic System Models</atitle><date>2023-09-01</date><risdate>2023</risdate><abstract>In this paper, we evaluate the use of a trained Long Short-Term Memory (LSTM) network as a surrogate for a Euler-Bernoulli beam model, and then we describe and characterize an FPGA-based deployment of the model for use in real-time structural health monitoring applications. The focus of our efforts is the DROPBEAR (Dynamic Reproduction of Projectiles in Ballistic Environments for Advanced Research) dataset, which was generated as a benchmark for the study of real-time structural modeling applications. The purpose of DROPBEAR is to evaluate models that take vibration data as input and give the initial conditions of the cantilever beam on which the measurements were taken as output. DROPBEAR is meant to serve an exemplar for emerging high-rate "active structures" that can be actively controlled with feedback latencies of less than one microsecond. Although the Euler-Bernoulli beam model is a well-known solution to this modeling problem, its computational cost is prohibitive for the time scales of interest. It has been previously shown that a properly structured LSTM network can achieve comparable accuracy with less workload, but achieving sub-microsecond model latency remains a challenge. Our approach is to deploy the LSTM optimized specifically for latency on FPGA. We designed the model using both high-level synthesis (HLS) and hardware description language (HDL). The lowest latency of 1.42 $\mu$S and the highest throughput of 7.87 Gops/s were achieved on Alveo U55C platform for HDL design.</abstract><doi>10.48550/arxiv.2309.00801</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2309.00801
ispartof
issn
language eng
recordid cdi_arxiv_primary_2309_00801
source arXiv.org
subjects Computer Science - Hardware Architecture
title Accelerating LSTM-based High-Rate Dynamic System Models
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-04T14%3A09%3A43IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Accelerating%20LSTM-based%20High-Rate%20Dynamic%20System%20Models&rft.au=Kabir,%20Ehsan&rft.date=2023-09-01&rft_id=info:doi/10.48550/arxiv.2309.00801&rft_dat=%3Carxiv_GOX%3E2309_00801%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true