scispace - formally typeset
Search or ask a question
Author

Suhas Suresh

Bio: Suhas Suresh is an academic researcher from Siemens. The author has contributed to research in topics: Process control & Cyber-physical system. The author has an hindex of 1, co-authored 2 publications receiving 4 citations.

Papers
More filters
Proceedings ArticleDOI
01 Aug 2015
TL;DR: A Virtual Testing Environment (VTE) for process industry CPS performance assessment based on Amesim simulation software is presented and a set of innovative techniques are used to effectively search for the best and worst system performances within the unlimited solution space.
Abstract: Process industry's modern process control valves are complex Cyber-Physical Systems (CPS) A valve positioner may be connected to thousands of different types of process valves, and operate under a large variety of environmental conditions; therefore it is unfeasible to conduct exhaustive physical tests The end user expects the system to work under any conditions and configurations and deliver high performance whenever required This paper presents a Virtual Testing Environment (VTE) for process industry CPS performance assessment based on Amesim simulation software A set of innovative techniques are used to effectively search for the best and worst system performances within the unlimited solution space In order to speed up the simulation, we combine model order reduction techniques with an extensible optimizer using with different solvers as Amesim plug-ins A Scheduler was designed to execute simulation tasks on multiple servers in parallel This paper presents the preliminary results based on Particle Swarm Optimization (PSO) and over 5 billion configurations

3 citations

Proceedings ArticleDOI
30 Aug 2018
TL;DR: This work proposes using L moments to model the spread of data and shows that the proposed approach works better than the classical robust design formulation.
Abstract: Uncertainties in the input variables are inevitable in any design process. As a consequence, the output responses are also uncertain. Robust design is one of the sought after approach to treat such uncertainties for controlling the variation in the output responses, while maximizing the mean performance. Variation is modeled by a measure of data spread. Often, the details of the uncertainties in the input space are not available readily and they are usually characterized from scarce sample realizations. In addition, there could also be outliers in the realizations. These will increase the error in the measure of spread of the output response. Hence, it is desirable that an approach that is insensitive to outliers but can characterize the spread of data is developed for robust design. In this work we propose using L moments to model the spread of data. The classical robust design formulation is reformulated using the second L moment (l2). The proposed approach is demonstrated on a turbine disk design with 17 design and random variables. The details of the uncertainties are not known. A DoE of 200 samples is used and at each DoE point, we propagate the uncertainties using scarce samples, which include outliers. Robust design is performed and it is shown that the proposed approach works better than the classical robust design formulation.

3 citations


Cited by
More filters
Proceedings ArticleDOI
22 Mar 2021
TL;DR: In this paper, the authors investigated the use of Machine Learning algorithms to predict acceptable behavior in system performance of a new software release, based on previous knowledge and actual run-time information, the proposed approach predicts the response time that can be considered acceptable for the new software version, and used to identify problematic releases.
Abstract: Remote software deployment and updating has long been commonplace in many different fields, but now, the increasing expansion of IoT and CPSoS (Cyber-Physcal System of Systems) has highlighted the need for additional mechanisms in these systems, to ensure the correct behaviour of the deployed software version after deployment. In this sense, this paper investigates the use of Machine Learning algorithms to predict acceptable behaviour in system performance of a new software release. By monitoring the real performance, eventual unexpected problems can be identified. Based on previous knowledge and actual run-time information, the proposed approach predicts the response time that can be considered acceptable for the new software release, and this information is used to identify problematic releases. The mechanism has been applied to the post-deployment monitoring of traffic algorithms in elevator systems. To evaluate the approach, we have used performance mutation testing, obtaining good results. This paper makes two contributions. First, it proposes several regression learners that have been trained with different types of traffic profiles to efficiently predict response time of the traffic dispatching algorithm. This prediction is then compared with the actual response time of the new algorithm release, and provides a verdict about its performance. Secondly, a comparison of the different learners is performed.

8 citations

Journal ArticleDOI
TL;DR: L-moment ratio diagram that uses higher order L-moments is adopted to choose the appropriate distribution, for uncertainty quantification and the probabilistic estimates obtained are found to be less sensitive to the extremes in the data, compared to the results obtained from the conventional moments approach.
Abstract: Sampling-based uncertainty quantification demands large data. Hence, when the available sample is scarce, it is customary to assume a distribution and estimate its moments from scarce data, to characterize the uncertainties. Nonetheless, inaccurate assumption about the distribution leads to flawed decisions. In addition, extremes, if present in the scarce data, are prone to be classified as outliers and neglected which leads to wrong estimation of the moments. Therefore, it is desirable to develop a method that is (i) distribution independent or allows distribution identification with scarce samples and (ii) accounts for the extremes in data and yet be insensitive or less sensitive to moments estimation. We propose using L-moments to develop a distribution-independent, robust moment estimation approach to characterize the uncertainty and propagate it through the system model. L-moment ratio diagram that uses higher order L-moments is adopted to choose the appropriate distribution, for uncertainty quantification. This allows for better characterization of the output distribution and the probabilistic estimates obtained using L-moments are found to be less sensitive to the extremes in the data, compared to the results obtained from the conventional moments approach. The efficacy of the proposed approach is demonstrated on conventional distributions covering all types of tails and several engineering examples. Engineering examples include a sheet metal manufacturing process, 7 variable speed reducer, and probabilistic fatigue life estimation.

3 citations

Proceedings ArticleDOI
25 May 2021
TL;DR: In this paper, the authors propose a microservice-based method to detect performance problems in Cyber-Physical Systems (CPSs) based on Machine Learning algorithms, which predict the performance of a new software release based on knowledge from previous releases.
Abstract: Software embedded in Cyber-Physical Systems (CPSs) usually has a large life-cycle and is continuously evolving. The increasing expansion of IoT and CPSs has highlighted the need for additional mechanisms for remote deployment and updating of this software, to ensure its correct behaviour. Performance problems require special attention, as they may appear in operation due to limitations in lab testing and environmental conditions. In this context, we propose a microservice-based method to detect performance problems in CPSs. These microservices will be deployed in installation to detect performance problems in run-time when new software versions are deployed. The problem detection is based on Machine Learning algorithms, which predict the performance of a new software release based on knowledge from previous releases. This permits taking corrective actions so that system reliability is guaranteed.

2 citations

Proceedings ArticleDOI
15 Aug 2021
TL;DR: In this article, a catalogo de requisitos for CPSs in the context of industrial 4.0 is presented, with the goal of providing a guia for desenvolvimento e aprimoramento of the CPSs.
Abstract: Os Sistemas Ciber-Fisicos (CPS), mesmo com beneficios em termos de automacao e gestao na industria 4.0, apresentam desafios que precisam ser explorados - como a ausencia de padronizacao na definicao das caracteristicas e requisitos de seu desenvolvimento. Acreditamos que a definicao de um catalogo de requisitos especificos para estes sistemas e uma das etapas relevantes para garantir uma maior adequacao e qualidade das praticas industriais. Assim, este projeto tem como objetivo propor um catalogo de requisitos para os CPS no contexto da industria 4.0, servindo como um guia para o desenvolvimento e aprimoramento destes sistemas. O catalogo foi desenvolvido com base na execucao de uma revisao sistematica e na experiencia de especialistas (incluindo experiencias de avaliacao e elicitacao para os requisitos do CPS). O objetivo deste catalogo e auxiliar os usuarios, gerentes ou diretores, a comparar, selecionar e aprimorar o entendimento sobre este tipo de sistema - sendo compativel com os criterios produtivos, socioculturais, funcionais e economicos, garantindo maior qualidade e padronizacao para as praticas da industria 4.0.