Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247595
Luís Silva, Pedro Gonçalves, R. Marau, P. Pedreiras, L. Almeida
Emerging concepts such as Smart Production, Industrial Internet of Things and Industry 4.0 bring a radically new set of requirements to the way industrial systems are engineered. In what concerns the communication infrastructure, support to dynamic environments, interoperability and heterogeneity, combined with a significant increase in the number of devices, are just a few of the challenges that must be faced. Software-defined networking is a disruptive networking paradigm that emerged on campus networks but was soon considered for use at industrial level. This paper presents a set of extensions to the Software Defined Networking (SDN) OpenFlow protocol that complement its functionality, namely supporting real-time reservations, which is one of its more notorious limitations when considering industrial scenarios. We explain how the extensions are implemented in the OpenFlow side and enforced using a Flexible Time-Triggered Ethernet network. The extensions are validated experimentally, showing that the platform supports dynamically reconfigurable heterogeneous traffic classes.
{"title":"Extending OpenFlow with flexible time-triggered real-time communication services","authors":"Luís Silva, Pedro Gonçalves, R. Marau, P. Pedreiras, L. Almeida","doi":"10.1109/ETFA.2017.8247595","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247595","url":null,"abstract":"Emerging concepts such as Smart Production, Industrial Internet of Things and Industry 4.0 bring a radically new set of requirements to the way industrial systems are engineered. In what concerns the communication infrastructure, support to dynamic environments, interoperability and heterogeneity, combined with a significant increase in the number of devices, are just a few of the challenges that must be faced. Software-defined networking is a disruptive networking paradigm that emerged on campus networks but was soon considered for use at industrial level. This paper presents a set of extensions to the Software Defined Networking (SDN) OpenFlow protocol that complement its functionality, namely supporting real-time reservations, which is one of its more notorious limitations when considering industrial scenarios. We explain how the extensions are implemented in the OpenFlow side and enforced using a Flexible Time-Triggered Ethernet network. The extensions are validated experimentally, showing that the platform supports dynamically reconfigurable heterogeneous traffic classes.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"36 1","pages":"1-8"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78464620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247703
Felix Auris, S. Süss, Andreas Schlag, C. Diedrich
Nowadays typical validation in the development process of automated production plants includes Virtual Commissioning (VC) to test the control logic against the modelled plant behaviour. To reduce modelling efforts, the generation of models for VC is in practical often based on the existing control logic, e.g. hardware configuration of the PLC (used for automated signal mapping). This results in the limitation of testing at the end of the development process with resulting disadvantages like late detection of errors. In the meantime early automation concept evaluations are mainly done by empirical knowledge due to the lack of reliable models. In this paper we will present a component model and work-flow for coupling a 3D-CAD tool with a behavioural model of a mechatronic component for generating an initial virtual plant model allowing examining the feasibility of the selected components already during the mechanic design phase. The created model could also be used for iterative development of the control logic with an comprehensive VC at the end of the process utilizing the already developed model and thus reducing modelling efforts and enabling shorter validation cycles throughout the whole process.
{"title":"Towards shorter validation cycles by considering mechatronic component behaviour in early design stages","authors":"Felix Auris, S. Süss, Andreas Schlag, C. Diedrich","doi":"10.1109/ETFA.2017.8247703","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247703","url":null,"abstract":"Nowadays typical validation in the development process of automated production plants includes Virtual Commissioning (VC) to test the control logic against the modelled plant behaviour. To reduce modelling efforts, the generation of models for VC is in practical often based on the existing control logic, e.g. hardware configuration of the PLC (used for automated signal mapping). This results in the limitation of testing at the end of the development process with resulting disadvantages like late detection of errors. In the meantime early automation concept evaluations are mainly done by empirical knowledge due to the lack of reliable models. In this paper we will present a component model and work-flow for coupling a 3D-CAD tool with a behavioural model of a mechatronic component for generating an initial virtual plant model allowing examining the feasibility of the selected components already during the mechanic design phase. The created model could also be used for iterative development of the control logic with an comprehensive VC at the end of the process utilizing the already developed model and thus reducing modelling efforts and enabling shorter validation cycles throughout the whole process.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"59 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77009628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247729
D. Tandur, M. Gandhi, Himashri Kour, Rahul N. Gore
With the emergence of IoT and low power wireless technologies, many of the factory floor devices now have wireless interfaces. Bluetooth wireless technology is increasingly being used in these devices for communicating device data to a floor operator. The operator brings a Bluetooth enabled mobile device such as a tablet from where the respective devices can be monitored or controlled. As Bluetooth has only a limited coverage, an operator has to bring the mobile device within a close proximity of the device. With the increase in the number of Bluetooth enabled factory devices, the task of device synchronization can become cumbersome and time consuming. In this paper we demonstrate an IoT infrastructure solution for factory environment that leverages the presence of Bluetooth enabled factory floor devices along with additional Bluetooth and Wi-Fi infrastructure nodes in order to provide context based data to the factory floor personnel. The proposed IoT platform gathers data over the entire factory floor in an automated fashion resulting in additional services that will aid in improving efficiency on the factory floor.
{"title":"An IoT infrastructure solution for factories","authors":"D. Tandur, M. Gandhi, Himashri Kour, Rahul N. Gore","doi":"10.1109/ETFA.2017.8247729","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247729","url":null,"abstract":"With the emergence of IoT and low power wireless technologies, many of the factory floor devices now have wireless interfaces. Bluetooth wireless technology is increasingly being used in these devices for communicating device data to a floor operator. The operator brings a Bluetooth enabled mobile device such as a tablet from where the respective devices can be monitored or controlled. As Bluetooth has only a limited coverage, an operator has to bring the mobile device within a close proximity of the device. With the increase in the number of Bluetooth enabled factory devices, the task of device synchronization can become cumbersome and time consuming. In this paper we demonstrate an IoT infrastructure solution for factory environment that leverages the presence of Bluetooth enabled factory floor devices along with additional Bluetooth and Wi-Fi infrastructure nodes in order to provide context based data to the factory floor personnel. The proposed IoT platform gathers data over the entire factory floor in an automated fashion resulting in additional services that will aid in improving efficiency on the factory floor.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"36 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85429483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247727
Steffen Pfrang, David Meier, Valentin Kautz
Industrial automation and control systems (IACS) play a key role in modern production facilities. On the one hand, they provide real-time functionality to the connected field devices. On the other hand, they get more and more connected to local networks and the internet in order to facilitate use cases promoted by “Industry 4.0”. This makes IACS susceptible to cyber-attacks which exploit vulnerabilities, for example in order to interrupt the automation process. Security testing targets at discovering those vulnerabilities before they are exploited. In order to enable IACS manufacturers and integrators to perform security testing for their devices, we present ISuTest, a modular security testing framework for IACS. ISuTest is designed to be extendable regarding all kinds of automation protocols, different connection paths as well as evaluating arbitrary outputs of the tested devices. This paper describes the fundamental ideas behind ISuTest, its design and a basic evaluation in which the ISuTest framework was able to discover a vulnerability in a programmable logic controller (PLC). The paper concludes with a broad overview of the planned future work.
{"title":"Towards a modular security testing framework for industrial automation and control systems: ISuTest","authors":"Steffen Pfrang, David Meier, Valentin Kautz","doi":"10.1109/ETFA.2017.8247727","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247727","url":null,"abstract":"Industrial automation and control systems (IACS) play a key role in modern production facilities. On the one hand, they provide real-time functionality to the connected field devices. On the other hand, they get more and more connected to local networks and the internet in order to facilitate use cases promoted by “Industry 4.0”. This makes IACS susceptible to cyber-attacks which exploit vulnerabilities, for example in order to interrupt the automation process. Security testing targets at discovering those vulnerabilities before they are exploited. In order to enable IACS manufacturers and integrators to perform security testing for their devices, we present ISuTest, a modular security testing framework for IACS. ISuTest is designed to be extendable regarding all kinds of automation protocols, different connection paths as well as evaluating arbitrary outputs of the tested devices. This paper describes the fundamental ideas behind ISuTest, its design and a basic evaluation in which the ISuTest framework was able to discover a vulnerability in a programmable logic controller (PLC). The paper concludes with a broad overview of the planned future work.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"15 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88033719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247674
Dimitrios Amaxilatis, O. Akrivopoulos, I. Chatzigiannakis, C. Tselios
The world of machine-to-machine (M2M) communication is gradually moving from vertical single purpose solutions to multi-purpose and collaborative applications interacting across industry verticals, organizations and people — a world of Internet of Things (IoT). The dominant approach for delivering IoT applications relies on the development of cloud-based IoT platforms that collect all the data generated by the sensing elements and centrally process the information to create real business value. In this paper, we present a system that follows the Fog Computing paradigm where the sensor resources, as well as the intermediate layers between embedded devices and cloud computing datacenters, participate by providing computational, storage, and control. We discuss the design aspects of our system and present a pilot deployment for the evaluating the performance in a real-world environment. Our findings indicate that Fog Computing can address the ever-increasing amount of data that is inherent in an IoT world by effective communication among all elements of the architecture.
{"title":"Enabling stream processing for people-centric IoT based on the fog computing paradigm","authors":"Dimitrios Amaxilatis, O. Akrivopoulos, I. Chatzigiannakis, C. Tselios","doi":"10.1109/ETFA.2017.8247674","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247674","url":null,"abstract":"The world of machine-to-machine (M2M) communication is gradually moving from vertical single purpose solutions to multi-purpose and collaborative applications interacting across industry verticals, organizations and people — a world of Internet of Things (IoT). The dominant approach for delivering IoT applications relies on the development of cloud-based IoT platforms that collect all the data generated by the sensing elements and centrally process the information to create real business value. In this paper, we present a system that follows the Fog Computing paradigm where the sensor resources, as well as the intermediate layers between embedded devices and cloud computing datacenters, participate by providing computational, storage, and control. We discuss the design aspects of our system and present a pilot deployment for the evaluating the performance in a real-world environment. Our findings indicate that Fog Computing can address the ever-increasing amount of data that is inherent in an IoT world by effective communication among all elements of the architecture.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"4 1","pages":"1-8"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90247110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247742
A. Pfeffer, L. Urbas
Modular process plants are small or medium scale plants which consist of separately engineered and automated modules. To speed up the start of production, a modular process plant is composed from several of such modules. Nowadays, safety engineering refers to the whole process plant, is very individual, time consuming, and hardly integrated. For safe modular process plants, the safety engineering has to be performed and implemented on module level to keep the engineering processes of the modules and the modular process plant independent. The safety engineering has to be more integrated to keep the advantage of modular process plants. This paper presents the challenges of such a modular safety engineering and our approach to cope with them. We use a case study of an exemplary modular process plant to show the results of a modular HAZOP study with limited information and how to combine the modular HAZOP studies to fill the gaps.
{"title":"HAZOP studies for engineering safe modular process plants","authors":"A. Pfeffer, L. Urbas","doi":"10.1109/ETFA.2017.8247742","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247742","url":null,"abstract":"Modular process plants are small or medium scale plants which consist of separately engineered and automated modules. To speed up the start of production, a modular process plant is composed from several of such modules. Nowadays, safety engineering refers to the whole process plant, is very individual, time consuming, and hardly integrated. For safe modular process plants, the safety engineering has to be performed and implemented on module level to keep the engineering processes of the modules and the modular process plant independent. The safety engineering has to be more integrated to keep the advantage of modular process plants. This paper presents the challenges of such a modular safety engineering and our approach to cope with them. We use a case study of an exemplary modular process plant to show the results of a modular HAZOP study with limited information and how to combine the modular HAZOP studies to fill the gaps.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"53 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85362051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247777
Javier Mesonero, C. Bielza, P. Larrañaga
Anomaly detection is an increasingly common task in many industrial environments. Cyber-physical systems stand out in this field due to their unique position in industrial areas. This paper introduces a new architecture aimed to detect anomalies in a real laser heating surface process, which is designed for field-programmable gate arrays (FPGAs). The FPGA design offers advantages of highly parallelized and pipelined architectures. The system will classify one process into normal or abnormal taking into account spatial information about where the laser spot is. The proposed design estimates a probability density function from data; then it performs an image convolution transforming the probability density function into a kernel density estimation function. This estimated function should be able to classify in real time.
{"title":"Architecture for anomaly detection in a laser heating surface process","authors":"Javier Mesonero, C. Bielza, P. Larrañaga","doi":"10.1109/ETFA.2017.8247777","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247777","url":null,"abstract":"Anomaly detection is an increasingly common task in many industrial environments. Cyber-physical systems stand out in this field due to their unique position in industrial areas. This paper introduces a new architecture aimed to detect anomalies in a real laser heating surface process, which is designed for field-programmable gate arrays (FPGAs). The FPGA design offers advantages of highly parallelized and pipelined architectures. The system will classify one process into normal or abnormal taking into account spatial information about where the laser spot is. The proposed design estimates a probability density function from data; then it performs an image convolution transforming the probability density function into a kernel density estimation function. This estimated function should be able to classify in real time.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"43 1","pages":"1-4"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83183898","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247713
S. Cavalieri, D. D. Stefano, Marco Giuseppe Salafia, Marco Stefano Scroppo
The paper presents a web-based platform able to offer access to OPC UA Servers. The proposed platform may be used by web-users to exchange data with OPC UA Server without any knowledge of the standard. The software solution described in the paper is available on GitHub.
{"title":"A web-based platform for OPC UA integration in IIoT environment","authors":"S. Cavalieri, D. D. Stefano, Marco Giuseppe Salafia, Marco Stefano Scroppo","doi":"10.1109/ETFA.2017.8247713","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247713","url":null,"abstract":"The paper presents a web-based platform able to offer access to OPC UA Servers. The proposed platform may be used by web-users to exchange data with OPC UA Server without any knowledge of the standard. The software solution described in the paper is available on GitHub.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"15 1","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81834445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247619
Gavneet Singh Chadha, Andreas Schwung
Process monitoring and fault diagnosis methods are used to detect abnormal events in industrial processes. Process breakdowns hinder the overall productivity of the system which makes the early detection of faults very critical. Due to the highly non-linear nature of modern industrial processes, deep neural networks with several layers of non-linear complex representations fit aptly for contemporary fault diagnosis. Although deep neural networks have found wide array of application areas such as image recognition and speech recognition, their effectiveness in fault detection has not been tested substantially. In this study, a comparison between two deep neural network architectures, namely Deep Stacking Networks and Sparse Stacked Autoencoders for fault detection from process data is presented. The Tennessee Eastman benchmark process is considered to test the effectiveness of these deep architectures. A detailed comparison between the two architectures is illustrated with different hyperparameters. The experiment results show that the Sparse Stacked Autoencoders model has superior average fault detection capability and is also more stable as it has less variation in fault detection rate.
{"title":"Comparison of deep neural network architectures for fault detection in Tennessee Eastman process","authors":"Gavneet Singh Chadha, Andreas Schwung","doi":"10.1109/ETFA.2017.8247619","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247619","url":null,"abstract":"Process monitoring and fault diagnosis methods are used to detect abnormal events in industrial processes. Process breakdowns hinder the overall productivity of the system which makes the early detection of faults very critical. Due to the highly non-linear nature of modern industrial processes, deep neural networks with several layers of non-linear complex representations fit aptly for contemporary fault diagnosis. Although deep neural networks have found wide array of application areas such as image recognition and speech recognition, their effectiveness in fault detection has not been tested substantially. In this study, a comparison between two deep neural network architectures, namely Deep Stacking Networks and Sparse Stacked Autoencoders for fault detection from process data is presented. The Tennessee Eastman benchmark process is considered to test the effectiveness of these deep architectures. A detailed comparison between the two architectures is illustrated with different hyperparameters. The experiment results show that the Sparse Stacked Autoencoders model has superior average fault detection capability and is also more stable as it has less variation in fault detection rate.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"59 1","pages":"1-8"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84601459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-09-01DOI: 10.1109/ETFA.2017.8247693
Stefan Windmann, Dorota Lang, O. Niggemann
A large part of the programmable logic controls (PLCs) used in industrial automation systems is based on automata, which are employed to model the different stages of the automated processes and to determine the discrete control signals. Complex PLCs are typically composed of several parallel automata, which are related to a subset of the IO signals, respectively. In this paper, a novel model learning approach is proposed, which allows to learn the parallel automata from the discrete IO signals during normal operation of the PLC. Learning the parallel automata is accomplished by means of a synchronous side-by-side decomposition of the overall system model. The side-by-side decomposition is based on the clustering of the correlation matrix computed between the individual IO signals. The learnt automata can be employed for automatic fault detection and visualization of the normal operation of the PLC. Evaluations are conducted for both a baseline method, where a single automaton is learned as model for the complete system, and the proposed learning algorithm for parallel automata. Experimental results show that the computed parallel automata are superior to a single automaton with respect to compactness, accuracy and fault detection capabilities.
{"title":"Learning parallel automata of PLCs","authors":"Stefan Windmann, Dorota Lang, O. Niggemann","doi":"10.1109/ETFA.2017.8247693","DOIUrl":"https://doi.org/10.1109/ETFA.2017.8247693","url":null,"abstract":"A large part of the programmable logic controls (PLCs) used in industrial automation systems is based on automata, which are employed to model the different stages of the automated processes and to determine the discrete control signals. Complex PLCs are typically composed of several parallel automata, which are related to a subset of the IO signals, respectively. In this paper, a novel model learning approach is proposed, which allows to learn the parallel automata from the discrete IO signals during normal operation of the PLC. Learning the parallel automata is accomplished by means of a synchronous side-by-side decomposition of the overall system model. The side-by-side decomposition is based on the clustering of the correlation matrix computed between the individual IO signals. The learnt automata can be employed for automatic fault detection and visualization of the normal operation of the PLC. Evaluations are conducted for both a baseline method, where a single automaton is learned as model for the complete system, and the proposed learning algorithm for parallel automata. Experimental results show that the computed parallel automata are superior to a single automaton with respect to compactness, accuracy and fault detection capabilities.","PeriodicalId":6522,"journal":{"name":"2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)","volume":"168 1","pages":"1-7"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83531611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：