Pub Date : 2017-05-01DOI: 10.1109/EAIS.2017.7954828
D. Dovžan
Evolving methods for on-line learning of nonlinear models can play an important role in future monitoring and fault detection systems. The ability to model nonlinear relationships between the measured variables and to adapt the model to changing variable relations can decrease the number of false alarms and ensure a more robust and stable monitoring system. In this paper an example of the waste water treatment process monitoring system based on evolving fuzzy model is presented.
{"title":"Evolving fuzzy model in fault detection system","authors":"D. Dovžan","doi":"10.1109/EAIS.2017.7954828","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954828","url":null,"abstract":"Evolving methods for on-line learning of nonlinear models can play an important role in future monitoring and fault detection systems. The ability to model nonlinear relationships between the measured variables and to adapt the model to changing variable relations can decrease the number of false alarms and ensure a more robust and stable monitoring system. In this paper an example of the waste water treatment process monitoring system based on evolving fuzzy model is presented.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127603634","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954829
G. Andonovski, B. Costa
This paper presents a practical implementation of the robust evolving controller RECCo for a two-tank pilot plant. The RECCo algorithm is a fuzzy PID type of controller and starts with empty parameters which are adapted in an online manner. Also the fuzzy cloud-based structure of the controller is initialized with the first data point received and evolves during time. The algorithm was additionally improved with introducing a protection of integral saturation (anti windup), which was necessary for this type of process. A real two-tank pilot plant was used to test the effectiveness of the controller. The process represents a real industrial environment through the OLE for Process Control (OPC) communication protocol. It has been demonstrated that the algorithm is capable of controlling the plant using the default values of the design parameters.
本文提出了一种鲁棒进化控制器RECCo在双罐中试装置中的实际实现。RECCo算法是一种模糊PID类型的控制器，以在线方式自适应的空参数开始。控制器的模糊云结构也随着接收到的第一个数据点初始化，并随着时间的推移而演变。该算法还通过引入积分饱和保护(防上卷)进行了改进，这是此类过程所必需的。在一个实际的双罐中试装置上测试了控制器的有效性。该过程通过OLE for process Control (OPC)通信协议代表了一个真实的工业环境。结果表明，该算法能够使用设计参数的默认值来控制对象。
{"title":"Robust evolving control of a two-tanks pilot plant","authors":"G. Andonovski, B. Costa","doi":"10.1109/EAIS.2017.7954829","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954829","url":null,"abstract":"This paper presents a practical implementation of the robust evolving controller RECCo for a two-tank pilot plant. The RECCo algorithm is a fuzzy PID type of controller and starts with empty parameters which are adapted in an online manner. Also the fuzzy cloud-based structure of the controller is initialized with the first data point received and evolves during time. The algorithm was additionally improved with introducing a protection of integral saturation (anti windup), which was necessary for this type of process. A real two-tank pilot plant was used to test the effectiveness of the controller. The process represents a real industrial environment through the OLE for Process Control (OPC) communication protocol. It has been demonstrated that the algorithm is capable of controlling the plant using the default values of the design parameters.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133549793","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954837
Zhao-Xu Yang, Hai-Jun Rong, Guangshe Zhao, Jing Yang
This paper presents a self-evolving kernel recursive least squares (KRLS) algorithm which implements the modelling of unknown nonlinear systems in reproducing kernel Hilbert spaces (RKHS). The prime motivation of this development is a reformulation of the well known KRLS algorithm which inevitably increases the computational complexity to the cases where data arrive sequentially. The self-evolving KRLS algorithm utilizes the measurement of kernel evaluation and adaptive approximation error to determine the learning system with a structure of a suitable size that involves recruiting and dimension reduction of the kernel vector during the adaptive learning phase without predefining them. This self-evolving procedure allows the algorithm to operate online, often in real time, reducing the computational time and improving the learning performance. This algorithm is finally utilized in the applications of online adaptive control and time series prediction where the system is described as a unknown function by Nonlinear AutoRegressive with Exogenous inputs model. Simulation results from an inverted pendulum system and Time Series Data Library demonstrate the satisfactory performance of the proposed self-evolving KRLS algorithm.
{"title":"Self-evolving kernel recursive least squares algorithm for control and prediction","authors":"Zhao-Xu Yang, Hai-Jun Rong, Guangshe Zhao, Jing Yang","doi":"10.1109/EAIS.2017.7954837","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954837","url":null,"abstract":"This paper presents a self-evolving kernel recursive least squares (KRLS) algorithm which implements the modelling of unknown nonlinear systems in reproducing kernel Hilbert spaces (RKHS). The prime motivation of this development is a reformulation of the well known KRLS algorithm which inevitably increases the computational complexity to the cases where data arrive sequentially. The self-evolving KRLS algorithm utilizes the measurement of kernel evaluation and adaptive approximation error to determine the learning system with a structure of a suitable size that involves recruiting and dimension reduction of the kernel vector during the adaptive learning phase without predefining them. This self-evolving procedure allows the algorithm to operate online, often in real time, reducing the computational time and improving the learning performance. This algorithm is finally utilized in the applications of online adaptive control and time series prediction where the system is described as a unknown function by Nonlinear AutoRegressive with Exogenous inputs model. Simulation results from an inverted pendulum system and Time Series Data Library demonstrate the satisfactory performance of the proposed self-evolving KRLS algorithm.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129288373","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954825
I. Škrjanc, A. Sanchis, J. A. Iglesias, Agapito Ledezma, D. Dovžan
In this paper the idea of evolving Cauchy clustering based on cosine similarity is given. It is used for monitoring in the case of cyber attacks. The proposed idea is for that kind of processes very interesting because it is very efficient when the data are noisy and when the outliers appear frequently and this is the case when dealing with cyber attacks data. The algorithm is given in an evolving form to be able to deal with big-data sets. One of the important features of the described clustering algorithm is that it deals with just few tuning parameters, such as maximal density. In this paper, the proposed approach was realized on DARPA data base and promising results have been achieved.
{"title":"Evolving Cauchy possibilistic clustering based on cosine similarity for monitoring cyber systems","authors":"I. Škrjanc, A. Sanchis, J. A. Iglesias, Agapito Ledezma, D. Dovžan","doi":"10.1109/EAIS.2017.7954825","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954825","url":null,"abstract":"In this paper the idea of evolving Cauchy clustering based on cosine similarity is given. It is used for monitoring in the case of cyber attacks. The proposed idea is for that kind of processes very interesting because it is very efficient when the data are noisy and when the outliers appear frequently and this is the case when dealing with cyber attacks data. The algorithm is given in an evolving form to be able to deal with big-data sets. One of the important features of the described clustering algorithm is that it deals with just few tuning parameters, such as maximal density. In this paper, the proposed approach was realized on DARPA data base and promising results have been achieved.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115318720","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954822
G. Černe
This paper introduces adaptive TS model developed with upgraded recursive Gustafson-Kessel (rGK) clustering in the field of short-term load forecasting (STLF), which is one of the most essential parts for electrical distributors. The problem of STLF is to forecast load consumption for a day ahead based on the weather forecast and the type of the day. Until now, most of the forecasting methods based on fuzzy logic needed a lot of expert knowledge to build and adapt the model, where rGK clustering lowers the need of this expert knowledge because of the automatic partitioning of the domain. In addition to rGK clustering, proposed solution also moves from directly forecasting the average load to forecasting the change of load from current to the next day, which is the fastest way to adapt the model to the change in electrical load system. To improve domain separation of clustering, improved membership function based both on input and output distance is also proposed.
{"title":"Introduction of adaptive TS model using recursive Gustafson-Kessel algorithm in short term load forecasting","authors":"G. Černe","doi":"10.1109/EAIS.2017.7954822","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954822","url":null,"abstract":"This paper introduces adaptive TS model developed with upgraded recursive Gustafson-Kessel (rGK) clustering in the field of short-term load forecasting (STLF), which is one of the most essential parts for electrical distributors. The problem of STLF is to forecast load consumption for a day ahead based on the weather forecast and the type of the day. Until now, most of the forecasting methods based on fuzzy logic needed a lot of expert knowledge to build and adapt the model, where rGK clustering lowers the need of this expert knowledge because of the automatic partitioning of the domain. In addition to rGK clustering, proposed solution also moves from directly forecasting the average load to forecasting the change of load from current to the next day, which is the fastest way to adapt the model to the change in electrical load system. To improve domain separation of clustering, improved membership function based both on input and output distance is also proposed.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125561926","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954842
Marsela Polic, Ziad Salem, Karlo Griparic, S. Bogdan, T. Schmickl
As a part of ASSISIbf project, with a final goal of forming a collective adaptive bio-hybrid society of animals and robots, an artificial neural network based on LSTM architecture was designed and trained for bee density estimation. During experiments, the bees are placed inside a plastic arena covered with wax, where they interact with and adapt to specialized static robotic units, CASUs, designed specially for this project. In order to interact with honeybees, the CASUs require the capability i) to produce and perceive the stimuli, i.e., environmental cues, that are relevant to honeybee behaviour, and ii) to sense the honeybees presence. The second requirement is implemented through 6 proximity sensors mounted on the upper part of CASU. In this paper we present estimation of honeybees (moving agents) density in 2D space (experimental arena) that is based on LSTM neural network. When compared to previous work done in this field, experiments demonstrate satisfactory results in estimating sizes of bee groups placed in the arena within a larger scope of outputs. Two different approaches were tested: regression and classification, with classification yielding higher accuracy.
{"title":"Estimation of moving agents density in 2D space based on LSTM neural network","authors":"Marsela Polic, Ziad Salem, Karlo Griparic, S. Bogdan, T. Schmickl","doi":"10.1109/EAIS.2017.7954842","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954842","url":null,"abstract":"As a part of ASSISIbf project, with a final goal of forming a collective adaptive bio-hybrid society of animals and robots, an artificial neural network based on LSTM architecture was designed and trained for bee density estimation. During experiments, the bees are placed inside a plastic arena covered with wax, where they interact with and adapt to specialized static robotic units, CASUs, designed specially for this project. In order to interact with honeybees, the CASUs require the capability i) to produce and perceive the stimuli, i.e., environmental cues, that are relevant to honeybee behaviour, and ii) to sense the honeybees presence. The second requirement is implemented through 6 proximity sensors mounted on the upper part of CASU. In this paper we present estimation of honeybees (moving agents) density in 2D space (experimental arena) that is based on LSTM neural network. When compared to previous work done in this field, experiments demonstrate satisfactory results in estimating sizes of bee groups placed in the arena within a larger scope of outputs. Two different approaches were tested: regression and classification, with classification yielding higher accuracy.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116838203","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954840
Suleyman Gemici, T. Gabel, Benjamin Loffler, A. Tharwat
Self-adaptation and the ability to assimilate new knowledge are two fundamental characteristics of intelligent systems. In this paper we leverage methods from evolutionary optimization and from case-based reasoning to construct an agent that is able to evolve in such a way that it is able to successfully master the popular video game Angry Birds.
{"title":"Combining evolutionary algorithms and case-based reasoning for learning high-quality shooting strategies in AI birds","authors":"Suleyman Gemici, T. Gabel, Benjamin Loffler, A. Tharwat","doi":"10.1109/EAIS.2017.7954840","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954840","url":null,"abstract":"Self-adaptation and the ability to assimilate new knowledge are two fundamental characteristics of intelligent systems. In this paper we leverage methods from evolutionary optimization and from case-based reasoning to construct an agent that is able to evolve in such a way that it is able to successfully master the popular video game Angry Birds.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121562161","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954827
Eckhard Gauterin, F. Pöschke, Nico Goldschmidt, H. Schulte
A new approach for optimal observer design of nonlinear systems, the so-called Nonlinear Quadratic Estimator is proposed. This approach employs the minimisation of a quadratic cost functional, thereby comprising two design parameters: Selective weighting of specific error state components and estimated upper bound minimisation. The new approach works without dual system transformation, achieving significant error state minimisation with optimised error dynamics and enabling a selective error state minimisation. Within this proceeding the observer and estimator design method, respectively, is derived from a Lyapunov stability condition of nonlinear, time-continuous systems in Takagi-Sugeno model structure, solved with linear matrix inequalities. Its capability is illustrated for an academical example of a nonlinear system with observer based stabilisation.
{"title":"Nonlinear Quadratic Estimator with selective error state weighting","authors":"Eckhard Gauterin, F. Pöschke, Nico Goldschmidt, H. Schulte","doi":"10.1109/EAIS.2017.7954827","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954827","url":null,"abstract":"A new approach for optimal observer design of nonlinear systems, the so-called Nonlinear Quadratic Estimator is proposed. This approach employs the minimisation of a quadratic cost functional, thereby comprising two design parameters: Selective weighting of specific error state components and estimated upper bound minimisation. The new approach works without dual system transformation, achieving significant error state minimisation with optimised error dynamics and enabling a selective error state minimisation. Within this proceeding the observer and estimator design method, respectively, is derived from a Lyapunov stability condition of nonlinear, time-continuous systems in Takagi-Sugeno model structure, solved with linear matrix inequalities. Its capability is illustrated for an academical example of a nonlinear system with observer based stabilisation.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"210 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116523571","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954834
Matevž Bošnak
This paper is accompanying the proposed implementation of the updated Evolving Principle Component Clustering (EPCC) algorithm for segmenting LRF (laser range finder) measurements into linear prototypes. The paper describes the target application for the algorithm, the algorithm itself and its implementation in C++ using Qt framework. The implementation is provided for both the proposed EPCC algorithm as well as for the popular split-and-merge (SAM) line segmenting algorithm and comparison is given in terms of computational complexity and results quality. The evolving nature of the proposed algorithm is most expressed in clustering approach itself and an on-line adaptation of cluster membership thresholds based on data observed in the past. The results conclusively show improvement over SAM in both the processing load and its stability in terms of low variations in how long the algorithm take to cluster various data sets.
{"title":"Evolving principal component clustering for 2-D LIDAR data","authors":"Matevž Bošnak","doi":"10.1109/EAIS.2017.7954834","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954834","url":null,"abstract":"This paper is accompanying the proposed implementation of the updated Evolving Principle Component Clustering (EPCC) algorithm for segmenting LRF (laser range finder) measurements into linear prototypes. The paper describes the target application for the algorithm, the algorithm itself and its implementation in C++ using Qt framework. The implementation is provided for both the proposed EPCC algorithm as well as for the popular split-and-merge (SAM) line segmenting algorithm and comparison is given in terms of computational complexity and results quality. The evolving nature of the proposed algorithm is most expressed in clustering approach itself and an on-line adaptation of cluster membership thresholds based on data observed in the past. The results conclusively show improvement over SAM in both the processing load and its stability in terms of low variations in how long the algorithm take to cluster various data sets.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126826706","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 : 2017-05-01DOI: 10.1109/EAIS.2017.7954838
Jose Eduardo Gomez, David Camilo Corrales, J. Corrales, A. Sanchis, Agapito Ledezma, J. A. Iglesias
In this paper we present a large set of data obtained from the volcanic surveillance of Puracé volcano (in Colombia). The proposed data are the result of the real-time data from the seismological stations which are close to this volcano. These data are extracted from the Colombian Geological Survey (SGC) and we have processed all of them in order to create the proposed dataset. This dataset will help to test if a learning algorithm can learn quickly and if it can extract knowledge from data streams in real time. In the proposed link, the presented dataset is available for any researcher.
{"title":"Monitoring of vulcano Puracé through seismic signals: Description of a real dataset","authors":"Jose Eduardo Gomez, David Camilo Corrales, J. Corrales, A. Sanchis, Agapito Ledezma, J. A. Iglesias","doi":"10.1109/EAIS.2017.7954838","DOIUrl":"https://doi.org/10.1109/EAIS.2017.7954838","url":null,"abstract":"In this paper we present a large set of data obtained from the volcanic surveillance of Puracé volcano (in Colombia). The proposed data are the result of the real-time data from the seismological stations which are close to this volcano. These data are extracted from the Colombian Geological Survey (SGC) and we have processed all of them in order to create the proposed dataset. This dataset will help to test if a learning algorithm can learn quickly and if it can extract knowledge from data streams in real time. In the proposed link, the presented dataset is available for any researcher.","PeriodicalId":286312,"journal":{"name":"2017 Evolving and Adaptive Intelligent Systems (EAIS)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122839515","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：