Pub Date : 2017-12-01DOI: 10.1109/ICMLA.2017.0-122
R. A. Bridges, Kelly M. T. Huffer, Corinne L. Jones, Michael D. Iannacone, J. Goodall
We address a crucial element of applied information extraction—accurate identification of basic security entities in text-—by evaluating previous methods and presenting new labelers. Our survey reveals that the previous efforts have not been tested on documents similar to the targeted sources (news articles, blogs, tweets, etc.) and that no sufficiently large publicly available annotated corpus of these documents exists. By assembling a representative test corpus, we perform a quantitative evaluation of previous methods in a realistic setting, revealing an overall lack of recall, and giving insight to the models' beneficial and inhibiting elements. In particular, our results show that many previous efforts overfit to the non-representative test corpora in this domain. Informed by this evaluation, we present three novel cyber entity extractors, which seek to leverage the available labeled data but remain worthwhile on the more diverse documents encountered in the wild. Each new model increases the state of the art in recall, with maximal or near maximal F1 score. Our results establish that the state of the art in cyber entity tagging is characterized by F1 = 0.61.
{"title":"Cybersecurity Automated Information Extraction Techniques: Drawbacks of Current Methods, and Enhanced Extractors","authors":"R. A. Bridges, Kelly M. T. Huffer, Corinne L. Jones, Michael D. Iannacone, J. Goodall","doi":"10.1109/ICMLA.2017.0-122","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.0-122","url":null,"abstract":"We address a crucial element of applied information extraction—accurate identification of basic security entities in text-—by evaluating previous methods and presenting new labelers. Our survey reveals that the previous efforts have not been tested on documents similar to the targeted sources (news articles, blogs, tweets, etc.) and that no sufficiently large publicly available annotated corpus of these documents exists. By assembling a representative test corpus, we perform a quantitative evaluation of previous methods in a realistic setting, revealing an overall lack of recall, and giving insight to the models' beneficial and inhibiting elements. In particular, our results show that many previous efforts overfit to the non-representative test corpora in this domain. Informed by this evaluation, we present three novel cyber entity extractors, which seek to leverage the available labeled data but remain worthwhile on the more diverse documents encountered in the wild. Each new model increases the state of the art in recall, with maximal or near maximal F1 score. Our results establish that the state of the art in cyber entity tagging is characterized by F1 = 0.61.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"126 1","pages":"437-442"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73727840","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.0-108
R. Bayindir, M. Yesilbudak, Medine Colak, N. Genç
Solar energy is one of the most affordable and clean renewable energy source in the world. Hence, the solar energy prediction is an inevitable requirement in order to get the maximum solar energy during the day time and to increase the efficiency of solar energy systems. For this purpose, this paper predicts the daily total energy generation of an installed photovoltaic system using the Naïve Bayes classifier. In the prediction process, one-year historical dataset including daily average temperature, daily total sunshine duration, daily total global solar radiation and daily total photovoltaic energy generation parameters are used as the categorical-valued attributes. By means of the Naïve Bayes application, the sensitivity and the accuracy measures are improved for the photovoltaic energy prediction and the effects of other solar attributes on the photovoltaic energy generation are evaluated.
{"title":"A Novel Application of Naive Bayes Classifier in Photovoltaic Energy Prediction","authors":"R. Bayindir, M. Yesilbudak, Medine Colak, N. Genç","doi":"10.1109/ICMLA.2017.0-108","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.0-108","url":null,"abstract":"Solar energy is one of the most affordable and clean renewable energy source in the world. Hence, the solar energy prediction is an inevitable requirement in order to get the maximum solar energy during the day time and to increase the efficiency of solar energy systems. For this purpose, this paper predicts the daily total energy generation of an installed photovoltaic system using the Naïve Bayes classifier. In the prediction process, one-year historical dataset including daily average temperature, daily total sunshine duration, daily total global solar radiation and daily total photovoltaic energy generation parameters are used as the categorical-valued attributes. By means of the Naïve Bayes application, the sensitivity and the accuracy measures are improved for the photovoltaic energy prediction and the effects of other solar attributes on the photovoltaic energy generation are evaluated.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"20 1","pages":"523-527"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75157623","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.00-69
K. Fujii, F. Hsieh, Cho-Jui Hsieh
We algorithmically compute and demonstrate multi-scale expert knowledge of computer gaming through pattern compositions on two levels of heterogeneity. Hierarchical clustering (HC) is applied to construct block-based heatmaps: colored matrices framed by two hierarchical trees imposed upon row and column axes. The computed heterogeneity is seen to induce different collections of viable gaming features pertaining to different map-clusters. On the game level, the map-dependent heterogeneity is seen to reveal which gaming-feature-pattern compositions are indeed viable for wins or losses with near-certainty, and which correspond to 50-50 uncertainty in outcome. Hence, such pattern compositions become the critical knowledge bases for pre-game prediction as well as ongoing-gaming strategy. The computer game, TagPro: Capture the Flag, is used as an illustrating example throughout the development of this paper.
我们通过算法计算和演示计算机游戏的多尺度专家知识，通过在两个异质性水平上的模式组合。分层聚类(HC)用于构建基于块的热图:由行轴和列轴上施加的两个分层树构成的彩色矩阵。计算出的异质性可以诱导出与不同地图集群相关的可行游戏功能的不同集合。在游戏层面上，地图依赖的异质性揭示了哪些游戏特征模式构成确实具有近乎确定性的获胜或失败，哪些对应于50% - 50%的结果不确定性。因此，这种模式组合成为游戏前预测和游戏中策略的关键知识基础。电脑游戏TagPro: Capture The Flag，在整个论文的开发过程中被用作一个说明例子。
{"title":"Computable Expert Knowledge in Computer Games","authors":"K. Fujii, F. Hsieh, Cho-Jui Hsieh","doi":"10.1109/ICMLA.2017.00-69","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.00-69","url":null,"abstract":"We algorithmically compute and demonstrate multi-scale expert knowledge of computer gaming through pattern compositions on two levels of heterogeneity. Hierarchical clustering (HC) is applied to construct block-based heatmaps: colored matrices framed by two hierarchical trees imposed upon row and column axes. The computed heterogeneity is seen to induce different collections of viable gaming features pertaining to different map-clusters. On the game level, the map-dependent heterogeneity is seen to reveal which gaming-feature-pattern compositions are indeed viable for wins or losses with near-certainty, and which correspond to 50-50 uncertainty in outcome. Hence, such pattern compositions become the critical knowledge bases for pre-game prediction as well as ongoing-gaming strategy. The computer game, TagPro: Capture the Flag, is used as an illustrating example throughout the development of this paper.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"99 1","pages":"749-754"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76780777","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.00-71
Anirban Das, Min-Yi Shen, M. Shashanka, Jisheng Wang
This paper proposes a method to detect two primary means of using the Domain Name System (DNS) for malicious purposes. We develop machine learning models to detect information exfiltration from compromised machines and the establishment of command & control (C&C) servers via tunneling. We validate our approach by experiments where we successfully detect a malware used in several recent Advanced Persistent Threat (APT) attacks [1]. The novelty of our method is its robustness, simplicity, scalability, and ease of deployment in a production environment.
{"title":"Detection of Exfiltration and Tunneling over DNS","authors":"Anirban Das, Min-Yi Shen, M. Shashanka, Jisheng Wang","doi":"10.1109/ICMLA.2017.00-71","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.00-71","url":null,"abstract":"This paper proposes a method to detect two primary means of using the Domain Name System (DNS) for malicious purposes. We develop machine learning models to detect information exfiltration from compromised machines and the establishment of command & control (C&C) servers via tunneling. We validate our approach by experiments where we successfully detect a malware used in several recent Advanced Persistent Threat (APT) attacks [1]. The novelty of our method is its robustness, simplicity, scalability, and ease of deployment in a production environment.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"33 1","pages":"737-742"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75809124","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.0-109
Y. Yilmaz, S. Uludag
The impact of cybersecurity attacks on the Smart Grid may cause cyber as well as physical damages, as clearly shown in the recent attacks on the power grid in Ukraine where consumers were left without power. A set of recent successful Distributed Denial-of-Service (DDoS) attacks on the Internet, facilitated by the proliferation of the Internet-of-Things powered botnets, shows that it is just a matter of time before the Smart Grid, as one of the most attractive critical infrastructure systems, becomes the target and likely victim of similar attacks, potentially leaving catastrophic disruption of power service to millions of people. It is in this context that we propose a scalable mitigation approach, referred to as Minimally Invasive Attack Mitigation via Detection Isolation and Localization (MIAMI-DIL), under a hierarchical data collection infrastructure. We provide a proofof- concept by means of simulations which show the efficacy and scalability of the proposed approach.
{"title":"Mitigating IoT-based Cyberattacks on the Smart Grid","authors":"Y. Yilmaz, S. Uludag","doi":"10.1109/ICMLA.2017.0-109","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.0-109","url":null,"abstract":"The impact of cybersecurity attacks on the Smart Grid may cause cyber as well as physical damages, as clearly shown in the recent attacks on the power grid in Ukraine where consumers were left without power. A set of recent successful Distributed Denial-of-Service (DDoS) attacks on the Internet, facilitated by the proliferation of the Internet-of-Things powered botnets, shows that it is just a matter of time before the Smart Grid, as one of the most attractive critical infrastructure systems, becomes the target and likely victim of similar attacks, potentially leaving catastrophic disruption of power service to millions of people. It is in this context that we propose a scalable mitigation approach, referred to as Minimally Invasive Attack Mitigation via Detection Isolation and Localization (MIAMI-DIL), under a hierarchical data collection infrastructure. We provide a proofof- concept by means of simulations which show the efficacy and scalability of the proposed approach.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"61 1","pages":"517-522"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76811393","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.00-81
W. Fisher, B. Jackson, T. Camp, V. Krzhizhanovskaya
As earth dams and levees (EDLs) across the United States reach the end of their design lives, effectively monitoring their structural integrity is of critical importance. This paper investigates automatic detection of anomalous events in passive seismic data as a step towards continuous real-time monitoring of EDL health. We use a multivariate Gaussian machine-learning model to identify anomalies in experimental data from two different laboratory earth embankments. Additionally, we explore five wavelet transform methods for signal denoising; removing different signal components. The best performance is achieved with the Haar wavelets (removing the Level 3 component). We achieve up to 97.3% overall accuracy and less than 1.4% false negatives in anomaly detection. These promising approaches could eventually provide a means for identifying internal erosion events in aging EDLs earlier than is currently possible, thereby allowing more time to prevent or mitigate catastrophic failures.
{"title":"Anomaly Detection in Earth Dam and Levee Passive Seismic Data Using Multivariate Gaussian","authors":"W. Fisher, B. Jackson, T. Camp, V. Krzhizhanovskaya","doi":"10.1109/ICMLA.2017.00-81","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.00-81","url":null,"abstract":"As earth dams and levees (EDLs) across the United States reach the end of their design lives, effectively monitoring their structural integrity is of critical importance. This paper investigates automatic detection of anomalous events in passive seismic data as a step towards continuous real-time monitoring of EDL health. We use a multivariate Gaussian machine-learning model to identify anomalies in experimental data from two different laboratory earth embankments. Additionally, we explore five wavelet transform methods for signal denoising; removing different signal components. The best performance is achieved with the Haar wavelets (removing the Level 3 component). We achieve up to 97.3% overall accuracy and less than 1.4% false negatives in anomaly detection. These promising approaches could eventually provide a means for identifying internal erosion events in aging EDLs earlier than is currently possible, thereby allowing more time to prevent or mitigate catastrophic failures.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"41 1","pages":"685-690"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75722328","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.00-63
T. Nguyen, S. Mukhopadhyay
Multidisciplinary Optimization (MDO) is one of the most popular techniques in aerospace engineering, where the system is complex and includes the knowledge from multiple fields. However, according to the best of our knowledge, MDO has not been widely applied in decentralized reinforcement learning (RL) due to the ‘unknown’ nature of the RL problems. In this work, we apply the MDO in decentralized RL. In our MDO design, each learning agent uses system identification to closely approximate the environment and tackle the ‘unknown’ nature of the RL. Then, the agents apply the MDO principles to compute the control solution using Monte Carlo and Markov Decision Process techniques. We examined two options of MDO designs: the multidisciplinary feasible and the individual discipline feasible options, which are suitable for multi-agent learning. Our results show that the MDO individual discipline feasible option could successfully learn how to control the system. The MDO approach shows better performance than the completely decentralization and centralization approaches.
{"title":"Multidisciplinary Optimization in Decentralized Reinforcement Learning","authors":"T. Nguyen, S. Mukhopadhyay","doi":"10.1109/ICMLA.2017.00-63","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.00-63","url":null,"abstract":"Multidisciplinary Optimization (MDO) is one of the most popular techniques in aerospace engineering, where the system is complex and includes the knowledge from multiple fields. However, according to the best of our knowledge, MDO has not been widely applied in decentralized reinforcement learning (RL) due to the ‘unknown’ nature of the RL problems. In this work, we apply the MDO in decentralized RL. In our MDO design, each learning agent uses system identification to closely approximate the environment and tackle the ‘unknown’ nature of the RL. Then, the agents apply the MDO principles to compute the control solution using Monte Carlo and Markov Decision Process techniques. We examined two options of MDO designs: the multidisciplinary feasible and the individual discipline feasible options, which are suitable for multi-agent learning. Our results show that the MDO individual discipline feasible option could successfully learn how to control the system. The MDO approach shows better performance than the completely decentralization and centralization approaches.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"41 1","pages":"779-784"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79898904","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.0-158
Fady Medhat, D. Chesmore, John A. Robinson
Automatic feature extraction using neural networks has accomplished remarkable success for images, but for sound recognition, these models are usually modified to fit the nature of the multi-dimensional temporal representation of the audio signal in spectrograms. This may not efficiently harness the time-frequency representation of the signal. The ConditionaL Neural Network (CLNN) takes into consideration the interrelation between the temporal frames, and the Masked ConditionaL Neural Network (MCLNN) extends upon the CLNN by forcing a systematic sparseness over the network’s weights using a binary mask. The masking allows the network to learn about frequency bands rather than bins, mimicking a filterbank used in signal transformations such as MFCC. Additionally, the Mask is designed to consider various combinations of features, which automates the feature hand-crafting process. We applied the MCLNN for the Environmental Sound Recognition problem using the Urbansound8k, YorNoise, ESC-10 and ESC-50 datasets. The MCLNN have achieved competitive performance compared to state-of-the-art Convolutional Neural Networks and hand-crafted attempts.
{"title":"Recognition of Acoustic Events Using Masked Conditional Neural Networks","authors":"Fady Medhat, D. Chesmore, John A. Robinson","doi":"10.1109/ICMLA.2017.0-158","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.0-158","url":null,"abstract":"Automatic feature extraction using neural networks has accomplished remarkable success for images, but for sound recognition, these models are usually modified to fit the nature of the multi-dimensional temporal representation of the audio signal in spectrograms. This may not efficiently harness the time-frequency representation of the signal. The ConditionaL Neural Network (CLNN) takes into consideration the interrelation between the temporal frames, and the Masked ConditionaL Neural Network (MCLNN) extends upon the CLNN by forcing a systematic sparseness over the network’s weights using a binary mask. The masking allows the network to learn about frequency bands rather than bins, mimicking a filterbank used in signal transformations such as MFCC. Additionally, the Mask is designed to consider various combinations of features, which automates the feature hand-crafting process. We applied the MCLNN for the Environmental Sound Recognition problem using the Urbansound8k, YorNoise, ESC-10 and ESC-50 datasets. The MCLNN have achieved competitive performance compared to state-of-the-art Convolutional Neural Networks and hand-crafted attempts.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"43 1","pages":"199-206"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80555502","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.0-107
R. T. Pereira, Javier Caicedo Zambrano
The results of the research project that aims to identify patterns of student dropout from socioeconomic, academic, disciplinary and institutional data of students from undergraduate programs at the University of Nariño from Pasto city (Colombia), using data mining techniques are presented. Built a data repository with the records of students who were admitted in the period from the first half of 2004 and the second semester of 2006. Three complete cohorts were analyzed with an observation period of six years until 2011. Socioeconomic and academic student dropout profiles were discovered using classification technique based on decision trees. The knowledge generated will support effective decision-making of university staff focused to develop policies and strategies related to student retention programs that are currently set.
{"title":"Application of Decision Trees for Detection of Student Dropout Profiles","authors":"R. T. Pereira, Javier Caicedo Zambrano","doi":"10.1109/ICMLA.2017.0-107","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.0-107","url":null,"abstract":"The results of the research project that aims to identify patterns of student dropout from socioeconomic, academic, disciplinary and institutional data of students from undergraduate programs at the University of Nariño from Pasto city (Colombia), using data mining techniques are presented. Built a data repository with the records of students who were admitted in the period from the first half of 2004 and the second semester of 2006. Three complete cohorts were analyzed with an observation period of six years until 2011. Socioeconomic and academic student dropout profiles were discovered using classification technique based on decision trees. The knowledge generated will support effective decision-making of university staff focused to develop policies and strategies related to student retention programs that are currently set.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"20 1","pages":"528-531"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87448704","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 : 2017-12-01DOI: 10.1109/ICMLA.2017.0-114
R. Benkercha, S. Moulahoum, I. Colak
The Grid Connected Photovoltaic System (GCPV) has become more used system in renewable energy. Several researches have been carried out to improve the efficiency and the decrease of energy losses. One of the important components used to increase the efficiency is the DC/DC boost converter. In this paper, a new hybrid model is proposed to control the DC/DC converter, this new controller is built on the fuzzy logic controller (FLC) and artificial neural network (ANN). The pathway taken to build the model is divided into three steps, the first step is to generate a data based on the FLC, the next step is to choose an ANN structure for modeling the FLC and the last step is the test and the validation of the obtained model. The phase of building an ANN is achieved by supervised learning based on back-propagation algorithm. This algorithm is used to train the ANN model by searching of the optimal weights and thresholds that has been a minimal root mean square error between the FLC output and the ANN model. The validation test was performed with various irradiation values between the both intelligent controllers and classical P&O algorithm simultaneously.
{"title":"Modelling of Fuzzy Logic Controller of a Maximum Power Point Tracker Based on Artificial Neural Network","authors":"R. Benkercha, S. Moulahoum, I. Colak","doi":"10.1109/ICMLA.2017.0-114","DOIUrl":"https://doi.org/10.1109/ICMLA.2017.0-114","url":null,"abstract":"The Grid Connected Photovoltaic System (GCPV) has become more used system in renewable energy. Several researches have been carried out to improve the efficiency and the decrease of energy losses. One of the important components used to increase the efficiency is the DC/DC boost converter. In this paper, a new hybrid model is proposed to control the DC/DC converter, this new controller is built on the fuzzy logic controller (FLC) and artificial neural network (ANN). The pathway taken to build the model is divided into three steps, the first step is to generate a data based on the FLC, the next step is to choose an ANN structure for modeling the FLC and the last step is the test and the validation of the obtained model. The phase of building an ANN is achieved by supervised learning based on back-propagation algorithm. This algorithm is used to train the ANN model by searching of the optimal weights and thresholds that has been a minimal root mean square error between the FLC output and the ANN model. The validation test was performed with various irradiation values between the both intelligent controllers and classical P&O algorithm simultaneously.","PeriodicalId":6636,"journal":{"name":"2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA)","volume":"68 1","pages":"485-492"},"PeriodicalIF":0.0,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78644867","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：