Pub Date : 2018-05-01DOI: 10.1109/COLCACI.2018.8484851
Juan de Dios Fuentes, A. Orjuela-Cañón, Héctor Iván Tangarife Escobar
This paper contains a proposal to determine the kind of nonlinear load when different appliances are connected to the solar generation system. A database built with sampled signals from the photovoltaic systems of the National Learning Service (SENA) in Bogota was employed. The methodology used information from harmonic distortion extracted from nonlinear loads, which was used as input in an artificial neural network with supervised learning. Two proposals were implemented. First one was based on energy information and second one was worked with wave peaks information. Results show that a classification rate of 95% could be reached in a problem with eight classes.
{"title":"Nonlinear loads determination using harmonic information in photovoltaic generation systems","authors":"Juan de Dios Fuentes, A. Orjuela-Cañón, Héctor Iván Tangarife Escobar","doi":"10.1109/COLCACI.2018.8484851","DOIUrl":"https://doi.org/10.1109/COLCACI.2018.8484851","url":null,"abstract":"This paper contains a proposal to determine the kind of nonlinear load when different appliances are connected to the solar generation system. A database built with sampled signals from the photovoltaic systems of the National Learning Service (SENA) in Bogota was employed. The methodology used information from harmonic distortion extracted from nonlinear loads, which was used as input in an artificial neural network with supervised learning. Two proposals were implemented. First one was based on energy information and second one was worked with wave peaks information. Results show that a classification rate of 95% could be reached in a problem with eight classes.","PeriodicalId":138992,"journal":{"name":"2018 IEEE 1st Colombian Conference on Applications in Computational Intelligence (ColCACI)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133911576","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 : 2018-05-01DOI: 10.1109/COLCACI.2018.8484844
H. Morales-Lopez, Israel Cruz-Vega, J. Ramírez-Cortés, H. Peregrina-Barreto, J. Rangel-Magdaleno
Many computer algorithms have been developed, providing an initial aided diagnosis to the medical expertise. Most important previous stage in the automatic classificatión to grading diseases using images is to obtain a well-segmented región of interest from. Several related research in image classificatión uses a great number of image processing techniques previous to the classificatión stage. In this paper, we compare the automatic segmentatión based on two leading machine learning techniques: Differential Evolutión (DE) and the Self-Organizing Multilayer (SOM) Neural Network (NN) methods. The results are also compared with K-means algorithm for multi-level segmentatión from slit-lamp images. Segmented images were obtained relying on a thresholding approach based on fuzzy partitións of the image histogram and a fuzzy entropy measure optimized via a neural process and by the evolutive technique. The resulting approaches were also compared with the classical Shannon entropy.
{"title":"Multi-Level Image Segmentatión in Slit-Lamp Images: A Comparison Between two Machine Learning Techniques","authors":"H. Morales-Lopez, Israel Cruz-Vega, J. Ramírez-Cortés, H. Peregrina-Barreto, J. Rangel-Magdaleno","doi":"10.1109/COLCACI.2018.8484844","DOIUrl":"https://doi.org/10.1109/COLCACI.2018.8484844","url":null,"abstract":"Many computer algorithms have been developed, providing an initial aided diagnosis to the medical expertise. Most important previous stage in the automatic classificatión to grading diseases using images is to obtain a well-segmented región of interest from. Several related research in image classificatión uses a great number of image processing techniques previous to the classificatión stage. In this paper, we compare the automatic segmentatión based on two leading machine learning techniques: Differential Evolutión (DE) and the Self-Organizing Multilayer (SOM) Neural Network (NN) methods. The results are also compared with K-means algorithm for multi-level segmentatión from slit-lamp images. Segmented images were obtained relying on a thresholding approach based on fuzzy partitións of the image histogram and a fuzzy entropy measure optimized via a neural process and by the evolutive technique. The resulting approaches were also compared with the classical Shannon entropy.","PeriodicalId":138992,"journal":{"name":"2018 IEEE 1st Colombian Conference on Applications in Computational Intelligence (ColCACI)","volume":"67 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120990184","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 : 2018-05-01DOI: 10.1109/COLCACI.2018.8484859
E. Inga, M. Campaña, R. Hincapié
This work presents a model of optimal sizing of electrical distribution networks that uses real scenarios, georeferenced and contrasted by simulation processes that analyze the deployment and variables within the planning of electrical networks, considering a scalable demand of users and the voltage levels of the electrical distribution network. The work exposes within the dimensioning of the radial electrical network the possible conditions to avoid a load imbalance and in this way to prevent a system failure.
{"title":"Optimal Sizing of Electrical Distribution Networks considering Scalable Demand and Voltage","authors":"E. Inga, M. Campaña, R. Hincapié","doi":"10.1109/COLCACI.2018.8484859","DOIUrl":"https://doi.org/10.1109/COLCACI.2018.8484859","url":null,"abstract":"This work presents a model of optimal sizing of electrical distribution networks that uses real scenarios, georeferenced and contrasted by simulation processes that analyze the deployment and variables within the planning of electrical networks, considering a scalable demand of users and the voltage levels of the electrical distribution network. The work exposes within the dimensioning of the radial electrical network the possible conditions to avoid a load imbalance and in this way to prevent a system failure.","PeriodicalId":138992,"journal":{"name":"2018 IEEE 1st Colombian Conference on Applications in Computational Intelligence (ColCACI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134423298","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 : 2018-05-01DOI: 10.1109/COLCACI.2018.8484848
A. Jerez, Hans Garcia, H. Arguello
Compressive spectral imaging (CSI) allows the acquisition of the spectral information of a three dimensional scene by using two dimensional coded projections. However, compressed sampling of information with simultaneously high spatial and high spectral resolution demands expensive highresolution sensors. Single pixel imaging is an approach that has had a high impact in spectroscopy, due to its low-cost implementation compared to architectures with larger sensors. One of the main challenges in CSI is to obtain high quality image reconstructions using low-cost architectures. Recent works have been shown that image fusion using measurements from a CSI sensor based on side information leads to improvement in the quality of the fused image. This work proposes a methodology that combines the spectral information of a single pixel camera (SPC) and the side information of a grayscale sensor in order to improve the reconstruction quality of the spatio-spectral data cube. Simulations and experimental results for the proposed method are shown, and its performance is compared with respect to the traditional approach of upsampling the single pixel image reconstruction through bilinear interpolation.
{"title":"Single Pixel Spectral Image Fusion with Side Information from a Grayscale Sensor","authors":"A. Jerez, Hans Garcia, H. Arguello","doi":"10.1109/COLCACI.2018.8484848","DOIUrl":"https://doi.org/10.1109/COLCACI.2018.8484848","url":null,"abstract":"Compressive spectral imaging (CSI) allows the acquisition of the spectral information of a three dimensional scene by using two dimensional coded projections. However, compressed sampling of information with simultaneously high spatial and high spectral resolution demands expensive highresolution sensors. Single pixel imaging is an approach that has had a high impact in spectroscopy, due to its low-cost implementation compared to architectures with larger sensors. One of the main challenges in CSI is to obtain high quality image reconstructions using low-cost architectures. Recent works have been shown that image fusion using measurements from a CSI sensor based on side information leads to improvement in the quality of the fused image. This work proposes a methodology that combines the spectral information of a single pixel camera (SPC) and the side information of a grayscale sensor in order to improve the reconstruction quality of the spatio-spectral data cube. Simulations and experimental results for the proposed method are shown, and its performance is compared with respect to the traditional approach of upsampling the single pixel image reconstruction through bilinear interpolation.","PeriodicalId":138992,"journal":{"name":"2018 IEEE 1st Colombian Conference on Applications in Computational Intelligence (ColCACI)","volume":"176 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131839162","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 : 2018-05-01DOI: 10.1109/COLCACI.2018.8484860
Juan Carlos Figueroa–García, José Jairo Soriano-Mendez, Miguel Alberto Melgarejo-Rey
This paper presents a proposal for computing the variance of a fuzzy number based on the Yager index for convex fuzzy sets. We compare the proposal to the sample variance (based on α-cuts) to see its behavior over triangular, trapezoidal and Gaussian fuzzy numbers. Some considerations about the obtained results are provided and some recommendations are given.
{"title":"On the variance of a fuzzy number based on the Yager index","authors":"Juan Carlos Figueroa–García, José Jairo Soriano-Mendez, Miguel Alberto Melgarejo-Rey","doi":"10.1109/COLCACI.2018.8484860","DOIUrl":"https://doi.org/10.1109/COLCACI.2018.8484860","url":null,"abstract":"This paper presents a proposal for computing the variance of a fuzzy number based on the Yager index for convex fuzzy sets. We compare the proposal to the sample variance (based on α-cuts) to see its behavior over triangular, trapezoidal and Gaussian fuzzy numbers. Some considerations about the obtained results are provided and some recommendations are given.","PeriodicalId":138992,"journal":{"name":"2018 IEEE 1st Colombian Conference on Applications in Computational Intelligence (ColCACI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115633103","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 : 2018-04-28DOI: 10.1109/COLCACI.2018.8484854
J. Hernandez, Andres G. Abad
A restricted Boltzmann machine (RBM) is a generative neural-network model with many novel applications such as collaborative filtering and acoustic modeling. An RBM lacks the capacity to retain memory, making it inappropriate for dynamic data modeling as in time-series analysis. In this paper we address this issue by proposing the p-RBM model, a generalization of the regular RBM model, capable of retaining memory of p past states. We further show how to train the p-RBM model using contrastive divergence and test our model on the problem of predicting the stock market direction considering 100 stocks of the NASDAQ-100 index. Obtained results show that the p-RBM offer promising prediction potential.
{"title":"Learning from multivariate discrete sequential data using a restricted Boltzmann machine model","authors":"J. Hernandez, Andres G. Abad","doi":"10.1109/COLCACI.2018.8484854","DOIUrl":"https://doi.org/10.1109/COLCACI.2018.8484854","url":null,"abstract":"A restricted Boltzmann machine (RBM) is a generative neural-network model with many novel applications such as collaborative filtering and acoustic modeling. An RBM lacks the capacity to retain memory, making it inappropriate for dynamic data modeling as in time-series analysis. In this paper we address this issue by proposing the p-RBM model, a generalization of the regular RBM model, capable of retaining memory of p past states. We further show how to train the p-RBM model using contrastive divergence and test our model on the problem of predicting the stock market direction considering 100 stocks of the NASDAQ-100 index. Obtained results show that the p-RBM offer promising prediction potential.","PeriodicalId":138992,"journal":{"name":"2018 IEEE 1st Colombian Conference on Applications in Computational Intelligence (ColCACI)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127421714","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：