Pub Date : 2016-11-01DOI: 10.1109/INDEL.2016.7797784
Marko Č. Bošković, T. Šekara, M. Rapaić, Vidan Govedarica
This paper presents a modification of recently reported general rules for designing complex controllers. The presented approach is here applied to control stable processes including dead-time with a dominant left half-plane zero. The introduced modification enables improvements in the form of better disturbance rejection and smaller IAE (Integrated Absolute Error). The solution to design procedure leads to practically realizable controllers. The effectiveness of the proposed method is verified via numerical simulations on several examples resulting into better performance and robustness indices of the closed-loop system compared to recently reported procedure which does not take into consideration the effect of dominant left half-plane zero.
{"title":"Novel tuning rules for stable dead-time processes with dominant left half-plane zero","authors":"Marko Č. Bošković, T. Šekara, M. Rapaić, Vidan Govedarica","doi":"10.1109/INDEL.2016.7797784","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797784","url":null,"abstract":"This paper presents a modification of recently reported general rules for designing complex controllers. The presented approach is here applied to control stable processes including dead-time with a dominant left half-plane zero. The introduced modification enables improvements in the form of better disturbance rejection and smaller IAE (Integrated Absolute Error). The solution to design procedure leads to practically realizable controllers. The effectiveness of the proposed method is verified via numerical simulations on several examples resulting into better performance and robustness indices of the closed-loop system compared to recently reported procedure which does not take into consideration the effect of dominant left half-plane zero.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125154550","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797797
S. Gusev, V. Oboskalov
Large computational load is one of the key issues of an electrical power system security analysis. Currently, most of the researchers are trying to find out how to reduce the computational load. The contingency analysis is one of the most promising approaches for solving the problem. In fact, there are several ways to improve the effectiveness of the contingency analysis. The first way is to reduce the number of simulated situations via contingency searching or contingency screening. Another way is to increase the efficiency of power flow computations. This paper presents a novel recursion-based approach for solving the contingency analysis problem. The approach improves the effectiveness of the contingency analysis by the above-mentioned ways. It relies on a specific contingency sorting procedure based on the recursive call of the calculating procedure. The proposed approach has been tested on the 14-bus IEEE test system. The results obtained show almost a twofold increase in the computational efficiency in comparison with the conventional contingency sorting procedure.
{"title":"Recursion based contingency analysis of an electrical power system","authors":"S. Gusev, V. Oboskalov","doi":"10.1109/INDEL.2016.7797797","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797797","url":null,"abstract":"Large computational load is one of the key issues of an electrical power system security analysis. Currently, most of the researchers are trying to find out how to reduce the computational load. The contingency analysis is one of the most promising approaches for solving the problem. In fact, there are several ways to improve the effectiveness of the contingency analysis. The first way is to reduce the number of simulated situations via contingency searching or contingency screening. Another way is to increase the efficiency of power flow computations. This paper presents a novel recursion-based approach for solving the contingency analysis problem. The approach improves the effectiveness of the contingency analysis by the above-mentioned ways. It relies on a specific contingency sorting procedure based on the recursive call of the calculating procedure. The proposed approach has been tested on the 14-bus IEEE test system. The results obtained show almost a twofold increase in the computational efficiency in comparison with the conventional contingency sorting procedure.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122478917","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797775
Nico Stati, Mohamed Abdelrahem, Muhammad Hosnee Mobarak, R. Kennel
In this paper, a finite control set-model predictive control (FCS-MPC) scheme with on-line parameter estimation for permanent magnet synchronous generators (PMSGs) based variable-speed wind turbines is proposed. For designing the proposed FCS-MPC, the discrete time model of the PMSG is considered and the future behavior is predicted. Afterward, the switching action that produces the minimum value of a predefined cost function is applied in the next sampling instant. In order to increase the robustness of the proposed FCS-MPC against parameter variations of the PMSG, an extended Kalman filter (EKF) is implemented for estimating the stator resistance and inductance of the PMSG. The performance of the proposed FCS-MPC with on-line parameter estimation is proved via simulation results for all operation conditions.
{"title":"Finite control set-model predictive control with on-line parameter estimation for variable-speed wind energy conversion systems","authors":"Nico Stati, Mohamed Abdelrahem, Muhammad Hosnee Mobarak, R. Kennel","doi":"10.1109/INDEL.2016.7797775","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797775","url":null,"abstract":"In this paper, a finite control set-model predictive control (FCS-MPC) scheme with on-line parameter estimation for permanent magnet synchronous generators (PMSGs) based variable-speed wind turbines is proposed. For designing the proposed FCS-MPC, the discrete time model of the PMSG is considered and the future behavior is predicted. Afterward, the switching action that produces the minimum value of a predefined cost function is applied in the next sampling instant. In order to increase the robustness of the proposed FCS-MPC against parameter variations of the PMSG, an extended Kalman filter (EKF) is implemented for estimating the stator resistance and inductance of the PMSG. The performance of the proposed FCS-MPC with on-line parameter estimation is proved via simulation results for all operation conditions.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123577138","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797815
K. Kasas-Lazetic, D. Herceg, D. Kljajić, G. Mijatović, M. Prsa
Electric characteristics of Aluminum Conductor Steel Reinforced (ACSR) components, aluminum and steel conductivity values, are usually provided by the ACSR manufacturers, but not the steel core permeability. In order to explore how important the steel core permeability is, the paper deals with the impact of this characteristic on ACSR impedance per unit length. For the estimation of this impact, numerous calculations were performed and the results are shown in this paper. The value of relative steel core permeability was varied from 10 to 50000 inside three different standardized ACSR conductors, at eleven different frequencies, up to 2500 Hz (50th harmonic of fundamental frequency). All calculations were carried out numerically, applying COMSOL Multiphysics 3.5a computer program package on simplified two-dimensional model. ACSR resistance per unit length was calculated from heating losses, while the reactance and the inductance per unit length were determined via magnetic energy stored in ACSR magnetic field. Even in the case of great relative permeability value range, the paper shows that the influence of steel core permeability on any ACS conductor is much less than expected, especially at lower frequencies. This fact is the consequence of the big difference in aluminum and steel conductivity values.
{"title":"Impact of steel permeability on ACSR impedance per unit length","authors":"K. Kasas-Lazetic, D. Herceg, D. Kljajić, G. Mijatović, M. Prsa","doi":"10.1109/INDEL.2016.7797815","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797815","url":null,"abstract":"Electric characteristics of Aluminum Conductor Steel Reinforced (ACSR) components, aluminum and steel conductivity values, are usually provided by the ACSR manufacturers, but not the steel core permeability. In order to explore how important the steel core permeability is, the paper deals with the impact of this characteristic on ACSR impedance per unit length. For the estimation of this impact, numerous calculations were performed and the results are shown in this paper. The value of relative steel core permeability was varied from 10 to 50000 inside three different standardized ACSR conductors, at eleven different frequencies, up to 2500 Hz (50th harmonic of fundamental frequency). All calculations were carried out numerically, applying COMSOL Multiphysics 3.5a computer program package on simplified two-dimensional model. ACSR resistance per unit length was calculated from heating losses, while the reactance and the inductance per unit length were determined via magnetic energy stored in ACSR magnetic field. Even in the case of great relative permeability value range, the paper shows that the influence of steel core permeability on any ACS conductor is much less than expected, especially at lower frequencies. This fact is the consequence of the big difference in aluminum and steel conductivity values.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121139339","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797776
Muhammad Hosnee Mobarak, Mohamed Abdelrahem, Nico Stati, R. Kennel
In this paper, a finite control set-model predictive control (FCS-MPC) approach for low-voltage ride-through (LVRT) capability enhancement of permanent magnet synchronous generators (PMSGs) based on variable-speed wind turbines is proposed. According to the FCS-MPC principles, the discrete states of the converter are taken into account and the future converter performance is predicted. Subsequently, the voltage vector that optimizes a predefined cost function is selected to be applied in the next sampling interval. The proposed LVRT method exploits the wind turbine-PMSG rotor inertia to store the surplus power during the grid voltage dips. Simulation results are presented to validate the proposed LVRT strategy for all operation conditions.
{"title":"Model predictive control for low-voltage ride-through capability improvement of variable-speed wind energy conversion systems","authors":"Muhammad Hosnee Mobarak, Mohamed Abdelrahem, Nico Stati, R. Kennel","doi":"10.1109/INDEL.2016.7797776","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797776","url":null,"abstract":"In this paper, a finite control set-model predictive control (FCS-MPC) approach for low-voltage ride-through (LVRT) capability enhancement of permanent magnet synchronous generators (PMSGs) based on variable-speed wind turbines is proposed. According to the FCS-MPC principles, the discrete states of the converter are taken into account and the future converter performance is predicted. Subsequently, the voltage vector that optimizes a predefined cost function is selected to be applied in the next sampling interval. The proposed LVRT method exploits the wind turbine-PMSG rotor inertia to store the surplus power during the grid voltage dips. Simulation results are presented to validate the proposed LVRT strategy for all operation conditions.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130641207","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797772
Dejan R. Ivic, Drazenko Macanovic, D. Šošić, P. Stefanov
From the viewpoint of the emerging concept of smart grid, distribution network will require fast power flow solution that must be resolved as efficiently as possible. Presence of distributed generators and power electronic based compensating devices, in modern medium and low voltage distribution networks, additionaly expands requirements for performances of algorithms for power flow calculations which are an essential part of effective smart distribution system analysis tools. The solution which can respond to most of defined requirements is proposed in this paper. This paper presents improved algorithm for power flow calculations for radial and weakly meshed distribution network with distributed generation and compensating devices. Proposed algorithm is based on the classical implicit impedance matrix (Zbus) Gauss method. Algorithm, presented in this paper, is enforceable to any one-sided supplied distribution network topology and can calculate the impact of distributed generation from renewable energy sources on electrical power losses and voltage magnitudes. Algorithm is applied to IEEE 33 bus test distribution network, modified by adding tie lines to get meshed network. Efficiency of proposed algorithm is confirmed by calculations for different switching scenarios within analysis of system abilities made to meet requirements for integration of distributed generators in defined system nodes. Described analysis were carried out for both systems, with and without compensation devices included.
{"title":"Weakly meshed distribution networks with distributed generation — power flow analysis using improved impedance matrix based algorithm","authors":"Dejan R. Ivic, Drazenko Macanovic, D. Šošić, P. Stefanov","doi":"10.1109/INDEL.2016.7797772","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797772","url":null,"abstract":"From the viewpoint of the emerging concept of smart grid, distribution network will require fast power flow solution that must be resolved as efficiently as possible. Presence of distributed generators and power electronic based compensating devices, in modern medium and low voltage distribution networks, additionaly expands requirements for performances of algorithms for power flow calculations which are an essential part of effective smart distribution system analysis tools. The solution which can respond to most of defined requirements is proposed in this paper. This paper presents improved algorithm for power flow calculations for radial and weakly meshed distribution network with distributed generation and compensating devices. Proposed algorithm is based on the classical implicit impedance matrix (Zbus) Gauss method. Algorithm, presented in this paper, is enforceable to any one-sided supplied distribution network topology and can calculate the impact of distributed generation from renewable energy sources on electrical power losses and voltage magnitudes. Algorithm is applied to IEEE 33 bus test distribution network, modified by adding tie lines to get meshed network. Efficiency of proposed algorithm is confirmed by calculations for different switching scenarios within analysis of system abilities made to meet requirements for integration of distributed generators in defined system nodes. Described analysis were carried out for both systems, with and without compensation devices included.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130872079","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797789
E. A. Plesnyaev, A. Berdin, P. Kryuchkov, P. A. Kuzin, P. Kovalenko
The problem of the power consumption forecast calculation is solved for all time intervals with the subsequent increase in the accuracy while the prediction intervals are decreasing. The lack of complete information on all types of node loads of the controlled power system makes it necessary to solve the problem of their determination. The composition of the nodes and methods of loads modelling are determined by the equivalent circuit of the power grid applied. A System of Node Load Models that provides a solution of the whole complex of forecasting problems on the basis of a common methodological basis, using nonlinear additive models is proposed in the paper.
{"title":"The system of node load models in the problem of electrical power production planning","authors":"E. A. Plesnyaev, A. Berdin, P. Kryuchkov, P. A. Kuzin, P. Kovalenko","doi":"10.1109/INDEL.2016.7797789","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797789","url":null,"abstract":"The problem of the power consumption forecast calculation is solved for all time intervals with the subsequent increase in the accuracy while the prediction intervals are decreasing. The lack of complete information on all types of node loads of the controlled power system makes it necessary to solve the problem of their determination. The composition of the nodes and methods of loads modelling are determined by the equivalent circuit of the power grid applied. A System of Node Load Models that provides a solution of the whole complex of forecasting problems on the basis of a common methodological basis, using nonlinear additive models is proposed in the paper.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134355998","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797774
Danijela Vuković, S. Vučenović, J. Šetrajčić, D. Rodić
Based on microscopic (exciton) theory of optical properties of symmetrically perturbed ultrathin molecular films, the absorption, reflection and transparency indices were formulated and presented in the function of frequencies of external electromagnetic field in near infrared (IR) region. It has been showed that all optical properties depend on the position of the crystal plane with regard on boundary planes of the film. We have determined and analyzed optical properties relations for the whole film structure based on the consideration for multiple reflection, absorption and transparency in those multilayered structure. The three-layered dielectric nanofilms with different boundary conditions on surfaces were analyzed and some discrete resonant absorption lines were obtained. Their number, position and distribution depend on the boundary parameter values, i.e. on the type and the technological process of their preparation/fabrication. Practically monochromatic absorption may occur. Unlike the corresponding bulk-samples which are total absorbers throughout the near IR region, in ultrathin films will appear selective and discrete reflection and transparency too. These results could give a great contribution in optical engineering of nanostructures, especially in technology of designing of new electronic and photonic equipment, and for nanoparticles construction for drug carrier/delivery in nano-medicine.
{"title":"Discretization of optical properties in symmetrical nanofilm-structures","authors":"Danijela Vuković, S. Vučenović, J. Šetrajčić, D. Rodić","doi":"10.1109/INDEL.2016.7797774","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797774","url":null,"abstract":"Based on microscopic (exciton) theory of optical properties of symmetrically perturbed ultrathin molecular films, the absorption, reflection and transparency indices were formulated and presented in the function of frequencies of external electromagnetic field in near infrared (IR) region. It has been showed that all optical properties depend on the position of the crystal plane with regard on boundary planes of the film. We have determined and analyzed optical properties relations for the whole film structure based on the consideration for multiple reflection, absorption and transparency in those multilayered structure. The three-layered dielectric nanofilms with different boundary conditions on surfaces were analyzed and some discrete resonant absorption lines were obtained. Their number, position and distribution depend on the boundary parameter values, i.e. on the type and the technological process of their preparation/fabrication. Practically monochromatic absorption may occur. Unlike the corresponding bulk-samples which are total absorbers throughout the near IR region, in ultrathin films will appear selective and discrete reflection and transparency too. These results could give a great contribution in optical engineering of nanostructures, especially in technology of designing of new electronic and photonic equipment, and for nanoparticles construction for drug carrier/delivery in nano-medicine.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116283946","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797787
P. Kovalenko, A. Berdin, D. Bliznyuk, E. A. Plesnyaev
The problem of identifying a disturbance and transient duration in power systems at the appropriate time is crucial. The necessity of an early disturbance detection is driven by the essential aims of dispatching control and automatic emergency control systems, both local and centralized. The paper proposes a flexible algorithm for identifying a disturbance and transient duration, based on the "hinge joint" technique: the initial data interval is fitted iteratively by two linear sections connected one after another. The key element of the algorithm is tracking the difference between the sections' angles of descent. Applying the technique allows to detect the point where the signal changes rapidly and significantly, which indicates the transient start. The significant changes in the state parameters behavior may be driven by a power imbalance when power signals are considered or by other emergencies along with relay protection and automatic control systems actions. The primary aim of the algorithm is to identify the disturbance with the highest possible accuracy in terms of time based on the power system state parameters measurements. The algorithm is applied to processing both Wide-Area Monitoring System (WAMS) data of the sampling rate corresponding to the rated network frequency (obtained from Phasor Measurement Units - PMUs) and instantaneous state parameters measurements (obtained from Digital Fault Recorders - DFRs) as well. The algorithm is validated involving physical simulation results along with the actual WAMS data collected from power system entities. The algorithm processing speed is sufficient to ensure operation on a real time basis.
{"title":"The flexible algorithm for identifying a disturbance and transient duration in power systems","authors":"P. Kovalenko, A. Berdin, D. Bliznyuk, E. A. Plesnyaev","doi":"10.1109/INDEL.2016.7797787","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797787","url":null,"abstract":"The problem of identifying a disturbance and transient duration in power systems at the appropriate time is crucial. The necessity of an early disturbance detection is driven by the essential aims of dispatching control and automatic emergency control systems, both local and centralized. The paper proposes a flexible algorithm for identifying a disturbance and transient duration, based on the \"hinge joint\" technique: the initial data interval is fitted iteratively by two linear sections connected one after another. The key element of the algorithm is tracking the difference between the sections' angles of descent. Applying the technique allows to detect the point where the signal changes rapidly and significantly, which indicates the transient start. The significant changes in the state parameters behavior may be driven by a power imbalance when power signals are considered or by other emergencies along with relay protection and automatic control systems actions. The primary aim of the algorithm is to identify the disturbance with the highest possible accuracy in terms of time based on the power system state parameters measurements. The algorithm is applied to processing both Wide-Area Monitoring System (WAMS) data of the sampling rate corresponding to the rated network frequency (obtained from Phasor Measurement Units - PMUs) and instantaneous state parameters measurements (obtained from Digital Fault Recorders - DFRs) as well. The algorithm is validated involving physical simulation results along with the actual WAMS data collected from power system entities. The algorithm processing speed is sufficient to ensure operation on a real time basis.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134288713","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 : 2016-11-01DOI: 10.1109/INDEL.2016.7797794
Igor Sevo, Tijana Mijatovic
Figure detection, separation and image classification are common problems occurring in various fields, especially medicine. Since image databases are usually large, manual classification would be a demanding task. In this paper, we proposed a method for automatic compound figure detection and separation, and gave a comparison between other recognition methods, such as convolutional neural networks. The proposed method is based on differentiating objects in the image, and merging object artifacts with the nearest large object. Parameters of size and distance were varied, and different criteria for determining the object boundaries were tested. Using this method, an accuracy of 90.20% was achieved on a test set of 500 images, with an average processing time less than 600ms per image for the given combination of parameters.
{"title":"Segmentation-based compound figure detection and separation methods","authors":"Igor Sevo, Tijana Mijatovic","doi":"10.1109/INDEL.2016.7797794","DOIUrl":"https://doi.org/10.1109/INDEL.2016.7797794","url":null,"abstract":"Figure detection, separation and image classification are common problems occurring in various fields, especially medicine. Since image databases are usually large, manual classification would be a demanding task. In this paper, we proposed a method for automatic compound figure detection and separation, and gave a comparison between other recognition methods, such as convolutional neural networks. The proposed method is based on differentiating objects in the image, and merging object artifacts with the nearest large object. Parameters of size and distance were varied, and different criteria for determining the object boundaries were tested. Using this method, an accuracy of 90.20% was achieved on a test set of 500 images, with an average processing time less than 600ms per image for the given combination of parameters.","PeriodicalId":273613,"journal":{"name":"2016 International Symposium on Industrial Electronics (INDEL)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122086774","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：