Pub Date : 2018-02-23DOI: 10.5152/iujeee.2018.1810
H. Ankishan
Voice data has demonstrated chaotic behavior in previous studies. Therefore, studying the linear properties alone does not yield successful results. This is valid for the examination of voice data as well. Therefore, conducting studies including chaotic features as well as existing technologies is inevitable. The main purpose of this study is to detect voice pathologies with fewer special features using new chaotic features. Both linear and nonlinear characteristics were used in this study. In this context, the largest Lyapunov exponents and entropy are preferred as chaotic properties because of their success in previous studies. Very few results with 100% accuracy were obtained in the experimental studies. In this study, multiple support vector machines (SVMs) were selected as a classifier because of their success in previous similar data types. Thus, the desired accuracy level was achieved using fewer features. Resultantly, the process complexity decreased and the system speed increased.
{"title":"A New Approach for Detection of Pathological Voice Disorders with Reduced Parameters","authors":"H. Ankishan","doi":"10.5152/iujeee.2018.1810","DOIUrl":"https://doi.org/10.5152/iujeee.2018.1810","url":null,"abstract":"Voice data has demonstrated chaotic behavior in previous studies. Therefore, studying the linear properties alone does not yield successful results. This is valid for the examination of voice data as well. Therefore, conducting studies including chaotic features as well as existing technologies is inevitable. The main purpose of this study is to detect voice pathologies with fewer special features using new chaotic features. Both linear and nonlinear characteristics were used in this study. In this context, the largest Lyapunov exponents and entropy are preferred as chaotic properties because of their success in previous studies. Very few results with 100% accuracy were obtained in the experimental studies. In this study, multiple support vector machines (SVMs) were selected as a classifier because of their success in previous similar data types. Thus, the desired accuracy level was achieved using fewer features. Resultantly, the process complexity decreased and the system speed increased.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129493560","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 : 2018-02-23DOI: 10.5152/iujeee.2018.1816
A. Boyacı, A. R. Ekti, Serhan Yarkan, M. Aydin
This paper discusses the current aspects of the surveillance and management of the electromagnetic spectrum under various topics that are vital to wireless communications. On the basis of the recent technological advancements in conjunction with the discussions on the global approaches toward electromagnetic spectrum monitoring, the paper addresses short-, medium-, and long-term strategies in wireless communications for monitoring, governing, and managing the electromagnetic spectrum of leading countries and multinational organizations. Furthermore, it proposes a novel spectrum monitoring strategy for next-generation wireless systems and outlines the implications of the proposed strategy.
{"title":"Monitoring, Surveillance, and Management of the Electromagnetic Spectrum: Current Issues in Electromagnetic Spectrum Monitoring","authors":"A. Boyacı, A. R. Ekti, Serhan Yarkan, M. Aydin","doi":"10.5152/iujeee.2018.1816","DOIUrl":"https://doi.org/10.5152/iujeee.2018.1816","url":null,"abstract":"This paper discusses the current aspects of the surveillance and management of the electromagnetic spectrum under various topics that are vital to wireless communications. On the basis of the recent technological advancements in conjunction with the discussions on the global approaches toward electromagnetic spectrum monitoring, the paper addresses short-, medium-, and long-term strategies in wireless communications for monitoring, governing, and managing the electromagnetic spectrum of leading countries and multinational organizations. Furthermore, it proposes a novel spectrum monitoring strategy for next-generation wireless systems and outlines the implications of the proposed strategy.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133194750","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 : 2018-02-23DOI: 10.5152/IUJEEE.2018.1808
V. Sravani, K. Santhosh, S. Bhargava, Verina D'Almeida
Agricultural produce significantly depends on many crop parameters such as humidity, pH, temperature, sunlight, microbial activity, soil ions, air quality, and water quality. A higher production of crop can be achieved via maintaining all these parameters in the desired range. A smart system was developed to control the environmental parameters in the desired range via incorporating a multisensor to measure the parameters such as humidity, temperature, and sunlight; in addition, also a suitable controller was designed to control these parameters in the desired range. Sensors were placed to continuously monitor the field parameters such as temperature, humidity, sunlight, and soil moisture. All these parameters were remotely acquired using ZigBee to PC through myRIO boards. Fuzzy-based controllers were designed to operate the actuators to maintain the set point. The designed system on implementation was tested on a real-life model. The results show that the proposed technique maintained the parameters at the desired state and reduced human intervention and labor.
{"title":"Design and Implementation of a Smart Controller in Agriculture for Improved Productivity","authors":"V. Sravani, K. Santhosh, S. Bhargava, Verina D'Almeida","doi":"10.5152/IUJEEE.2018.1808","DOIUrl":"https://doi.org/10.5152/IUJEEE.2018.1808","url":null,"abstract":"Agricultural produce significantly depends on many crop parameters such as humidity, pH, temperature, sunlight, microbial activity, soil ions, air quality, and water quality. A higher production of crop can be achieved via maintaining all these parameters in the desired range. A smart system was developed to control the environmental parameters in the desired range via incorporating a multisensor to measure the parameters such as humidity, temperature, and sunlight; in addition, also a suitable controller was designed to control these parameters in the desired range. Sensors were placed to continuously monitor the field parameters such as temperature, humidity, sunlight, and soil moisture. All these parameters were remotely acquired using ZigBee to PC through myRIO boards. Fuzzy-based controllers were designed to operate the actuators to maintain the set point. The designed system on implementation was tested on a real-life model. The results show that the proposed technique maintained the parameters at the desired state and reduced human intervention and labor.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"231 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124288025","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 : 2018-02-23DOI: 10.5152/IUJEEE.2018.1807
F. S. Utku, Ozan Özdemir, M. Bakay
A molecularly imprinted polymer (MIP)-based impedimetric biosensor was developed for the electrochemical analysis of low-weight biological molecules. Synthetic polymeric matrices with specific and selective recognition sites, which are complementary to the shapes and sizes of the functional groups of analytes, can be prepared using the molecular imprinting method. In this study, a small molecule, tris(hydroxymethyl)aminomethane (TRIS), was used to coat a graphite pencil tip with a TRIS-containing polyacrylamide gel to fabricate a working electrode. The electrode modification and performance were evaluated using cyclic voltammetry and electrochemical impedance spectroscopy. The electrochemical properties of the modified electrodes were observed using an electrochemical cell comprising a Ag/AgCl reference electrode, a Pt wire as the counter electrode, and a pencil graphite tip as the working electrode using a redox-phosphate buffer solution with different concentrations of TRIS and Ethylenediaminetetraacetic acid (EDTA). The I–V and impedance performance of the chemically modified graphite pencil-tip electrodes exhibited decreased conductance and increased impedance correlating with the increase in TRIS concentration. Thus, MIP-based small-molecule biosensor prototypes can be promising economical replacements over other expensive sensors.
{"title":"A Molecularly Imprinted Polymer Based Biosensor for Electrochemical Impedance Spectroscopic Analysis","authors":"F. S. Utku, Ozan Özdemir, M. Bakay","doi":"10.5152/IUJEEE.2018.1807","DOIUrl":"https://doi.org/10.5152/IUJEEE.2018.1807","url":null,"abstract":"A molecularly imprinted polymer (MIP)-based impedimetric biosensor was developed for the electrochemical analysis of low-weight biological molecules. Synthetic polymeric matrices with specific and selective recognition sites, which are complementary to the shapes and sizes of the functional groups of analytes, can be prepared using the molecular imprinting method. In this study, a small molecule, tris(hydroxymethyl)aminomethane (TRIS), was used to coat a graphite pencil tip with a TRIS-containing polyacrylamide gel to fabricate a working electrode. The electrode modification and performance were evaluated using cyclic voltammetry and electrochemical impedance spectroscopy. The electrochemical properties of the modified electrodes were observed using an electrochemical cell comprising a Ag/AgCl reference electrode, a Pt wire as the counter electrode, and a pencil graphite tip as the working electrode using a redox-phosphate buffer solution with different concentrations of TRIS and Ethylenediaminetetraacetic acid (EDTA). The I–V and impedance performance of the chemically modified graphite pencil-tip electrodes exhibited decreased conductance and increased impedance correlating with the increase in TRIS concentration. Thus, MIP-based small-molecule biosensor prototypes can be promising economical replacements over other expensive sensors.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"32 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120863660","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 : 2018-02-23DOI: 10.5152/IUJEEE.2018.1806
Yunus Babacan
Previous studies have focused on the new mem-elements after the experimental realization of memristor which is the the fourth passive circuit element. This research presents a simple operational transconductance amplifier (OTA)-based memcapacitor and meminductor emulator circuit that does not comprise a mutator circuit. Each emulator was verified using the LTspice simulation program, and the simulation results were found to be compatible with the characteristic behaviors of the memcapacitor and meminductor.
{"title":"An Operational Transconductance Amplifier-based Memcapacitor and Meminductor","authors":"Yunus Babacan","doi":"10.5152/IUJEEE.2018.1806","DOIUrl":"https://doi.org/10.5152/IUJEEE.2018.1806","url":null,"abstract":"Previous studies have focused on the new mem-elements after the experimental realization of memristor which is the the fourth passive circuit element. This research presents a simple operational transconductance amplifier (OTA)-based memcapacitor and meminductor emulator circuit that does not comprise a mutator circuit. Each emulator was verified using the LTspice simulation program, and the simulation results were found to be compatible with the characteristic behaviors of the memcapacitor and meminductor.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122125873","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 : 2018-02-23DOI: 10.5152/IUJEEE.2018.1811
C. P. Uzunoğlu
Signal quality is the key issue for maintaining effective power transmission in electrical networks. In most cases, a high voltage (HV) is transmitted in power systems to decrease power loss. Power quality disturbances are monitored by observing the noise degradation of HV signals. Increased oscillations and high-frequency components of power signals exhibit nonstationary signal characteristics. In this study, a comparative analysis of empirical mode decomposition (EMD) and variational mode decomposition (VMD) was conducted on noisy discharge signals. These techniques were used for adaptive signal decomposition in the time domain, facilitating the evaluation of deeper characteristics of the investigated signal. The HV discharges were obtained using 0.4/40 kV and 8 kVA transformers in a laboratory, and all the current and voltage signal waveforms were recorded using high-frequency current and high-voltage probes. The results demonstrate distinct calculations of EMD and VMD techniques in terms of signal decomposition and extracting intrinsic mode functions (IMFs), which define low- and high-frequency components.
{"title":"A Comparative Study of Empirical and Variational Mode Decomposition on High Voltage Discharges","authors":"C. P. Uzunoğlu","doi":"10.5152/IUJEEE.2018.1811","DOIUrl":"https://doi.org/10.5152/IUJEEE.2018.1811","url":null,"abstract":"Signal quality is the key issue for maintaining effective power transmission in electrical networks. In most cases, a high voltage (HV) is transmitted in power systems to decrease power loss. Power quality disturbances are monitored by observing the noise degradation of HV signals. Increased oscillations and high-frequency components of power signals exhibit nonstationary signal characteristics. In this study, a comparative analysis of empirical mode decomposition (EMD) and variational mode decomposition (VMD) was conducted on noisy discharge signals. These techniques were used for adaptive signal decomposition in the time domain, facilitating the evaluation of deeper characteristics of the investigated signal. The HV discharges were obtained using 0.4/40 kV and 8 kVA transformers in a laboratory, and all the current and voltage signal waveforms were recorded using high-frequency current and high-voltage probes. The results demonstrate distinct calculations of EMD and VMD techniques in terms of signal decomposition and extracting intrinsic mode functions (IMFs), which define low- and high-frequency components.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122160464","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 : 2018-02-23DOI: 10.5152/IUJEEE.2018.1809
Burcu Özbay, S. Cekli
A Viterbi decoder system comprises a convolutional encoder and Viterbi decoder. In general, the code words generated from the input series of convolutional encoder arrive at the decoder through a noisy channel; however, the channel noise can cause corruption of code words. The Viterbi decoder extracts the original input message from the corrupted data using the Viterbi algorithm based on the maximum likelihood principle. A Viterbi decoder mainly comprises four essential units: a branch metrics unit, add-compare-select unit, path metrics unit, and survivor-path memory unit. Related complex calculations are repeated in these units at each clock cycle. In this study, a power- and area-efficient Viterbi decoder architecture that also reduces the computational complexity is proposed. Initially, a hard-decision Viterbi decoder system architecture design for Very Large Scale Integration (VLSI) realization was fulfilled without any further improvement to compare the performance of fundamental and improved designs with respect to power consumption. The initial design constitutes an essential base for the improved power- and area-efficient Viterbi decoder architecture. The improvements were made to achieve the less complex and power-efficient architectural system design. The performance of the proposed architecture was tested by a fieldprogrammable gate array (FPGA) platform, and the results have been reported. The architectural design is described using the Verilog hardware description language for comparing the related tests and performance of FPGA platform.
{"title":"Power-Efficient Viterbi Decoder Architecture and Field Programmeble Gate Arrays Fpga Implementation","authors":"Burcu Özbay, S. Cekli","doi":"10.5152/IUJEEE.2018.1809","DOIUrl":"https://doi.org/10.5152/IUJEEE.2018.1809","url":null,"abstract":"A Viterbi decoder system comprises a convolutional encoder and Viterbi decoder. In general, the code words generated from the input series of convolutional encoder arrive at the decoder through a noisy channel; however, the channel noise can cause corruption of code words. The Viterbi decoder extracts the original input message from the corrupted data using the Viterbi algorithm based on the maximum likelihood principle. A Viterbi decoder mainly comprises four essential units: a branch metrics unit, add-compare-select unit, path metrics unit, and survivor-path memory unit. Related complex calculations are repeated in these units at each clock cycle. In this study, a power- and area-efficient Viterbi decoder architecture that also reduces the computational complexity is proposed. Initially, a hard-decision Viterbi decoder system architecture design for Very Large Scale Integration (VLSI) realization was fulfilled without any further improvement to compare the performance of fundamental and improved designs with respect to power consumption. The initial design constitutes an essential base for the improved power- and area-efficient Viterbi decoder architecture. The improvements were made to achieve the less complex and power-efficient architectural system design. The performance of the proposed architecture was tested by a fieldprogrammable gate array (FPGA) platform, and the results have been reported. The architectural design is described using the Verilog hardware description language for comparing the related tests and performance of FPGA platform.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116783614","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 : 2018-02-23DOI: 10.5152/IUJEEE.2018.1803
Aysel Ersoy Yilmaz, M. Ispirli
Polymer materials that are used in outdoor enviroments are exposed to humidity and polution. In addition, some insulation materials installed in water or embeded to slush containing serious pollutants for a long time. These conditions adversely affect the dielectric properties of insulating materials. In this study, the prepared unsaturated polyester (UPE) samples are stored in an ionized water solution for 8, 24, and 168 hours. This ionized water solution is considered to represent the polluted water in the outdoor environments. Further, the dielectric behaviors of the prepared polyester samples are investigated using the inclined plane test (IPT) method, whereas the leakage current data that are recorded during the experiment are analyzed using the recurrence plot (RP) method. The results of applying the inclined plane test on the unsaturated polyester samples depict that the lifetime of the samples decreased with an increase in the period of storage in the ionized water solution. Further, the results of the recurrence plot analysis reveal that the recurrence plot point density also depicted a decrease with an increase in the period of storage in the ionized water solution.
{"title":"Recurrence Plot Analysis of Unsatured Polyester Samples Subjected to Contamination","authors":"Aysel Ersoy Yilmaz, M. Ispirli","doi":"10.5152/IUJEEE.2018.1803","DOIUrl":"https://doi.org/10.5152/IUJEEE.2018.1803","url":null,"abstract":"Polymer materials that are used in outdoor enviroments are exposed to humidity and polution. In addition, some insulation materials installed in water or embeded to slush containing serious pollutants for a long time. These conditions adversely affect the dielectric properties of insulating materials. In this study, the prepared unsaturated polyester (UPE) samples are stored in an ionized water solution for 8, 24, and 168 hours. This ionized water solution is considered to represent the polluted water in the outdoor environments. Further, the dielectric behaviors of the prepared polyester samples are investigated using the inclined plane test (IPT) method, whereas the leakage current data that are recorded during the experiment are analyzed using the recurrence plot (RP) method. The results of applying the inclined plane test on the unsaturated polyester samples depict that the lifetime of the samples decreased with an increase in the period of storage in the ionized water solution. Further, the results of the recurrence plot analysis reveal that the recurrence plot point density also depicted a decrease with an increase in the period of storage in the ionized water solution.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116069979","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 : 2018-02-23DOI: 10.5152/iujeee.2018.1801
N. Kabaoğlu, R. O. Kabaoglu
Modeling a dead-time system is a common issue in engineering applications. To address this issue, existing research has employed neural networks and fuzzy logic-based intelligent systems. Herein, a dead-time system modeled with the aid of support vector machine regression, which has a good generalization feature, was investigated. The performance of the method proposed herein was examined with different parameters in linear and nonlinear dead-time systems.
{"title":"An Analysis of the Effects of SVM Parameters on the Dead-Time System Modeling","authors":"N. Kabaoğlu, R. O. Kabaoglu","doi":"10.5152/iujeee.2018.1801","DOIUrl":"https://doi.org/10.5152/iujeee.2018.1801","url":null,"abstract":"Modeling a dead-time system is a common issue in engineering applications. To address this issue, existing research has employed neural networks and fuzzy logic-based intelligent systems. Herein, a dead-time system modeled with the aid of support vector machine regression, which has a good generalization feature, was investigated. The performance of the method proposed herein was examined with different parameters in linear and nonlinear dead-time systems.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129803429","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 : 2018-02-23DOI: 10.5152/IUJEEE.2018.1805
I. Kocaarslan, M. Akçay, Abdurrahim Akgundogdu, Hasan Tiryaki
Railway electrification systems are designed with regard to the operating data and design parameters. The minimum voltage rating required by traction during the operation should be provided. The maximum voltage drop on a line determines the minimum traction voltage. This voltage should be maintened within certain limits for the continuity of operation. In this study, the maximum voltage drop generated via traction was determined using artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) for a 25-kV AC-supplied railway. The voltage drop on line was calculated with regard to the operating data using ANN and ANFIS. ANN and ANFIS were explained, and the results were compared. The Levenberg–Marquardt (LM) algorithm was used for the ANN model. The LM algorithm is preferred because of the speed and stability it provides for the training of ANNs. The data created for one-way supply status were examined for simulation.
{"title":"Comparison of ANN and ANFIS Methods for the Voltage-Drop Prediction on an Electric Railway Line","authors":"I. Kocaarslan, M. Akçay, Abdurrahim Akgundogdu, Hasan Tiryaki","doi":"10.5152/IUJEEE.2018.1805","DOIUrl":"https://doi.org/10.5152/IUJEEE.2018.1805","url":null,"abstract":"Railway electrification systems are designed with regard to the operating data and design parameters. The minimum voltage rating required by traction during the operation should be provided. The maximum voltage drop on a line determines the minimum traction voltage. This voltage should be maintened within certain limits for the continuity of operation. In this study, the maximum voltage drop generated via traction was determined using artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) for a 25-kV AC-supplied railway. The voltage drop on line was calculated with regard to the operating data using ANN and ANFIS. ANN and ANFIS were explained, and the results were compared. The Levenberg–Marquardt (LM) algorithm was used for the ANN model. The LM algorithm is preferred because of the speed and stability it provides for the training of ANNs. The data created for one-way supply status were examined for simulation.","PeriodicalId":256344,"journal":{"name":"Istanbul University - Journal of Electrical and Electronics Engineering","volume":"40 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130640951","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：