Pub Date : 2019-09-16DOI: 10.1109/ICIEA.2019.8833893
N. Vahabi, Mohammed A. Alabdullah, D. Selviah
This study was conducted to estimate the downhole speed of flow in oil wells and determined the flow direction by analyzing acoustic data recorded by fibre optic distributed acoustic sensors. The signals generated from acoustic data are in the time versus distance domain that are then normalized and differentiated with respect to distance. A 2D Fast Fourier Transform is used to convert time to frequency and distance to wave-number for subsequent calculation. A Gamma correction function was employed to enhance an intensity of the signal in the frequency wevenumber domain. Also, decaying function was successfully applied to enhance the signals with a very low frequencies. We developed a novel method called integration along the radius in polar coordinate to measure the speed of sound and calculating the speed of oil flow. We compared the performance of our method with a Radon transform and proved our method outperforms an existing methods in both processing time and accuracy. The data sets used in this study are recorded from real oil and gas pipes which means there is no controlled environment and there are lots of noisy signals as a result of unpredicted events under the sea. The result of this study is applicable in Oil and Gas production energy industry, Hydraulic fracturing and shale gas extraction energy industry, Borehole water supply industry, Gas pipeline transportation energy industry and Carbon Dioxide Sequestration industry.
{"title":"Acoustic Sensors to Measure Speed of Oil Flow in Downhole Pipes","authors":"N. Vahabi, Mohammed A. Alabdullah, D. Selviah","doi":"10.1109/ICIEA.2019.8833893","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8833893","url":null,"abstract":"This study was conducted to estimate the downhole speed of flow in oil wells and determined the flow direction by analyzing acoustic data recorded by fibre optic distributed acoustic sensors. The signals generated from acoustic data are in the time versus distance domain that are then normalized and differentiated with respect to distance. A 2D Fast Fourier Transform is used to convert time to frequency and distance to wave-number for subsequent calculation. A Gamma correction function was employed to enhance an intensity of the signal in the frequency wevenumber domain. Also, decaying function was successfully applied to enhance the signals with a very low frequencies. We developed a novel method called integration along the radius in polar coordinate to measure the speed of sound and calculating the speed of oil flow. We compared the performance of our method with a Radon transform and proved our method outperforms an existing methods in both processing time and accuracy. The data sets used in this study are recorded from real oil and gas pipes which means there is no controlled environment and there are lots of noisy signals as a result of unpredicted events under the sea. The result of this study is applicable in Oil and Gas production energy industry, Hydraulic fracturing and shale gas extraction energy industry, Borehole water supply industry, Gas pipeline transportation energy industry and Carbon Dioxide Sequestration industry.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128417474","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8833879
Jun Guo, Xiaoxian Wang, C. Zhai, Jiahao Niu, Siliang Lu
In this study, a two-steps method is proposed to diagnose the wind turbine bearing fault under the variable speed condition without using a tachometer. In the first step, the synchrosqueezing wavelet transform is used to process the bearing vibration signal to extract the rotating phase from the time-frequency plane. In the second step, the original time-domain vibration signal is resampled according to the extracted rotating phase, and then the bearing fault indicator can be recognized in the envelope order spectrum for fault diagnosis. The effectiveness of the proposed method is validated on a direct-drive wind turbine test rig in which a fault bearing is installed on a permanent magnet synchronous generator.
{"title":"Fault diagnosis of wind turbine bearing using synchrosqueezing wavelet transform and order analysis","authors":"Jun Guo, Xiaoxian Wang, C. Zhai, Jiahao Niu, Siliang Lu","doi":"10.1109/ICIEA.2019.8833879","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8833879","url":null,"abstract":"In this study, a two-steps method is proposed to diagnose the wind turbine bearing fault under the variable speed condition without using a tachometer. In the first step, the synchrosqueezing wavelet transform is used to process the bearing vibration signal to extract the rotating phase from the time-frequency plane. In the second step, the original time-domain vibration signal is resampled according to the extracted rotating phase, and then the bearing fault indicator can be recognized in the envelope order spectrum for fault diagnosis. The effectiveness of the proposed method is validated on a direct-drive wind turbine test rig in which a fault bearing is installed on a permanent magnet synchronous generator.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115227678","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8833850
A. K. Fjetland, Jing Zhou, Darshana Abeyrathna, Jan Einar Gravdal
An uncontrolled or unobserved influx or kick during drilling has the potential to induce a well blowout, one of the most harmful incidences during drilling both in regards to economic and environmental cost. Since kicks during drilling are serious risks, it is important to improve kick and loss detection performance and capabilities and to develop automatic flux detection methodology. There are clear patterns during a influx incident. However, due to complex processes and sparse instrumentation it is difficult to predict the behaviour of kicks or losses based on sensor data combined with physical models alone. Emerging technologies within Deep Learning are however quite adapt at picking up on, and quantifying, subtle patterns in time series given enough data. In this paper, a new model is developed using Long Short-Term Memory (LSTM), a Recurrent Deep Neural Network, for kick detection and influx size estimation during drilling operations. The proposed detection methodology is based on simulated drilling dataand involves detecting and quantifying the influx of fluids between fractured formations and the well bore. The results show that the proposed methods are effective both to detect and estimate the influx size during drilling operations, so that corrective actions can be taken before any major problem occurs.
{"title":"Kick Detection and Influx Size Estimation during Offshore Drilling Operations using Deep Learning","authors":"A. K. Fjetland, Jing Zhou, Darshana Abeyrathna, Jan Einar Gravdal","doi":"10.1109/ICIEA.2019.8833850","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8833850","url":null,"abstract":"An uncontrolled or unobserved influx or kick during drilling has the potential to induce a well blowout, one of the most harmful incidences during drilling both in regards to economic and environmental cost. Since kicks during drilling are serious risks, it is important to improve kick and loss detection performance and capabilities and to develop automatic flux detection methodology. There are clear patterns during a influx incident. However, due to complex processes and sparse instrumentation it is difficult to predict the behaviour of kicks or losses based on sensor data combined with physical models alone. Emerging technologies within Deep Learning are however quite adapt at picking up on, and quantifying, subtle patterns in time series given enough data. In this paper, a new model is developed using Long Short-Term Memory (LSTM), a Recurrent Deep Neural Network, for kick detection and influx size estimation during drilling operations. The proposed detection methodology is based on simulated drilling dataand involves detecting and quantifying the influx of fluids between fractured formations and the well bore. The results show that the proposed methods are effective both to detect and estimate the influx size during drilling operations, so that corrective actions can be taken before any major problem occurs.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125113579","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8833996
Wenjun Song, Lijun Yun, Zaiqing Chen, Zhixia Ye
At present, most of the dehazing algorithms are the dehazing process of a single hazy image, and the hazy image resources taken by the binocular camera are scarce. Therefore, in order to get rid of the weather restrictions, to more clearly study the hazy images mechanism and increase the data source of binocular hazy images, so as to lay a foundation for the later research on the dehazing algorithm based on binocular stereo vision. This paper proposes a binocular clear image hazing algorithm based on depth map, which combines atmospheric attenuation models with the characteristics of hazing image depth, and derives a model of the hazy image and then simulates the effect of the hazy image on this basis. In practical applications, simply take a clear image and you can get the desired stereo hazy image with this hazing model. Thereby, it provides convenience for the shooting of haze scene effects such as 3D movies and TV dramas, and has a very large practical application value.
{"title":"Research on binocular clear image hazing algorithm based on depth map","authors":"Wenjun Song, Lijun Yun, Zaiqing Chen, Zhixia Ye","doi":"10.1109/ICIEA.2019.8833996","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8833996","url":null,"abstract":"At present, most of the dehazing algorithms are the dehazing process of a single hazy image, and the hazy image resources taken by the binocular camera are scarce. Therefore, in order to get rid of the weather restrictions, to more clearly study the hazy images mechanism and increase the data source of binocular hazy images, so as to lay a foundation for the later research on the dehazing algorithm based on binocular stereo vision. This paper proposes a binocular clear image hazing algorithm based on depth map, which combines atmospheric attenuation models with the characteristics of hazing image depth, and derives a model of the hazy image and then simulates the effect of the hazy image on this basis. In practical applications, simply take a clear image and you can get the desired stereo hazy image with this hazing model. Thereby, it provides convenience for the shooting of haze scene effects such as 3D movies and TV dramas, and has a very large practical application value.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"35 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122675778","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8834035
U.G.K. Mulleriyawage, W. Shen, Cungang Hu
Due to the intermittency of renewable energy sources (RESs), prosumers-owned DC microgrids (MGs) in a residential house require a battery energy storage system (ESS) to achieve significant energy efficiency improvements. This study investigates the durability and economical gain of different battery ESSs in a DC MG integrated with RESs, taking an Australian household as a case study. These batteries used in the ESS can be lead-acid, lithium-ion, and zinc-bromine chemistries. A model of a DC MG for a residential house has been established to simulate and evaluate performances of different types of batteries as an ESS in the DC MG. The simulation results show that lithium-ion phosphonate (LFP) batteries have more than 20 years of lifetime. Thus, economically only LFP batteries provide a positive return on investment (ROI) values compared to other types of batteries
{"title":"Battery system selection in DC microgrids for residential applications: an Australian case study","authors":"U.G.K. Mulleriyawage, W. Shen, Cungang Hu","doi":"10.1109/ICIEA.2019.8834035","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8834035","url":null,"abstract":"Due to the intermittency of renewable energy sources (RESs), prosumers-owned DC microgrids (MGs) in a residential house require a battery energy storage system (ESS) to achieve significant energy efficiency improvements. This study investigates the durability and economical gain of different battery ESSs in a DC MG integrated with RESs, taking an Australian household as a case study. These batteries used in the ESS can be lead-acid, lithium-ion, and zinc-bromine chemistries. A model of a DC MG for a residential house has been established to simulate and evaluate performances of different types of batteries as an ESS in the DC MG. The simulation results show that lithium-ion phosphonate (LFP) batteries have more than 20 years of lifetime. Thus, economically only LFP batteries provide a positive return on investment (ROI) values compared to other types of batteries","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124010631","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8834234
Chengfeng Zhu, Hongbo Tan, Yanzhong Li, Yang Nie, Jiamin Shi
In pursuit of high efficiency of energy conveying system, a new high temperature superconducting electric power transmission system cooled by LNG (liquefied natural gas) is investigated. LNG is used to contribute a low temperature environment for the superconducting cable operated in the temperature range of 85-95K. On the basis of previous research, the new structure of the combined transmission system consists of a flowing channel for sub-cooled LN2 (liquid nitrogen) as a kind of protective medium. In this paper, the remarkable advantages of the combined transmission system are demonstrated through the analysis. Comparison of efficiencies in different operation condition has been conducted. The results showed that the new system is of great significance for the high efficient transmission of energy.
{"title":"Energy efficiency analysis of a liquefied natural gas and electric power combined transmission system","authors":"Chengfeng Zhu, Hongbo Tan, Yanzhong Li, Yang Nie, Jiamin Shi","doi":"10.1109/ICIEA.2019.8834234","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8834234","url":null,"abstract":"In pursuit of high efficiency of energy conveying system, a new high temperature superconducting electric power transmission system cooled by LNG (liquefied natural gas) is investigated. LNG is used to contribute a low temperature environment for the superconducting cable operated in the temperature range of 85-95K. On the basis of previous research, the new structure of the combined transmission system consists of a flowing channel for sub-cooled LN2 (liquid nitrogen) as a kind of protective medium. In this paper, the remarkable advantages of the combined transmission system are demonstrated through the analysis. Comparison of efficiencies in different operation condition has been conducted. The results showed that the new system is of great significance for the high efficient transmission of energy.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114651109","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8834150
Manisha Verma, S. Gangavarapu, A. Rathore
The continuous urge for improvement in our standard of living has increased the consumption of electrical energy by leaps and bounds. This hike in energy consumption, draining of fossil fuels and degrading global environment has led to invention of green power generation systems. Thus, the global demand for renewable resources has led to flourishing of photovoltaic (PV) market. The enabling technology in the PV systems is the inverter, which could be either: 1) with transformer isolated or 2) without transformer non-isolated (transformer-less inverter). Recently, single phase transformerless voltage source inverters (VSI) have been extensively used for distributed photovoltaic grid tied systems. The objective of this paper is to review a few notable topologies and propose a new topology for transformer-less photovoltaic inverter. The analysis and design of the proposed topology is verified by simulating it on PSIM. Furthermore, the simulation results are validated by testing a proof-of-concept laboratory hardware prototype rated at 250 W.
{"title":"Analysis and Design of a Novel Transformer-less-Inverter in Grid-Connected Photovoltaic System","authors":"Manisha Verma, S. Gangavarapu, A. Rathore","doi":"10.1109/ICIEA.2019.8834150","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8834150","url":null,"abstract":"The continuous urge for improvement in our standard of living has increased the consumption of electrical energy by leaps and bounds. This hike in energy consumption, draining of fossil fuels and degrading global environment has led to invention of green power generation systems. Thus, the global demand for renewable resources has led to flourishing of photovoltaic (PV) market. The enabling technology in the PV systems is the inverter, which could be either: 1) with transformer isolated or 2) without transformer non-isolated (transformer-less inverter). Recently, single phase transformerless voltage source inverters (VSI) have been extensively used for distributed photovoltaic grid tied systems. The objective of this paper is to review a few notable topologies and propose a new topology for transformer-less photovoltaic inverter. The analysis and design of the proposed topology is verified by simulating it on PSIM. Furthermore, the simulation results are validated by testing a proof-of-concept laboratory hardware prototype rated at 250 W.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117312860","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8833898
W. Wenbin, Xin Jianbo, Fan Ruixiang, L. Qiong, Meng Tianqi, Wang Ziyi, Z. Ao
The zero voltage switching (ZVS) condition should be maintained in the inductive power transfer (IPT) system supplied by inverter. It can be obtained when current lags the voltage of the inverter. In this paper, a modeling and control strategy is presented to maintain the ZVS condition of the IPT system. The equivalent impedance angle of the system can be regulated through adjusting the operation frequency. An equivalent circuit model of the IPT system supplied by a half-bridge inverter is firstly developed for analysis. Then the model is linearized around the operation point using small-signal method and is reduced with balance realization method. Based on this low-order model, the transfer function of this model is obtained and a PI controller is presented in discrete-time domain to stabilize the impedance angle. Finally, the simulation results verify the correctness of modeling and the effectiveness of the proposed controller.
{"title":"Modeling and Control Strategy for Zero Voltage Switching Condition in Series-Series Compensated Inductive Power Transfer System","authors":"W. Wenbin, Xin Jianbo, Fan Ruixiang, L. Qiong, Meng Tianqi, Wang Ziyi, Z. Ao","doi":"10.1109/ICIEA.2019.8833898","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8833898","url":null,"abstract":"The zero voltage switching (ZVS) condition should be maintained in the inductive power transfer (IPT) system supplied by inverter. It can be obtained when current lags the voltage of the inverter. In this paper, a modeling and control strategy is presented to maintain the ZVS condition of the IPT system. The equivalent impedance angle of the system can be regulated through adjusting the operation frequency. An equivalent circuit model of the IPT system supplied by a half-bridge inverter is firstly developed for analysis. Then the model is linearized around the operation point using small-signal method and is reduced with balance realization method. Based on this low-order model, the transfer function of this model is obtained and a PI controller is presented in discrete-time domain to stabilize the impedance angle. Finally, the simulation results verify the correctness of modeling and the effectiveness of the proposed controller.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126281802","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8834366
Jun Jiang, Qiang Zhang, Jikui Liu, Yuanzi Zhou, Wen Wen, Wei Liu, Xuepeng Liu
Tip-tilt stage, also known as fast steering mirror, is a key assembly for many fine tuning applications. However, high performance tip-tilt stage often encounters the problem of parasitic error when moving in the desired directions. To address this problem, this paper presented the design and analysis of a novel tiptilt stage based on compliant remote center of motion mechanism. The proposed mechanism is proven to have less parasitic motion and with better pointing accuracy, meanwhile can suspend a relatively large payload (mirror). In this paper, the design details of the stage is elaborated, both theoretical analysis and numerical simulations are conducted to optimize and validate the design. The results indicate that the proposed tip-tilt stage can drive a 2-in size payload with 3nrad resolution and 300Hz bandwidth within 2mrad range in both axes. The proposed stage matches well with the requirements of most space applications.
{"title":"Design and analysis of a novel tip-tilt stage with high precision for space applications","authors":"Jun Jiang, Qiang Zhang, Jikui Liu, Yuanzi Zhou, Wen Wen, Wei Liu, Xuepeng Liu","doi":"10.1109/ICIEA.2019.8834366","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8834366","url":null,"abstract":"Tip-tilt stage, also known as fast steering mirror, is a key assembly for many fine tuning applications. However, high performance tip-tilt stage often encounters the problem of parasitic error when moving in the desired directions. To address this problem, this paper presented the design and analysis of a novel tiptilt stage based on compliant remote center of motion mechanism. The proposed mechanism is proven to have less parasitic motion and with better pointing accuracy, meanwhile can suspend a relatively large payload (mirror). In this paper, the design details of the stage is elaborated, both theoretical analysis and numerical simulations are conducted to optimize and validate the design. The results indicate that the proposed tip-tilt stage can drive a 2-in size payload with 3nrad resolution and 300Hz bandwidth within 2mrad range in both axes. The proposed stage matches well with the requirements of most space applications.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114959184","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 : 2019-06-19DOI: 10.1109/ICIEA.2019.8834085
Yalin Liu, Tianyou Yu, Yuanqing Li
Many brain decoding algorithms are based on the assumption that the training and test data are in the same feature space and have the same distribution. However, in many practical applications, the statistical distribution of brain signal varies across subjects as well as sessions, restricting the transferability of training data or training models between different subjects or sessions. In this study, we focus on this problem in magnetoencephalography (MEG) signal. By reviewing the transfer learning literatures that have achieved satisfactory results in brain decoding, we proposed a new method for MEG decoding across subjects. First, a new kernel is derived by establishing a connection between the Riemannian geometry and the sample covariance matrix (SCM), which is then applied to MEG transfer decoding by using an improved multi-task learning framework. The experiments on an MEG dataset of face vs. scramble decoding task demonstrated that the proposed approach is superior to other comparable methods.
{"title":"Enhance cross-subject MEG decoding by combining Riemannian Geometry feature extraction and multi-task learning","authors":"Yalin Liu, Tianyou Yu, Yuanqing Li","doi":"10.1109/ICIEA.2019.8834085","DOIUrl":"https://doi.org/10.1109/ICIEA.2019.8834085","url":null,"abstract":"Many brain decoding algorithms are based on the assumption that the training and test data are in the same feature space and have the same distribution. However, in many practical applications, the statistical distribution of brain signal varies across subjects as well as sessions, restricting the transferability of training data or training models between different subjects or sessions. In this study, we focus on this problem in magnetoencephalography (MEG) signal. By reviewing the transfer learning literatures that have achieved satisfactory results in brain decoding, we proposed a new method for MEG decoding across subjects. First, a new kernel is derived by establishing a connection between the Riemannian geometry and the sample covariance matrix (SCM), which is then applied to MEG transfer decoding by using an improved multi-task learning framework. The experiments on an MEG dataset of face vs. scramble decoding task demonstrated that the proposed approach is superior to other comparable methods.","PeriodicalId":311302,"journal":{"name":"2019 14th IEEE Conference on Industrial Electronics and Applications (ICIEA)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129815302","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：