2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)最新文献
Pub Date : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704734
E. Aladesanmi, D. Dorrell
This paper investigates and analyzes the impacts of reverse power flow on the power system connecting elements (transformers and cables). For effective investigation and analysis, the paper groups the network under investigation into Zones A and B. Zone A is directly connected to a wind farm. Zone B is far from the wind farm. Different wind energy penetration levels are simulated. The results show that the elements experience reverse power flow at different penetration levels depending on their location and distance from the point of common coupling. Reverse power flow incurs additional element losses. In Zone A, loading and losses increase with reverse power flow, while Zone B experiences loss reduction with increase in reverse power flow. The elements operate within acceptable thermal ratings.
{"title":"Investigation and Assessment of the Impacts of Reverse Power Flow on Power System Network Loading under High Penetration of Wind Energy","authors":"E. Aladesanmi, D. Dorrell","doi":"10.1109/ROBOMECH.2019.8704734","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704734","url":null,"abstract":"This paper investigates and analyzes the impacts of reverse power flow on the power system connecting elements (transformers and cables). For effective investigation and analysis, the paper groups the network under investigation into Zones A and B. Zone A is directly connected to a wind farm. Zone B is far from the wind farm. Different wind energy penetration levels are simulated. The results show that the elements experience reverse power flow at different penetration levels depending on their location and distance from the point of common coupling. Reverse power flow incurs additional element losses. In Zone A, loading and losses increase with reverse power flow, while Zone B experiences loss reduction with increase in reverse power flow. The elements operate within acceptable thermal ratings.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129617151","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704816]
Lulamile Mzamo, A.S. Helberg, Sonja E. Bosch
In this paper, branching entropy techniques and isiXhosa language heuristics are adapted to develop unsupervised morphological segmenters for isiXhosa. An overview of isiXhosa segmentation issues is given, followed by a discussion on previous work in automated segmentation, and segmentation of isiXhosa in particular. Two unsupervised isiXhosa segmenters are presented and compared to a random minimum baseline and Morfessor-Baseline, a standard in unsupervised word segmentation. Morfessor-Baseline outperforms both isiXhosa segmenters at 79.10% boundary identification accuracy. The IsiXhosa Branching Entropy Segmenter (XBES) performance varies depending on the segmentation mode used, with a maximum of 73.39%. The IsiXhosa Heuristic Maximum Likelihood Segmenter (XHMLS) achieves 72.42%. The study suggests that unsupervised isiXhosa morphological segmentation is feasible with better optimization of the current attempts.
{"title":"Towards an unsupervised morphological segmenter for isiXhosa","authors":"Lulamile Mzamo, A.S. Helberg, Sonja E. Bosch","doi":"10.1109/ROBOMECH.2019.8704816]","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704816]","url":null,"abstract":"In this paper, branching entropy techniques and isiXhosa language heuristics are adapted to develop unsupervised morphological segmenters for isiXhosa. An overview of isiXhosa segmentation issues is given, followed by a discussion on previous work in automated segmentation, and segmentation of isiXhosa in particular. Two unsupervised isiXhosa segmenters are presented and compared to a random minimum baseline and Morfessor-Baseline, a standard in unsupervised word segmentation. Morfessor-Baseline outperforms both isiXhosa segmenters at 79.10% boundary identification accuracy. The IsiXhosa Branching Entropy Segmenter (XBES) performance varies depending on the segmentation mode used, with a maximum of 73.39%. The IsiXhosa Heuristic Maximum Likelihood Segmenter (XHMLS) achieves 72.42%. The study suggests that unsupervised isiXhosa morphological segmentation is feasible with better optimization of the current attempts.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129743623","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704733
L. Waswa, B. Bekker
Increasing adoption of PV SSEGs and its integration into the grid has led to increased challenge in grid management. Accurate estimation of the amount of PV SSEGs connected to the grid is key in voltage regulation, management of electricity supply for local suppliers and for the general electricity network design and security.This paper proposes a method that uses data collected during the Eskom Domestic Load Research (DLR) project as a baseline to estimate the amount of PV SSEGs that is currently installed on a low voltage network feeder in South Africa. Customer class is used to identify similar areas. The proposed approach is a non-analytical method that uses aggregated demand data from a feeder, individual customer demand data from the DLR project as well as the solar PV irradiation data. It attempts to account for stochasticity in the loads and solar irradiation and hence model an estimation tool using a probabilistic approach. Ultimately, a methodology is proposed through which an assessment of the installed PV SSEGs on a feeder can be determined using feeder aggregated demand data for a specific set of customers.March data for the year 2000 from Helderberg area in Cape Town is used in the modelling. In this study, MATLAB is used.
{"title":"Estimating Installed PV SSEGs on an LV Feeder using Aggregated Load Demand Data","authors":"L. Waswa, B. Bekker","doi":"10.1109/ROBOMECH.2019.8704733","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704733","url":null,"abstract":"Increasing adoption of PV SSEGs and its integration into the grid has led to increased challenge in grid management. Accurate estimation of the amount of PV SSEGs connected to the grid is key in voltage regulation, management of electricity supply for local suppliers and for the general electricity network design and security.This paper proposes a method that uses data collected during the Eskom Domestic Load Research (DLR) project as a baseline to estimate the amount of PV SSEGs that is currently installed on a low voltage network feeder in South Africa. Customer class is used to identify similar areas. The proposed approach is a non-analytical method that uses aggregated demand data from a feeder, individual customer demand data from the DLR project as well as the solar PV irradiation data. It attempts to account for stochasticity in the loads and solar irradiation and hence model an estimation tool using a probabilistic approach. Ultimately, a methodology is proposed through which an assessment of the installed PV SSEGs on a feeder can be determined using feeder aggregated demand data for a specific set of customers.March data for the year 2000 from Helderberg area in Cape Town is used in the modelling. In this study, MATLAB is used.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130368956","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704780
Samuel Ogunniyi, D. Withey, Stephen Marais
Due to kinematic, physical, or operational constraints in terrain types, it is necessary for a mobile robot to be able to quantify the traversability characteristics of its environment to ensure safe and efficient navigation. The Gaussian Process Regression approach is a supervised method which promises to add completeness to one of the aspects of traversability analyses, namely environment modelling. This paper presents experimental results which demonstrate the effectiveness of Gaussian Process Regression in predicting the values of missing data for artificial environment features as well as actual collected point cloud data. The study concludes that when there are sufficient points the regression fits more closely to the features in the data set, with less error. Also, the prediction model produced by the Gaussian Process Regression method can be useful during robot operation to improve the terrain modelling.
{"title":"Using Gaussian Process Regression for the interpolation of missing 2.5D environment modelling data","authors":"Samuel Ogunniyi, D. Withey, Stephen Marais","doi":"10.1109/ROBOMECH.2019.8704780","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704780","url":null,"abstract":"Due to kinematic, physical, or operational constraints in terrain types, it is necessary for a mobile robot to be able to quantify the traversability characteristics of its environment to ensure safe and efficient navigation. The Gaussian Process Regression approach is a supervised method which promises to add completeness to one of the aspects of traversability analyses, namely environment modelling. This paper presents experimental results which demonstrate the effectiveness of Gaussian Process Regression in predicting the values of missing data for artificial environment features as well as actual collected point cloud data. The study concludes that when there are sufficient points the regression fits more closely to the features in the data set, with less error. Also, the prediction model produced by the Gaussian Process Regression method can be useful during robot operation to improve the terrain modelling.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121739426","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704764
Ntuba Irene Nkhasi, A. Saha
The conventional radial power distribution systems were initially not designed to accommodate distribution generation (DG). For an effective penetration of PV systems on a large-scale into the current distribution network, considerable work to investigate the nature of incompatibility problems has been done and research is being carried out to develop successful integration strategies. This paper aims to address the radial networks’ protection system challenges after embedding photovoltaic generation sources, investigation on the impacts of high PV penetrations on protection systems of distribution networks and lastly make modification and implement proper protection coordination of the grid-tied PV systems with some emphasis on anti-islanding protection. To accomplish the above-mentioned objectives, a radial distribution network is modelled, simulated and protection settings validated. The PV generation system is designed and added to specific distribution feeders and steady-state results obtained. The results show that addition of DGs cause the system to lose its radial power flow. There is an increase in fault contribution hence causing maloperation such as protection coordination mismatch. An overall protection scheme is proposed based on the addition of DG’s and an efficient adaptive protection system for the distribution networks with a considerable penetration of dispersed generations implemented. For the islanded mode, relaying considerations are provided and implementation of anti-islanding techniques achieved. The impact study is performed which is compared with the existing protection scheme and necessary modifications done. The entire analysis is simulated on a real-time digital simulator (RTDS) and results displayed in a MATLAB environment.
{"title":"Protection Coordination And Anti-Islanding Control Of Grid-Connected PV Systems","authors":"Ntuba Irene Nkhasi, A. Saha","doi":"10.1109/ROBOMECH.2019.8704764","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704764","url":null,"abstract":"The conventional radial power distribution systems were initially not designed to accommodate distribution generation (DG). For an effective penetration of PV systems on a large-scale into the current distribution network, considerable work to investigate the nature of incompatibility problems has been done and research is being carried out to develop successful integration strategies. This paper aims to address the radial networks’ protection system challenges after embedding photovoltaic generation sources, investigation on the impacts of high PV penetrations on protection systems of distribution networks and lastly make modification and implement proper protection coordination of the grid-tied PV systems with some emphasis on anti-islanding protection. To accomplish the above-mentioned objectives, a radial distribution network is modelled, simulated and protection settings validated. The PV generation system is designed and added to specific distribution feeders and steady-state results obtained. The results show that addition of DGs cause the system to lose its radial power flow. There is an increase in fault contribution hence causing maloperation such as protection coordination mismatch. An overall protection scheme is proposed based on the addition of DG’s and an efficient adaptive protection system for the distribution networks with a considerable penetration of dispersed generations implemented. For the islanded mode, relaying considerations are provided and implementation of anti-islanding techniques achieved. The impact study is performed which is compared with the existing protection scheme and necessary modifications done. The entire analysis is simulated on a real-time digital simulator (RTDS) and results displayed in a MATLAB environment.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126513160","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704741
S. Schoonveldt, B. Kotze
Swimming pools need some sort of alarm system to indicate if someone or an article have dropped into the water. This article conveys the methods and practices used in the detection and analysis of low frequency vibrations in water for this purpose. Logged data are used to develop a digital signal processing system that is able to distinguish between different vibration signatures to determine the source of the vibration. Conclusions are drawn from the results and the viability of the technology is evaluated.
{"title":"Detection and analysis of low frequency vibrations to determine their sources in a water medium","authors":"S. Schoonveldt, B. Kotze","doi":"10.1109/ROBOMECH.2019.8704741","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704741","url":null,"abstract":"Swimming pools need some sort of alarm system to indicate if someone or an article have dropped into the water. This article conveys the methods and practices used in the detection and analysis of low frequency vibrations in water for this purpose. Logged data are used to develop a digital signal processing system that is able to distinguish between different vibration signatures to determine the source of the vibration. Conclusions are drawn from the results and the viability of the technology is evaluated.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128055285","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 : 2019-01-01DOI: 10.1109/robomech.2019.8704791
{"title":"SAUPEC/RobMech/PRASA 2019 Program Committee","authors":"","doi":"10.1109/robomech.2019.8704791","DOIUrl":"https://doi.org/10.1109/robomech.2019.8704791","url":null,"abstract":"","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128140444","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704747
R. Stopforth, S. Davrajh
Recent research has shown that Yagi-patch antennae are able to act as non-contact electrodes, to read electro-magnetic signal generated from muscles. Using the same principle, this paper investigates the possibility of using similar electrodes, and the characteristic performance, when tested on an electroencephalogram headset. Tests were conducted on ten volunteers. From the results obtained, a suggested configuration is discussed to be used for future research, such as in robotic and mechatronics applications.
{"title":"Contactless Yagi-Patch Electrodes for Electroencephalogram (EEG) headsets, to be used for Robotic Applications","authors":"R. Stopforth, S. Davrajh","doi":"10.1109/ROBOMECH.2019.8704747","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704747","url":null,"abstract":"Recent research has shown that Yagi-patch antennae are able to act as non-contact electrodes, to read electro-magnetic signal generated from muscles. Using the same principle, this paper investigates the possibility of using similar electrodes, and the characteristic performance, when tested on an electroencephalogram headset. Tests were conducted on ten volunteers. From the results obtained, a suggested configuration is discussed to be used for future research, such as in robotic and mechatronics applications.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134582589","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704735
O. Oni, A. Swanson, R. P. Carpanen
This paper presents a small signal stability analysis of Line Commutated Converter, High Voltage Direct Current (LCC HVDC) system when interconnected at different buses in a Kundur two-area four-machine system. The swing equation of a single machine, which was represented as a generator connected to a large infinite network and two incoherent two-area machines were analysed. A time domain simulation study was carried out on the system during a small signal disturbance on a synchronous generator to determine the damping characteristics and the amplitude of oscillations when LCC-HVDC model was connected at different strategic bus on the network. The results show a more quickly damped oscillation and more stable operating mode with HVDC link compare to HVAC lines.
{"title":"Small Signal Stability Analysis of a four Machine System with Strategic Placement of Monopolar LCC-HVDC link","authors":"O. Oni, A. Swanson, R. P. Carpanen","doi":"10.1109/ROBOMECH.2019.8704735","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704735","url":null,"abstract":"This paper presents a small signal stability analysis of Line Commutated Converter, High Voltage Direct Current (LCC HVDC) system when interconnected at different buses in a Kundur two-area four-machine system. The swing equation of a single machine, which was represented as a generator connected to a large infinite network and two incoherent two-area machines were analysed. A time domain simulation study was carried out on the system during a small signal disturbance on a synchronous generator to determine the damping characteristics and the amplitude of oscillations when LCC-HVDC model was connected at different strategic bus on the network. The results show a more quickly damped oscillation and more stable operating mode with HVDC link compare to HVAC lines.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117066817","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 : 2019-01-01DOI: 10.1109/ROBOMECH.2019.8704835
J. Badenhorst, L. Martinus, Febe de Wet
Since the release of the National Centre for Human Language Technology (NCHLT) Speech corpus, very few additional resources for automatic speech recognition (ASR) system development have been created for South Africa’s eleven official languages. The NCHLT corpus contained a curated but limited subset of the collected data. In this study the auxiliary data that was not included in the released corpus was processed with the aim to improve the acoustic modelling of the NCHLT data. Recent advances in ASR modelling that incorporate deep learning approaches require even more data than previous techniques. Sophisticated neural models seem to accommodate the variability between related acoustic units better and are capable of exploiting speech resources containing more training examples. Our results show that time delay neural networks (TDNN) combined with bi-directional long short-term memory (BLSTM) models are effective, significantly reducing error rates across all languages with just 56 hours of training data. In addition, a cross-corpus evaluation of an Afrikaans system trained on the original NCHLT data plus harvested auxiliary data shows further improvements on this baseline.
{"title":"BLSTM harvesting of auxiliary NCHLT speech data","authors":"J. Badenhorst, L. Martinus, Febe de Wet","doi":"10.1109/ROBOMECH.2019.8704835","DOIUrl":"https://doi.org/10.1109/ROBOMECH.2019.8704835","url":null,"abstract":"Since the release of the National Centre for Human Language Technology (NCHLT) Speech corpus, very few additional resources for automatic speech recognition (ASR) system development have been created for South Africa’s eleven official languages. The NCHLT corpus contained a curated but limited subset of the collected data. In this study the auxiliary data that was not included in the released corpus was processed with the aim to improve the acoustic modelling of the NCHLT data. Recent advances in ASR modelling that incorporate deep learning approaches require even more data than previous techniques. Sophisticated neural models seem to accommodate the variability between related acoustic units better and are capable of exploiting speech resources containing more training examples. Our results show that time delay neural networks (TDNN) combined with bi-directional long short-term memory (BLSTM) models are effective, significantly reducing error rates across all languages with just 56 hours of training data. In addition, a cross-corpus evaluation of an Afrikaans system trained on the original NCHLT data plus harvested auxiliary data shows further improvements on this baseline.","PeriodicalId":344332,"journal":{"name":"2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116426351","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}
2019 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa (SAUPEC/RobMech/PRASA)
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：