Pub Date : 2022-04-22DOI: 10.1109/CEEPE55110.2022.9783378
Jingang Wang, Ya Liu, Tian Tian
To solve the problem that the input of single spectrogram cannot fully express the fault feature of the wind turbine gearbox, a fault diagnosis method of the wind turbine gearbox based on the fusion of the multi-sensor spectrogram and the improved CNN neural network is proposed. Firstly, in view of the problem of aliasing of vibration signal components of wind turbine gearboxes, the vibration signals of each sensor are sparsely decomposed to obtain high resonance components including gear harmonic components and low resonance components that may include bearing fault impact components. The high-resonance component and low-resonance component spectrograms of the sensor are fused as the input of the convolutional neural network; secondly, the fault diagnosis model of the wind turbine gearbox that fuses the multispectrogram and the improved CNN neural network is constructed and trained; finally, through QPZZ-II The experimental platform for fault diagnosis of rotating machinery verifies the effectiveness of the proposed method. The results show that the proposed method has a high accuracy of 98.55% for fault diagnosis of wind turbine gearboxes.
{"title":"Fault Diagnosis Method of Wind Turbine Gearbox Based on Fusion Multispectrogram and Improved CNN Neural Network","authors":"Jingang Wang, Ya Liu, Tian Tian","doi":"10.1109/CEEPE55110.2022.9783378","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783378","url":null,"abstract":"To solve the problem that the input of single spectrogram cannot fully express the fault feature of the wind turbine gearbox, a fault diagnosis method of the wind turbine gearbox based on the fusion of the multi-sensor spectrogram and the improved CNN neural network is proposed. Firstly, in view of the problem of aliasing of vibration signal components of wind turbine gearboxes, the vibration signals of each sensor are sparsely decomposed to obtain high resonance components including gear harmonic components and low resonance components that may include bearing fault impact components. The high-resonance component and low-resonance component spectrograms of the sensor are fused as the input of the convolutional neural network; secondly, the fault diagnosis model of the wind turbine gearbox that fuses the multispectrogram and the improved CNN neural network is constructed and trained; finally, through QPZZ-II The experimental platform for fault diagnosis of rotating machinery verifies the effectiveness of the proposed method. The results show that the proposed method has a high accuracy of 98.55% for fault diagnosis of wind turbine gearboxes.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131319062","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}
Currently, demand of precise and refined energy consumption service for residents is growing constantly. Aiming at this situation, according to the principles of International Recommendation 46, an electricity meter for non-intrusive load monitoring based on multicore modularity is designed to improve the level of detailed electrical data analysis. With general function module, metering module and load analysis module embedded separately, intelligent identification for type, turn-on/off time and electricity consumption of household appliances is achieved in triple cores management mode by load data sampling, characteristic quantity extraction, load matching and record upload. At the same time, a remote update scheme relying on electricity information acquisition system is proposed to ensure that load identification algorithm and load signature database could be updated online.
{"title":"Development of a Smart Meter with Non-intrusive Load Monitoring Function","authors":"Hanmiao Cheng, Zengkai Ouyang, Zecheng Ding, Kaijie Fang, Yixuan Huang, Wei Tang","doi":"10.1109/CEEPE55110.2022.9783385","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783385","url":null,"abstract":"Currently, demand of precise and refined energy consumption service for residents is growing constantly. Aiming at this situation, according to the principles of International Recommendation 46, an electricity meter for non-intrusive load monitoring based on multicore modularity is designed to improve the level of detailed electrical data analysis. With general function module, metering module and load analysis module embedded separately, intelligent identification for type, turn-on/off time and electricity consumption of household appliances is achieved in triple cores management mode by load data sampling, characteristic quantity extraction, load matching and record upload. At the same time, a remote update scheme relying on electricity information acquisition system is proposed to ensure that load identification algorithm and load signature database could be updated online.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115366782","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 : 2022-04-22DOI: 10.1109/CEEPE55110.2022.9783405
Zhang Pengcheng, Z. Xiu, Tian Tian, Luo Yan, L. Xiuguang, Sun Jun
Based on the advanced and flexible acoustic detection of electrical equipment, the characteristics of acoustic signal of the insulation defects discharge of the real dry-type reactor are studied in this paper. The contamination defects and metal foreign body defects are set up, and the acoustic signals of the discharge are obtained, and the acoustic characteristics are analyzed based on the power spectrum. The results show that the discharge acoustic signal cycle of metal foreign body defect is 20 ms, and the main frequency bands are 6 ~ 8 kHz and 11 ~ 19 kHz; The acoustic signal of encapsulated pollution discharge is in the shape of "Hump", mainly with high frequency of 12 ~ 17 kHz; It is found that the power spectrum curve of metal foreign body defects fluctuates greatly and has a larger Var, while the power spectrum of encapsulated pollution defects is mainly composed of high-frequency components and has a larger MF value.
{"title":"Research on Insulation Defect Discharge Characteristics of Dry-Type Reactor Based on Acoustic Power Spectral Density","authors":"Zhang Pengcheng, Z. Xiu, Tian Tian, Luo Yan, L. Xiuguang, Sun Jun","doi":"10.1109/CEEPE55110.2022.9783405","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783405","url":null,"abstract":"Based on the advanced and flexible acoustic detection of electrical equipment, the characteristics of acoustic signal of the insulation defects discharge of the real dry-type reactor are studied in this paper. The contamination defects and metal foreign body defects are set up, and the acoustic signals of the discharge are obtained, and the acoustic characteristics are analyzed based on the power spectrum. The results show that the discharge acoustic signal cycle of metal foreign body defect is 20 ms, and the main frequency bands are 6 ~ 8 kHz and 11 ~ 19 kHz; The acoustic signal of encapsulated pollution discharge is in the shape of \"Hump\", mainly with high frequency of 12 ~ 17 kHz; It is found that the power spectrum curve of metal foreign body defects fluctuates greatly and has a larger Var, while the power spectrum of encapsulated pollution defects is mainly composed of high-frequency components and has a larger MF value.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"4 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115683802","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 : 2022-04-22DOI: 10.1109/CEEPE55110.2022.9783284
Cong Zeng, Jizhong Zhu
To achieve more accurate operation of power system, the models are sometime represented as a non-linear, non-convex and even multi-peak function, such as the fuel cost function with valve-point effects. However, to the best of the author’s knowledge, the existing distributed optimization algorithms can only solve the convex problem. It is urged that proposing a novel distributed optimization technique to solve non-convex optimization problem. Due to the strong generalization ability of meta-heuristic algorithm, a novel distributed meta-heuristic optimization algorithm is proposed in this paper. In the meantime, a distributed power flow solution algorithm is embedded into each iteration. These two algorithms constitute a novel double-layer optimization mechanism to solve the non-convex optimal power flow (NCOPF), in which a set of optimizers (operators essentially) across the network to optimize the operation of each generation area in parallel while minimizing the total operational cost of the entire multi-area power system. The experimental results in two test systems demonstrate that the proposed algorithm does not implement distributed NCOPF solution only but also improve the solution accuracy, accelerate the convergence speed and enhance the robustness, especially in the large-scale system.
{"title":"Non-Convex Optimal Power Flow Implementation by Distributed Meta-Heuristic Optimization Algorithm","authors":"Cong Zeng, Jizhong Zhu","doi":"10.1109/CEEPE55110.2022.9783284","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783284","url":null,"abstract":"To achieve more accurate operation of power system, the models are sometime represented as a non-linear, non-convex and even multi-peak function, such as the fuel cost function with valve-point effects. However, to the best of the author’s knowledge, the existing distributed optimization algorithms can only solve the convex problem. It is urged that proposing a novel distributed optimization technique to solve non-convex optimization problem. Due to the strong generalization ability of meta-heuristic algorithm, a novel distributed meta-heuristic optimization algorithm is proposed in this paper. In the meantime, a distributed power flow solution algorithm is embedded into each iteration. These two algorithms constitute a novel double-layer optimization mechanism to solve the non-convex optimal power flow (NCOPF), in which a set of optimizers (operators essentially) across the network to optimize the operation of each generation area in parallel while minimizing the total operational cost of the entire multi-area power system. The experimental results in two test systems demonstrate that the proposed algorithm does not implement distributed NCOPF solution only but also improve the solution accuracy, accelerate the convergence speed and enhance the robustness, especially in the large-scale system.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114477983","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 : 2022-04-22DOI: 10.1109/CEEPE55110.2022.9783354
Dezheng Jiang, Yubi Jing
The fluoride organics will experience the decomposition under the high temperature and high incident energy, which would have an effect on the system and equipment. The preliminary decomposition of HFC143a is investigated via reactive molecular dynamics in the present work. The results indicate that incident energy will aggravate the decomposition of HFC143a. And the main product of the reaction is FH. Besides, the first-order kinetic analysis is employed to analysis the reaction and obtain the relationship between the decomposition rate, temperature and the incident energy.
{"title":"Preliminary Study on Decomposition of HFC143a: A Reactive Molecular Dynamic Study","authors":"Dezheng Jiang, Yubi Jing","doi":"10.1109/CEEPE55110.2022.9783354","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783354","url":null,"abstract":"The fluoride organics will experience the decomposition under the high temperature and high incident energy, which would have an effect on the system and equipment. The preliminary decomposition of HFC143a is investigated via reactive molecular dynamics in the present work. The results indicate that incident energy will aggravate the decomposition of HFC143a. And the main product of the reaction is FH. Besides, the first-order kinetic analysis is employed to analysis the reaction and obtain the relationship between the decomposition rate, temperature and the incident energy.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128257773","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 : 2022-04-22DOI: 10.1109/CEEPE55110.2022.9783440
Xingang Yang, Boyuan Cao, Zhibin Liu, Yang Du, Lingyu Guo, Zhongguang Yang
With the large-scale and cluster development of offshore wind farms, it has become an inevitable trend for the development of offshore wind farm integration to form an intensive mode of interconnection of offshore power transmission network and land power network. Aiming at the problem of fixed cost allocation for offshore wind farm integration system under multi-stakeholder investment., this paper proposes a fixed cost allocation method for offshore wind farm integration system considering the correlation of wind power output. First, considering the correlation between wind speeds in offshore wind farms, the PSO is used to optimize the weight coefficient of the mixed Copula function, and a wind speed correlation model based on the mixed Copula function is established. Secondly, for the fixed cost allocation problem of the offshore wind farm integration system, the fixed cost allocation method based on the Shapley value solution of cooperative game is used to establish the cooperative game model of the fixed cost allocation of the offshore wind farm integration system. Taking the IEEE30 node standard test system as an case, the simulation results of the case verify the effectiveness and superiority of the proposed method.
{"title":"Fixed Cost Allocation Method of Offshore Wind Farm Integration System Considering Wind Power output Correlation","authors":"Xingang Yang, Boyuan Cao, Zhibin Liu, Yang Du, Lingyu Guo, Zhongguang Yang","doi":"10.1109/CEEPE55110.2022.9783440","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783440","url":null,"abstract":"With the large-scale and cluster development of offshore wind farms, it has become an inevitable trend for the development of offshore wind farm integration to form an intensive mode of interconnection of offshore power transmission network and land power network. Aiming at the problem of fixed cost allocation for offshore wind farm integration system under multi-stakeholder investment., this paper proposes a fixed cost allocation method for offshore wind farm integration system considering the correlation of wind power output. First, considering the correlation between wind speeds in offshore wind farms, the PSO is used to optimize the weight coefficient of the mixed Copula function, and a wind speed correlation model based on the mixed Copula function is established. Secondly, for the fixed cost allocation problem of the offshore wind farm integration system, the fixed cost allocation method based on the Shapley value solution of cooperative game is used to establish the cooperative game model of the fixed cost allocation of the offshore wind farm integration system. Taking the IEEE30 node standard test system as an case, the simulation results of the case verify the effectiveness and superiority of the proposed method.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128742657","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 : 2022-04-22DOI: 10.1109/CEEPE55110.2022.9783300
Ke Sun, Jialin Yu, Bin Han, Yifei Wang
Flexible demand response operation optimization decision is one of the important features of smart distribution network. With the continuous enhancement of the mutual coupling of multiple energy forms on the distribution network side, comprehensive energy conversion devices represented by energy hubs have received extensive attention. How to consider the operation characteristics of the energy hub in the optimal operation of the demand side, so as to bring more flexibility and economy to the energy consumption of the distribution network and users, related research needs to be carried out urgently. This paper proposes a demand response optimization decision-making model considering comprehensive energy demand response, which combines the load profile of power-side users with the comprehensive energy demand of multiple energy network. Collaborative decision-making to achieve the overall optimization of the distribution network and the overall user. The case studies verify the effectiveness of the proposed model and algorithm.
{"title":"Energy Hub Considering the Customer Directrix Line","authors":"Ke Sun, Jialin Yu, Bin Han, Yifei Wang","doi":"10.1109/CEEPE55110.2022.9783300","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783300","url":null,"abstract":"Flexible demand response operation optimization decision is one of the important features of smart distribution network. With the continuous enhancement of the mutual coupling of multiple energy forms on the distribution network side, comprehensive energy conversion devices represented by energy hubs have received extensive attention. How to consider the operation characteristics of the energy hub in the optimal operation of the demand side, so as to bring more flexibility and economy to the energy consumption of the distribution network and users, related research needs to be carried out urgently. This paper proposes a demand response optimization decision-making model considering comprehensive energy demand response, which combines the load profile of power-side users with the comprehensive energy demand of multiple energy network. Collaborative decision-making to achieve the overall optimization of the distribution network and the overall user. The case studies verify the effectiveness of the proposed model and algorithm.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127381373","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}
In this paper, for the detection of missing images of power insulators, the production of image datasets and image classification methods are discussed. Using the drone aerial insulator images from the power grid company, images containing insulators in different scenarios such as high-voltage transmission lines and substations were collected, 2000 insulator images were extracted, and a power insulator database was constructed. Insulator-missing image classification model to verify the robustness of the EfficientNet algorithm. Use EfficientNet to build a transfer learning network, train it, and compare it with the commonly used classifier ResNet-50. By introducing classification evaluation indicators and class activation maps, the experimental results show that EfficientNet-b0 has good transfer ability and can significantly improve the model. Efficiency, better than ResNet-50 for insulator-missing image classification.
{"title":"Image Classification of Missing Insulators Based on EfficientNet","authors":"Jiang Wang, Jinpeng Tang, Jiyi Wei, Yi Wei, Hailin Wang, Mingsheng Qin","doi":"10.1109/CEEPE55110.2022.9783390","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783390","url":null,"abstract":"In this paper, for the detection of missing images of power insulators, the production of image datasets and image classification methods are discussed. Using the drone aerial insulator images from the power grid company, images containing insulators in different scenarios such as high-voltage transmission lines and substations were collected, 2000 insulator images were extracted, and a power insulator database was constructed. Insulator-missing image classification model to verify the robustness of the EfficientNet algorithm. Use EfficientNet to build a transfer learning network, train it, and compare it with the commonly used classifier ResNet-50. By introducing classification evaluation indicators and class activation maps, the experimental results show that EfficientNet-b0 has good transfer ability and can significantly improve the model. Efficiency, better than ResNet-50 for insulator-missing image classification.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132465135","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 : 2022-04-22DOI: 10.1109/CEEPE55110.2022.9783121
Dingwen Lu, Kai Liao, Yangwu Shen, Ming Ma
With the increasing wind power penetration level (WPPL), the frequency regulation support capacity of traditional power systems is insufficient, which seriously affects the safety and stability of system frequency. Aiming at this problem, this paper proposes a robust control strategy of wind turbines based on second-order sliding mode to support the frequency stability of the power system. Firstly, in order to reflect the dynamic behavior of wind turbines participating in frequency regulation support, the wind turbine linearization model is deduced. Secondly, a second-order sliding mode algorithm, namely the super-twisting algorithm, is introduced into the wind turbine frequency regulation support control, and the stability of the proposed strategy is proved by the Lyapunov method. Finally, the effectiveness of the proposed strategy and its advantages are verified by simulation results.
{"title":"A Robust Control Strategy of Wind Turbines for Frequency Regulation Support in Power Systems Based on Second-Order Sliding Mode","authors":"Dingwen Lu, Kai Liao, Yangwu Shen, Ming Ma","doi":"10.1109/CEEPE55110.2022.9783121","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783121","url":null,"abstract":"With the increasing wind power penetration level (WPPL), the frequency regulation support capacity of traditional power systems is insufficient, which seriously affects the safety and stability of system frequency. Aiming at this problem, this paper proposes a robust control strategy of wind turbines based on second-order sliding mode to support the frequency stability of the power system. Firstly, in order to reflect the dynamic behavior of wind turbines participating in frequency regulation support, the wind turbine linearization model is deduced. Secondly, a second-order sliding mode algorithm, namely the super-twisting algorithm, is introduced into the wind turbine frequency regulation support control, and the stability of the proposed strategy is proved by the Lyapunov method. Finally, the effectiveness of the proposed strategy and its advantages are verified by simulation results.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130439895","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}
In recent years, researchers have proposed the use of Near-Infrared Spectroscopy (NIRS) to detect the Degree of Polymerization (DP) of insulating paper, thus obtaining the aging of transformer insulation in a convenient, fast and nondestructive way. To cope with the problem that the previously established quantitative analysis models are no longer applicable to newly produced spectrometers due to the differences between spectrometers, also called calibration transfer problem, we proposed an robust multi-task learning (RMTL) method, which unites the multi-task learning model of trace norm regularization and l2,1 norm regularization to obtain the correlation relationships between tasks, improving the generalization ability of each task and reducing the risk of overfitting. Therefore, RMTL can use the large amount of data accumulated by the host spectrometer (HS) and the small amount of data from the slave spectrometer (SS) to train at the same time to obtain a relatively high-quality quantitative analysis model of the slave machine. In addition, we compare the RMTL method with the classical DS, PDS, MU-PLS, PLS with direct slave modeling, and three other multi-task learning methods with different norm regularization, and the results show that the proposed method has the best performance in terms of root mean square error (RMSE) and correlation coefficient(R) on the dataset.
{"title":"Robust Multi-task Learning for Calibration Transfer in DP Detection by NIRS of Insulating Paper","authors":"Han Li, Xie Jia, Wenbo Zhang, Shaorui Qin, Yuan Li, Guanjun Zhang","doi":"10.1109/CEEPE55110.2022.9783363","DOIUrl":"https://doi.org/10.1109/CEEPE55110.2022.9783363","url":null,"abstract":"In recent years, researchers have proposed the use of Near-Infrared Spectroscopy (NIRS) to detect the Degree of Polymerization (DP) of insulating paper, thus obtaining the aging of transformer insulation in a convenient, fast and nondestructive way. To cope with the problem that the previously established quantitative analysis models are no longer applicable to newly produced spectrometers due to the differences between spectrometers, also called calibration transfer problem, we proposed an robust multi-task learning (RMTL) method, which unites the multi-task learning model of trace norm regularization and l2,1 norm regularization to obtain the correlation relationships between tasks, improving the generalization ability of each task and reducing the risk of overfitting. Therefore, RMTL can use the large amount of data accumulated by the host spectrometer (HS) and the small amount of data from the slave spectrometer (SS) to train at the same time to obtain a relatively high-quality quantitative analysis model of the slave machine. In addition, we compare the RMTL method with the classical DS, PDS, MU-PLS, PLS with direct slave modeling, and three other multi-task learning methods with different norm regularization, and the results show that the proposed method has the best performance in terms of root mean square error (RMSE) and correlation coefficient(R) on the dataset.","PeriodicalId":118143,"journal":{"name":"2022 5th International Conference on Energy, Electrical and Power Engineering (CEEPE)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129252256","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：