Xing Kun-peng, Xue Yang, Kong De-yan, Dong Wei, Ji Zhen-yan
Although the fine-tuning pre-training model technique has obtained tremendous success in the domains of named entity recognition and relation extraction, realistic scenarios exist with many triples of nested entities and overlapping relations. Existing works focus on solving the overlapping triple problem where multiple relational triples in the same sentence share the same entity. In this work, we introduce a joint entity-relation extraction framework based on hybrid feature representation. Our framework consists of five primary parts: constructing hybrid feature representations, bidirectional LSTM encoder, head entity recognition module, entity type classification, and relation tail entity recognition. First, we fuse character-level vector and word-level vector representations via a max-pooling operation to enrich text feature information. Second, the hybrid feature representation is fed into a bidirectional LSTM to capture the correlation between characters and entities. Third, the head entity recognition module employs two identical binary classifiers to detect the start and end positions of entities separately. Then the entity type classification module filters out entities classified as non-entity types by softmax. Finally, we regard relation tail entity recognition as a machine reading comprehension task to eliminate the problem of entity overlap. Specifically, we regard the combination of head entities and relations as queries to query possible tail entities from the text. This framework efficiently handles the polysemy problem, considerably enhances knowledge extraction efficiency, and accurately extracts overlapping triples in domain texts with complicated relationships.
{"title":"Joint Extraction of Entities and Relations Based on Hybrid Feature Representations","authors":"Xing Kun-peng, Xue Yang, Kong De-yan, Dong Wei, Ji Zhen-yan","doi":"10.1115/icone29-93152","DOIUrl":"https://doi.org/10.1115/icone29-93152","url":null,"abstract":"\u0000 Although the fine-tuning pre-training model technique has obtained tremendous success in the domains of named entity recognition and relation extraction, realistic scenarios exist with many triples of nested entities and overlapping relations. Existing works focus on solving the overlapping triple problem where multiple relational triples in the same sentence share the same entity. In this work, we introduce a joint entity-relation extraction framework based on hybrid feature representation. Our framework consists of five primary parts: constructing hybrid feature representations, bidirectional LSTM encoder, head entity recognition module, entity type classification, and relation tail entity recognition. First, we fuse character-level vector and word-level vector representations via a max-pooling operation to enrich text feature information. Second, the hybrid feature representation is fed into a bidirectional LSTM to capture the correlation between characters and entities. Third, the head entity recognition module employs two identical binary classifiers to detect the start and end positions of entities separately. Then the entity type classification module filters out entities classified as non-entity types by softmax. Finally, we regard relation tail entity recognition as a machine reading comprehension task to eliminate the problem of entity overlap. Specifically, we regard the combination of head entities and relations as queries to query possible tail entities from the text. This framework efficiently handles the polysemy problem, considerably enhances knowledge extraction efficiency, and accurately extracts overlapping triples in domain texts with complicated relationships.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133317534","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}
Building Information Modeling theory has been more and more welcome all over the world, and has been introduced into nuclear power industry. The so-called ‘smart power station’ being established by many nuclear power enterprises is the embodiment of BIM. With the deepening of digital design and application, the design data has become more and more detailed and accurate, to support all design work such as data exchange between systems and software, automatic drafting, automatic material and working counting, and so on. In addition to design, how should the procurement, construction, installation and power station operation engineers use these data to improve their work efficiency and accuracy is a research topic, which has been highly followed by engineering companies at present. At the same time, the design data is more for design itself. Whether its expression form, depth and breadth can meet the requirements of downstream engineers and how to improve efficiently when they are not satisfied is an important issue that all project participants should pay attention to. This paper gives a sketch of 3D design data, with pipework as an example, to elaborate how to use design data to help other engineers to break down old-style of working mode, and gives some thinking about optimizing design data efficiently to fulfill the requirement of other project participants.
{"title":"Application of Detail Design Data of Nuclear Power Plant in Project Construction and Operation Under the Background of Digitization","authors":"Huizhen Huang, Licheng Tian, Jiafu Yan","doi":"10.1115/icone29-93125","DOIUrl":"https://doi.org/10.1115/icone29-93125","url":null,"abstract":"\u0000 Building Information Modeling theory has been more and more welcome all over the world, and has been introduced into nuclear power industry. The so-called ‘smart power station’ being established by many nuclear power enterprises is the embodiment of BIM. With the deepening of digital design and application, the design data has become more and more detailed and accurate, to support all design work such as data exchange between systems and software, automatic drafting, automatic material and working counting, and so on. In addition to design, how should the procurement, construction, installation and power station operation engineers use these data to improve their work efficiency and accuracy is a research topic, which has been highly followed by engineering companies at present. At the same time, the design data is more for design itself. Whether its expression form, depth and breadth can meet the requirements of downstream engineers and how to improve efficiently when they are not satisfied is an important issue that all project participants should pay attention to. This paper gives a sketch of 3D design data, with pipework as an example, to elaborate how to use design data to help other engineers to break down old-style of working mode, and gives some thinking about optimizing design data efficiently to fulfill the requirement of other project participants.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"458 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133469012","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}
Bing Chen, Guidan Zhao, Zhiping Zhong, Yong Zheng, Charlie Yu
Intelligence is the trend of industrial development in today’s world, which will inevitably affect the nuclear power field, thus promoting nuclear power industry from digital to intelligent forward.5G and Wi-Fi6, as a new generation of wireless communication technologies, have important advantages in terms of technical performance and security due to the characteristics of dense personnel, wide equipment distribution, complex facility environment and special materials used in nuclear power scenarios. Considering the high requirement of electromagnetic compatibility in nuclear power environment, it is urgent to further study the electromagnetic interference and protection requirements of 5G, Wi-Fi6 and other new technologies on nuclear power unit equipment in nuclear power environment. In this paper, combined with relevant national document standards and industry standards, we will sort out the requirements of Immunity about Radio Frequency Radiation (RFR) for the important equipment in nuclear power plants, and use signal generators to simulate new radio signals such as 5G and Wi-Fi6 to test the immunity of nuclear power instruments and meters. Combined the results of test, we will study the Electro Magnetic Compatibility (EMC) between new wireless technologies and nuclear power plant equipment from different factors, such as Power, bandwidth, modulation, distance, direction. And then, we will summarize the protection requirements about nuclear power equipment.
{"title":"Research on Electromagnetic Interference Protection of Nuclear Power Unit Equipment Under the Application of New Wireless Technology in Smart Nuclear Power","authors":"Bing Chen, Guidan Zhao, Zhiping Zhong, Yong Zheng, Charlie Yu","doi":"10.1115/icone29-91926","DOIUrl":"https://doi.org/10.1115/icone29-91926","url":null,"abstract":"\u0000 Intelligence is the trend of industrial development in today’s world, which will inevitably affect the nuclear power field, thus promoting nuclear power industry from digital to intelligent forward.5G and Wi-Fi6, as a new generation of wireless communication technologies, have important advantages in terms of technical performance and security due to the characteristics of dense personnel, wide equipment distribution, complex facility environment and special materials used in nuclear power scenarios. Considering the high requirement of electromagnetic compatibility in nuclear power environment, it is urgent to further study the electromagnetic interference and protection requirements of 5G, Wi-Fi6 and other new technologies on nuclear power unit equipment in nuclear power environment.\u0000 In this paper, combined with relevant national document standards and industry standards, we will sort out the requirements of Immunity about Radio Frequency Radiation (RFR) for the important equipment in nuclear power plants, and use signal generators to simulate new radio signals such as 5G and Wi-Fi6 to test the immunity of nuclear power instruments and meters. Combined the results of test, we will study the Electro Magnetic Compatibility (EMC) between new wireless technologies and nuclear power plant equipment from different factors, such as Power, bandwidth, modulation, distance, direction. And then, we will summarize the protection requirements about nuclear power equipment.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"438 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132895579","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}
Yong Liu, Xiangyu Li, Biao Liang, Bo Wang, Sichao Tan, P. Gao
Traditional machine learning algorithms have problems such as overfitting, low accuracy, and difficulty in hyperparameter optimization when performing fault diagnosis.In order to improve the accident diagnosis ability of nuclear power plant reactor system, this paper combines Bayesian optimization (BO) algorithm with eXtreme Gradient Boosting (XGBoost) algorithm to develop a reactor accident diagnosis model.First, data preprocessing and feature quantity analysis are performed on accident data samples.Then, the BO algorithm is used to optimize the hyperparameters of the XGBoost model. Finally, the BO-XGBoost model is used to diagnose the operating conditions of seven nuclear power plants, and the diagnostic effects of various traditional machine learning classification algorithms are compared and analyzed.The results show that the BO-XGBoost model can achieve more efficient and accurate identification of reactor accident types, and the model has better generalization ability.This research can help nuclear power plant operators to accurately identify the types of reactor accidents, assist decision-making, and ensure the safe operation of nuclear power plants.
{"title":"Research on Accident Diagnosis Method of Reactor System Based on XGBoost Using Bayesian Optimization","authors":"Yong Liu, Xiangyu Li, Biao Liang, Bo Wang, Sichao Tan, P. Gao","doi":"10.1115/icone29-92061","DOIUrl":"https://doi.org/10.1115/icone29-92061","url":null,"abstract":"\u0000 Traditional machine learning algorithms have problems such as overfitting, low accuracy, and difficulty in hyperparameter optimization when performing fault diagnosis.In order to improve the accident diagnosis ability of nuclear power plant reactor system, this paper combines Bayesian optimization (BO) algorithm with eXtreme Gradient Boosting (XGBoost) algorithm to develop a reactor accident diagnosis model.First, data preprocessing and feature quantity analysis are performed on accident data samples.Then, the BO algorithm is used to optimize the hyperparameters of the XGBoost model. Finally, the BO-XGBoost model is used to diagnose the operating conditions of seven nuclear power plants, and the diagnostic effects of various traditional machine learning classification algorithms are compared and analyzed.The results show that the BO-XGBoost model can achieve more efficient and accurate identification of reactor accident types, and the model has better generalization ability.This research can help nuclear power plant operators to accurately identify the types of reactor accidents, assist decision-making, and ensure the safe operation of nuclear power plants.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114244945","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}
The research and development of default value development and management platform of nuclear power plant control system can realize the functions of default value analysis, value selection, query and data management. The software standardizes the process and judgment factors of default value analysis process, makes qualitative and quantitative judgment on value selection factors, and combines the value selection algorithm embedded in the software, The auxiliary designer can automatically get the default value. The use of software can effectively improve the rationality and standardization of default value analysis. The software platform solves the problems of different default value analysis standards and chaotic data management in nuclear power plants. As the first special software for signal default value analysis and management of nuclear power control system in China, it has improved the intelligent level of the third generation nuclear power HPR1000.
{"title":"Research and Development of Default Value Analyses and Management Platform for Nuclear Power Plant Control System","authors":"Hongbin Zhao, Zhiyong Liu, Hemin Liu, Weifang Liu","doi":"10.1115/icone29-93009","DOIUrl":"https://doi.org/10.1115/icone29-93009","url":null,"abstract":"\u0000 The research and development of default value development and management platform of nuclear power plant control system can realize the functions of default value analysis, value selection, query and data management. The software standardizes the process and judgment factors of default value analysis process, makes qualitative and quantitative judgment on value selection factors, and combines the value selection algorithm embedded in the software, The auxiliary designer can automatically get the default value. The use of software can effectively improve the rationality and standardization of default value analysis. The software platform solves the problems of different default value analysis standards and chaotic data management in nuclear power plants. As the first special software for signal default value analysis and management of nuclear power control system in China, it has improved the intelligent level of the third generation nuclear power HPR1000.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133265409","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}
Jing Xie, Yueming Fu, Xun Zhang, Hengdi Zhang, Jianfei Hou, Xiangjie He, Fei Yu, Run Lin
Industrial digital twins have become popular in China and around the world in various fields in the past a few years. The application of industrial digital twins in nuclear power plants is to be researched in this paper. This article mainly summarizes the concept of industrial digital twin and its application status; sets up an application system of industrial digital twin in nuclear power plants, including the overall framework and application mode; introduces some typical application of digital twins in nuclear power plants; explores the expectation, challenges of the application of digital twins in the development of future intelligent nuclear power plant.
{"title":"Application of Industrial Digital Twins in Nuclear Power Plant","authors":"Jing Xie, Yueming Fu, Xun Zhang, Hengdi Zhang, Jianfei Hou, Xiangjie He, Fei Yu, Run Lin","doi":"10.1115/icone29-92912","DOIUrl":"https://doi.org/10.1115/icone29-92912","url":null,"abstract":"\u0000 Industrial digital twins have become popular in China and around the world in various fields in the past a few years. The application of industrial digital twins in nuclear power plants is to be researched in this paper.\u0000 This article mainly summarizes the concept of industrial digital twin and its application status; sets up an application system of industrial digital twin in nuclear power plants, including the overall framework and application mode; introduces some typical application of digital twins in nuclear power plants; explores the expectation, challenges of the application of digital twins in the development of future intelligent nuclear power plant.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130227093","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}
The top nozzle is an important part of the fuel assembly and is located at the core coolant outlet, which has an important impact on the uniformity of the coolant flow distribution at the core outlet and the overall pressure drop of the fuel assembly[1]. The peak stress of the top nozzle under each working condition appears at the position of the adapter plate, top nozzle and the pressure drop of the top nozzle is closely related to the flow area, height and local characteristics of the adapter plate. In this paper, the overall structure of the adapter plate is optimized to reduce the pressure drop as much as possible on the premise of ensuring that the stress meets the requirements of the standard limit. The 3D modeling software NX is used to establish the geometric model of the top nozzle, and it is transferred to ANSYS to establish a finite element model for simulation optimization design. On this basis, the key dimension parameters of the adapter plate (such as the thickness of the rib, etc.) are selected as design variables, and the law of its influence on the blocking area and stress is studied, and the structure size is further optimized.
{"title":"Research on Structural Optimization Technology of Fuel Assembly Parts Based on Topology and Optimal Design","authors":"Huabin Fan, Shiyin Xu, Huili Jia","doi":"10.1115/icone29-93368","DOIUrl":"https://doi.org/10.1115/icone29-93368","url":null,"abstract":"\u0000 The top nozzle is an important part of the fuel assembly and is located at the core coolant outlet, which has an important impact on the uniformity of the coolant flow distribution at the core outlet and the overall pressure drop of the fuel assembly[1]. The peak stress of the top nozzle under each working condition appears at the position of the adapter plate, top nozzle and the pressure drop of the top nozzle is closely related to the flow area, height and local characteristics of the adapter plate. In this paper, the overall structure of the adapter plate is optimized to reduce the pressure drop as much as possible on the premise of ensuring that the stress meets the requirements of the standard limit. The 3D modeling software NX is used to establish the geometric model of the top nozzle, and it is transferred to ANSYS to establish a finite element model for simulation optimization design. On this basis, the key dimension parameters of the adapter plate (such as the thickness of the rib, etc.) are selected as design variables, and the law of its influence on the blocking area and stress is studied, and the structure size is further optimized.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115518043","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}
Bin Wang, Bingtao Hu, Zhifeng Zhang, Yixiong Feng, Jianrong Tan
The layout design of nuclear power reactor coolant system requires a large amount of knowledge that satisfies many disciplines, which will waste designers a lot of time retrieving relevant knowledge in the design process. In order to obtain the knowledge efficiently and accurately, a knowledge retrieval method based on hypernetwork and deep structured semantic model (DSSM) was proposed. The knowledge hypernetwork model consisted of a designer sub-network, a design task subnetwork, and a design knowledge resource sub-network. Nodes in each sub-network and between different sub-networks were connected through special edges, which represented correlation degree information. Then an improved DSSM model was used to evaluate relevance at the semantic level between user query information and knowledge elements in hypernetwork. Correlation scores will be obtained based on relevance at the semantic level, and knowledge elements with lower scores will be removed during the process. Finally, the Bayesian method was used to calculate the knowledge recommendation probability to obtain the most relevant knowledge retrieval results. The knowledge retrieval results were sorted from high to low according to the calculated probability. A case study conducted in this work showed that the proposed approach was effective in capturing relevance at the semantic level and supporting efficient and accurate knowledge retrieval services.
{"title":"A Knowledge Retrieval Method for Layout Design of Nuclear Power Reactor Coolant System Based on Hypernetwork and Deep Structured Semantic Model","authors":"Bin Wang, Bingtao Hu, Zhifeng Zhang, Yixiong Feng, Jianrong Tan","doi":"10.1115/icone29-91827","DOIUrl":"https://doi.org/10.1115/icone29-91827","url":null,"abstract":"\u0000 The layout design of nuclear power reactor coolant system requires a large amount of knowledge that satisfies many disciplines, which will waste designers a lot of time retrieving relevant knowledge in the design process. In order to obtain the knowledge efficiently and accurately, a knowledge retrieval method based on hypernetwork and deep structured semantic model (DSSM) was proposed. The knowledge hypernetwork model consisted of a designer sub-network, a design task subnetwork, and a design knowledge resource sub-network. Nodes in each sub-network and between different sub-networks were connected through special edges, which represented correlation degree information. Then an improved DSSM model was used to evaluate relevance at the semantic level between user query information and knowledge elements in hypernetwork. Correlation scores will be obtained based on relevance at the semantic level, and knowledge elements with lower scores will be removed during the process. Finally, the Bayesian method was used to calculate the knowledge recommendation probability to obtain the most relevant knowledge retrieval results. The knowledge retrieval results were sorted from high to low according to the calculated probability. A case study conducted in this work showed that the proposed approach was effective in capturing relevance at the semantic level and supporting efficient and accurate knowledge retrieval services.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132340512","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}
Yixiong Feng, Yong Wang, Bingtao Hu, Zhaoxi Hong, Jianrong Tan
Nuclear power is an indispensable part of modern energy systems. To operate the nuclear power plants safely and reliably, it is crucial to greatly develop the predictive maintenance of nuclear infrastructure with the support of various smart sensors and big data analytics. To this end, this paper proposes a novel collaborative edge computing-enabled solution for predictive maintenance in nuclear power plants, from which a key problem of efficiently allocating some edge computing tasks is formulated. Specifically, considering huge amounts of industrial data are continuously generated during plant operations, we first present a three-layer predictive maintenance computing framework for nuclear power plants. Subsequently, to timely process these data in some distributed and heterogeneous industrial computing nodes, a complicated scheduling optimization model with some interdependent computational tasks is established. To lower the size of model, we also introduce some reduction strategies. Finally, an actual predictive maintenance scenario in nuclear power plant is chosen and some algorithms are taken for comparisons.
{"title":"A Collaborative Edge Computing Framework for Predictive Maintenance in Nuclear Power Plants","authors":"Yixiong Feng, Yong Wang, Bingtao Hu, Zhaoxi Hong, Jianrong Tan","doi":"10.1115/icone29-93370","DOIUrl":"https://doi.org/10.1115/icone29-93370","url":null,"abstract":"\u0000 Nuclear power is an indispensable part of modern energy systems. To operate the nuclear power plants safely and reliably, it is crucial to greatly develop the predictive maintenance of nuclear infrastructure with the support of various smart sensors and big data analytics. To this end, this paper proposes a novel collaborative edge computing-enabled solution for predictive maintenance in nuclear power plants, from which a key problem of efficiently allocating some edge computing tasks is formulated. Specifically, considering huge amounts of industrial data are continuously generated during plant operations, we first present a three-layer predictive maintenance computing framework for nuclear power plants. Subsequently, to timely process these data in some distributed and heterogeneous industrial computing nodes, a complicated scheduling optimization model with some interdependent computational tasks is established. To lower the size of model, we also introduce some reduction strategies. Finally, an actual predictive maintenance scenario in nuclear power plant is chosen and some algorithms are taken for comparisons.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127290497","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}
At present, the ultra-high voltage transmission lines in nuclear power plants mostly use Gas Insulated transmission lines (GIL). This type of transmission line transmits large natural power, and GILs are mostly laid through corridors and themselves are metal-enclosed gas -insulated structures, which are affected by the natural environment and have low reliability. high. However, due to the long length of each independent air chamber of the GIL, when it fails, it is accurately difficult to monitor the fault and locate the fault, which reduces the availability of the unit and brings about greater economic losses. In this paper, through the simulation test of the vibration signal propagation when the GIL fails, the signal attenuation law of the vibration signal passing through the insulator is tested, and the fault cell is located through the signal analysis of the vibration sensors in different directions. At the same time, through the simulation test research on the discarge faults of metal spikes, floating potential and metal particles in the GIL, after obtaining the partial discharge signal through the sensor, the corresponding partial discharge eigenvalues are extracted, and the eigenvalues are classified based on the SVM method to realize the GIL fault. Which can provide support for engineering practice.
{"title":"Research on GIL Fault Diagnosis and Location Method Based on Partial Discharge and Vibration Monitoring","authors":"Sen Liu","doi":"10.1115/icone29-92174","DOIUrl":"https://doi.org/10.1115/icone29-92174","url":null,"abstract":"\u0000 At present, the ultra-high voltage transmission lines in nuclear power plants mostly use Gas Insulated transmission lines (GIL). This type of transmission line transmits large natural power, and GILs are mostly laid through corridors and themselves are metal-enclosed gas -insulated structures, which are affected by the natural environment and have low reliability. high. However, due to the long length of each independent air chamber of the GIL, when it fails, it is accurately difficult to monitor the fault and locate the fault, which reduces the availability of the unit and brings about greater economic losses. In this paper, through the simulation test of the vibration signal propagation when the GIL fails, the signal attenuation law of the vibration signal passing through the insulator is tested, and the fault cell is located through the signal analysis of the vibration sensors in different directions. At the same time, through the simulation test research on the discarge faults of metal spikes, floating potential and metal particles in the GIL, after obtaining the partial discharge signal through the sensor, the corresponding partial discharge eigenvalues are extracted, and the eigenvalues are classified based on the SVM method to realize the GIL fault. Which can provide support for engineering practice.","PeriodicalId":422334,"journal":{"name":"Volume 12: Innovative and Smart Nuclear Power Plant Design","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132627635","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：