Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy Internet
In order to reduce the cost and improve efficiency of power line inspection, UAV (unmanned aerial vehicle), which use mobile edge computing technology to access and process service data, are used to inspect power lines in the energy internet.However, due to the dynamic changes of UAV data transmissi...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2021-05-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2021071/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841539181215481856 |
|---|---|
| author | Siya XU Yifei XING Shaoyong GUO Chao YANG Xuesong QIU Luoming MENG |
| author_facet | Siya XU Yifei XING Shaoyong GUO Chao YANG Xuesong QIU Luoming MENG |
| author_sort | Siya XU |
| collection | DOAJ |
| description | In order to reduce the cost and improve efficiency of power line inspection, UAV (unmanned aerial vehicle), which use mobile edge computing technology to access and process service data, are used to inspect power lines in the energy internet.However, due to the dynamic changes of UAV data transmission demand and geographical location, the edge server load will be unbalanced, which causes higher service processing delay and network energy consumption.Thus, an intelligent inspection task allocation mechanism for energy internet based on deep reinforcement learning was proposed.First, a two-layer edge network task offloading model was established to archive joint optimization of multi-objectives, such as delay and energy consumption.It was designed by comprehensively considering the route of UAV and edge nodes, different demands of services and limited service capabilities of edge nodes.Furthermore, based on Lyapunov optimization theory and dual-time-scaled mechanism, proximal policy optimization algorithm based deep reinforcement learning was used to solve the connection relationship and offloading strategy of edge servers between fixed edge sink layer and mobile edge access layer.The simulation results show that, the proposed mechanism can reduce the service request delay and system energy consumption while ensuring the stability of system. |
| format | Article |
| id | doaj-art-300bd0da545d4a978c376a60e0284265 |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2021-05-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-300bd0da545d4a978c376a60e02842652025-01-14T07:24:08ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2021-05-014219120459835480Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy InternetSiya XUYifei XINGShaoyong GUOChao YANGXuesong QIULuoming MENGIn order to reduce the cost and improve efficiency of power line inspection, UAV (unmanned aerial vehicle), which use mobile edge computing technology to access and process service data, are used to inspect power lines in the energy internet.However, due to the dynamic changes of UAV data transmission demand and geographical location, the edge server load will be unbalanced, which causes higher service processing delay and network energy consumption.Thus, an intelligent inspection task allocation mechanism for energy internet based on deep reinforcement learning was proposed.First, a two-layer edge network task offloading model was established to archive joint optimization of multi-objectives, such as delay and energy consumption.It was designed by comprehensively considering the route of UAV and edge nodes, different demands of services and limited service capabilities of edge nodes.Furthermore, based on Lyapunov optimization theory and dual-time-scaled mechanism, proximal policy optimization algorithm based deep reinforcement learning was used to solve the connection relationship and offloading strategy of edge servers between fixed edge sink layer and mobile edge access layer.The simulation results show that, the proposed mechanism can reduce the service request delay and system energy consumption while ensuring the stability of system.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2021071/patrol UAVask offloadingproximal policy optimizationLyapunov optimizationartificial intelligence |
| spellingShingle | Siya XU Yifei XING Shaoyong GUO Chao YANG Xuesong QIU Luoming MENG Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy Internet Tongxin xuebao patrol UAV ask offloading proximal policy optimization Lyapunov optimization artificial intelligence |
| title | Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy Internet |
| title_full | Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy Internet |
| title_fullStr | Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy Internet |
| title_full_unstemmed | Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy Internet |
| title_short | Deep reinforcement learning based task allocation mechanism for intelligent inspection in energy Internet |
| title_sort | deep reinforcement learning based task allocation mechanism for intelligent inspection in energy internet |
| topic | patrol UAV ask offloading proximal policy optimization Lyapunov optimization artificial intelligence |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2021071/ |
| work_keys_str_mv | AT siyaxu deepreinforcementlearningbasedtaskallocationmechanismforintelligentinspectioninenergyinternet AT yifeixing deepreinforcementlearningbasedtaskallocationmechanismforintelligentinspectioninenergyinternet AT shaoyongguo deepreinforcementlearningbasedtaskallocationmechanismforintelligentinspectioninenergyinternet AT chaoyang deepreinforcementlearningbasedtaskallocationmechanismforintelligentinspectioninenergyinternet AT xuesongqiu deepreinforcementlearningbasedtaskallocationmechanismforintelligentinspectioninenergyinternet AT luomingmeng deepreinforcementlearningbasedtaskallocationmechanismforintelligentinspectioninenergyinternet |