Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics
Abstract As more renewable energy generators and adjustable loads such as electric vehicles are being connected to the power grids, load modelling of the distribution network becomes more complicated. Therefore, this paper explores a dynamic equivalent modelling method for active distribution networ...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-12-01
|
| Series: | IET Generation, Transmission & Distribution |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/gtd2.13344 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849304114188517376 |
|---|---|
| author | Jingwen Wang Jiehui Zheng Zhigang Li Qing‐Hua Wu |
| author_facet | Jingwen Wang Jiehui Zheng Zhigang Li Qing‐Hua Wu |
| author_sort | Jingwen Wang |
| collection | DOAJ |
| description | Abstract As more renewable energy generators and adjustable loads such as electric vehicles are being connected to the power grids, load modelling of the distribution network becomes more complicated. Therefore, this paper explores a dynamic equivalent modelling method for active distribution network that takes into account electric vehicle charging. First of all the combination of integrated ZIP loads and motors is adopted as an equivalent model for active distribution networks. Subsequently, a four‐layer, tri‐stage deep reinforcement learning approach is used to solve the relevant key parameters of the proposed equivalent model. The method proposed in this paper fully utilizes the superiority of reinforcement learning in decision making, while the method combines the excellent feature extraction capability of deep learning. The method utilizes measurements obtained at boundary nodes to obtain an active distributed network equivalent model after a series of calculations. At the same time, adjustable loads are identified in detail. On the other hand, this method introduces a prioritized empirical playback mechanism, log‐cosh loss function, and adaptive operator to improve the computational efficiency of the method. From the simulation results, the present method is effective. |
| format | Article |
| id | doaj-art-4904f777a55e4d86a73151c3da19bc05 |
| institution | Kabale University |
| issn | 1751-8687 1751-8695 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Generation, Transmission & Distribution |
| spelling | doaj-art-4904f777a55e4d86a73151c3da19bc052025-08-20T03:55:49ZengWileyIET Generation, Transmission & Distribution1751-86871751-86952024-12-0118244154416710.1049/gtd2.13344Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristicsJingwen Wang0Jiehui Zheng1Zhigang Li2Qing‐Hua Wu3School of Electric Power EngineeringSouth China University of TechnologyGuangzhouGuangdongChinaSchool of Electric Power EngineeringSouth China University of TechnologyGuangzhouGuangdongChinaSchool of Electric Power EngineeringSouth China University of TechnologyGuangzhouGuangdongChinaSchool of Electric Power EngineeringSouth China University of TechnologyGuangzhouGuangdongChinaAbstract As more renewable energy generators and adjustable loads such as electric vehicles are being connected to the power grids, load modelling of the distribution network becomes more complicated. Therefore, this paper explores a dynamic equivalent modelling method for active distribution network that takes into account electric vehicle charging. First of all the combination of integrated ZIP loads and motors is adopted as an equivalent model for active distribution networks. Subsequently, a four‐layer, tri‐stage deep reinforcement learning approach is used to solve the relevant key parameters of the proposed equivalent model. The method proposed in this paper fully utilizes the superiority of reinforcement learning in decision making, while the method combines the excellent feature extraction capability of deep learning. The method utilizes measurements obtained at boundary nodes to obtain an active distributed network equivalent model after a series of calculations. At the same time, adjustable loads are identified in detail. On the other hand, this method introduces a prioritized empirical playback mechanism, log‐cosh loss function, and adaptive operator to improve the computational efficiency of the method. From the simulation results, the present method is effective.https://doi.org/10.1049/gtd2.13344demand side managementenergy resourcesoptimisationpower distributionpower system management |
| spellingShingle | Jingwen Wang Jiehui Zheng Zhigang Li Qing‐Hua Wu Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics IET Generation, Transmission & Distribution demand side management energy resources optimisation power distribution power system management |
| title | Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics |
| title_full | Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics |
| title_fullStr | Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics |
| title_full_unstemmed | Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics |
| title_short | Dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics |
| title_sort | dynamic equivalent modelling for active distributed network considering adjustable loads charging characteristics |
| topic | demand side management energy resources optimisation power distribution power system management |
| url | https://doi.org/10.1049/gtd2.13344 |
| work_keys_str_mv | AT jingwenwang dynamicequivalentmodellingforactivedistributednetworkconsideringadjustableloadschargingcharacteristics AT jiehuizheng dynamicequivalentmodellingforactivedistributednetworkconsideringadjustableloadschargingcharacteristics AT zhigangli dynamicequivalentmodellingforactivedistributednetworkconsideringadjustableloadschargingcharacteristics AT qinghuawu dynamicequivalentmodellingforactivedistributednetworkconsideringadjustableloadschargingcharacteristics |