Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction model
Power system dynamic state estimation plays an important role. However, rapid changes in states cause state estimation to become very hard. To reduce the residual between pseudo and real measurement, prediction models are adopted, which are strongly associated with the convergence rates and accuraci...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-11-01
|
| Series: | Energy Reports |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352484722020492 |
| _version_ | 1797952159252217856 |
|---|---|
| author | Yanjie Yu Qiang Li Chuchu Chen Xinze Zheng Yingjie Tan |
| author_facet | Yanjie Yu Qiang Li Chuchu Chen Xinze Zheng Yingjie Tan |
| author_sort | Yanjie Yu |
| collection | DOAJ |
| description | Power system dynamic state estimation plays an important role. However, rapid changes in states cause state estimation to become very hard. To reduce the residual between pseudo and real measurement, prediction models are adopted, which are strongly associated with the convergence rates and accuracies of estimation methods. In this paper, to improve the estimation accuracy, a prediction model that consists of the convolutional neural network and long short-term memory (CNN-LSTM) is employed and then integrated into the unscented Kalman filter (UKF). In the proposed UKF with CNN-LSTM, state vectors are considered as time-series data, so CNN performs feature extraction for data pre-processing first, and then the features go through LSTM to improve its forecast accuracy in real-time. Next, online training and error normalization are introduced to UKF, which increases the estimation accuracy effectively. Finally, simulations are carried out on the IEEE 33-bus system. Simulation results show that the accuracies of the CNN-LSTM prediction model are much higher than those of conventional methods. Furthermore, compared to widely used state estimation methods, our method decreases RMSE and MAPE by about 2 multiples. |
| first_indexed | 2024-04-10T22:41:55Z |
| format | Article |
| id | doaj.art-977f4a689d794afea736042e38471609 |
| institution | Directory Open Access Journal |
| issn | 2352-4847 |
| language | English |
| last_indexed | 2024-04-10T22:41:55Z |
| publishDate | 2022-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Energy Reports |
| spelling | doaj.art-977f4a689d794afea736042e384716092023-01-16T04:08:25ZengElsevierEnergy Reports2352-48472022-11-018364376Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction modelYanjie Yu0Qiang Li1Chuchu Chen2Xinze Zheng3Yingjie Tan4School of Electric Engineering, Chongqing University, Chongqing, 400030, ChinaSchool of Electric Engineering, Chongqing University, Chongqing, 400030, China; Corresponding author.School of Electric Engineering, Chongqing University, Chongqing, 400030, ChinaSchool of Electric Engineering, Chongqing University, Chongqing, 400030, ChinaElectric Power Research Institute, China Southern Power Grid Co., Ltd, Guangzhou, 510000, ChinaPower system dynamic state estimation plays an important role. However, rapid changes in states cause state estimation to become very hard. To reduce the residual between pseudo and real measurement, prediction models are adopted, which are strongly associated with the convergence rates and accuracies of estimation methods. In this paper, to improve the estimation accuracy, a prediction model that consists of the convolutional neural network and long short-term memory (CNN-LSTM) is employed and then integrated into the unscented Kalman filter (UKF). In the proposed UKF with CNN-LSTM, state vectors are considered as time-series data, so CNN performs feature extraction for data pre-processing first, and then the features go through LSTM to improve its forecast accuracy in real-time. Next, online training and error normalization are introduced to UKF, which increases the estimation accuracy effectively. Finally, simulations are carried out on the IEEE 33-bus system. Simulation results show that the accuracies of the CNN-LSTM prediction model are much higher than those of conventional methods. Furthermore, compared to widely used state estimation methods, our method decreases RMSE and MAPE by about 2 multiples.http://www.sciencedirect.com/science/article/pii/S2352484722020492Smart gridDynamic state estimationLong short-term memoryUnscented Kalman filterPower system operation |
| spellingShingle | Yanjie Yu Qiang Li Chuchu Chen Xinze Zheng Yingjie Tan Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction model Energy Reports Smart grid Dynamic state estimation Long short-term memory Unscented Kalman filter Power system operation |
| title | Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction model |
| title_full | Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction model |
| title_fullStr | Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction model |
| title_full_unstemmed | Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction model |
| title_short | Improved dynamic state estimation of power system using unscented Kalman filter with more accurate prediction model |
| title_sort | improved dynamic state estimation of power system using unscented kalman filter with more accurate prediction model |
| topic | Smart grid Dynamic state estimation Long short-term memory Unscented Kalman filter Power system operation |
| url | http://www.sciencedirect.com/science/article/pii/S2352484722020492 |
| work_keys_str_mv | AT yanjieyu improveddynamicstateestimationofpowersystemusingunscentedkalmanfilterwithmoreaccuratepredictionmodel AT qiangli improveddynamicstateestimationofpowersystemusingunscentedkalmanfilterwithmoreaccuratepredictionmodel AT chuchuchen improveddynamicstateestimationofpowersystemusingunscentedkalmanfilterwithmoreaccuratepredictionmodel AT xinzezheng improveddynamicstateestimationofpowersystemusingunscentedkalmanfilterwithmoreaccuratepredictionmodel AT yingjietan improveddynamicstateestimationofpowersystemusingunscentedkalmanfilterwithmoreaccuratepredictionmodel |