In terms of traditional oil-immersed transformer fault diagnosis and location, major works focus on data feature selection and classifier optimization currently. They are studied as two independent directions due to the difference of solving method. In this paper, ergodic-MAEPSO (EMAEPSO) is proposed, which inherits the ability of classifier parameter optimization from PSO, and by introducing ergodic comparison into Multi-scale Cooperative Mutation Self-adaptive Escape PSO (MAEPSO) to realize feature selection. Based on EMAEPSO, the idea of Multi-birth Optimization by merging two different scale problems simultaneously, feature selection and classifier parameter optimization, is presented to improve the accuracy of transformer fault diagnosis and location. Additionally, considering the scarcity of the fault dataset in some cases, the Random Seed of SMOTE is included into the Multi-birth Optimization for further improvement of diagnostic model. To this end, for the purpose of verifying the generalization and reliability of the idea of Multi-birth Optimization, different types of classifiers are carried out for comparison. Experimental results show that the model optimized by Multi-birth Optimization based on EMAEPSO has a higher diagnostic accuracy, no matter which type of classifier is involved.
{"title":"Multi-birth Optimization Based on Ergodic Multi-scale Cooperative Mutation Self-Adaptive Escape PSO for Transformer Fault Diagnosis and Location","authors":"Weiming Zheng, Chenchen Zhao, Guogang Zhang, Qianqian Zhu, Mingming Yang, Yingsan Geng","doi":"10.1109/AEEES56888.2023.10114235","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114235","url":null,"abstract":"In terms of traditional oil-immersed transformer fault diagnosis and location, major works focus on data feature selection and classifier optimization currently. They are studied as two independent directions due to the difference of solving method. In this paper, ergodic-MAEPSO (EMAEPSO) is proposed, which inherits the ability of classifier parameter optimization from PSO, and by introducing ergodic comparison into Multi-scale Cooperative Mutation Self-adaptive Escape PSO (MAEPSO) to realize feature selection. Based on EMAEPSO, the idea of Multi-birth Optimization by merging two different scale problems simultaneously, feature selection and classifier parameter optimization, is presented to improve the accuracy of transformer fault diagnosis and location. Additionally, considering the scarcity of the fault dataset in some cases, the Random Seed of SMOTE is included into the Multi-birth Optimization for further improvement of diagnostic model. To this end, for the purpose of verifying the generalization and reliability of the idea of Multi-birth Optimization, different types of classifiers are carried out for comparison. Experimental results show that the model optimized by Multi-birth Optimization based on EMAEPSO has a higher diagnostic accuracy, no matter which type of classifier is involved.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128331620","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114250
Lu Chen, D. Wang, Jie Ma, Yongliang Zhao, Weixiong Chen
Operational flexibility of the coal-fired power plants must be improved so as to deal with the unpredictability of renewable energy in the future. In this study, a 1100 MW supercritical coal-fired power plant was selected, and dynamic simulation model of the unit was established via GSE software. Moreover, control model was added. Dynamic response characteristics of key thermal parameters under different power ramp rates in the region of 30%-100% THA was obtained. The cumulative standard coal consumption rate and integral absolute error were selected as the economic evaluation index of the thermal system. The results show that the maximum power ramp rate of the unit in each region during loading up is smaller than that during loading down. In the loading down process, the standard coal consumption rate of the unit firstly decreases with the reduction of output power. When the load of the unit drops to the target load, the standard coal consumption rate rises to a stabler stage. The process of loading up is the contrary. The cumulative standard coal consumption rate of coal-fired power plant is the lowest, only 275 g•(kW•h)-1 during the loading down process of 100%-75% THA. However, the cumulative standard coal consumption rate is the highest, reaching 311 g(•kW•h)-1 during the loading up process of 30%-50% THA. Due to the output power of the unit fluctuating greatly, the integral absolute error of the unit is much higher than that of other load cycling transient processes during the loading up process of 50%-75% THA and 30%-50% THA. It can be concluded that the loading up process of the unit is more difficult to control stably. This study can provide sufficient theoretical and data guidance for improving the operational flexibility of the coal-fired power plants.
{"title":"Operational Flexibility Analysis of 1100 MW Supercritical Coal-Fired Power Plants during Load Cycling Transient Processes","authors":"Lu Chen, D. Wang, Jie Ma, Yongliang Zhao, Weixiong Chen","doi":"10.1109/AEEES56888.2023.10114250","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114250","url":null,"abstract":"Operational flexibility of the coal-fired power plants must be improved so as to deal with the unpredictability of renewable energy in the future. In this study, a 1100 MW supercritical coal-fired power plant was selected, and dynamic simulation model of the unit was established via GSE software. Moreover, control model was added. Dynamic response characteristics of key thermal parameters under different power ramp rates in the region of 30%-100% THA was obtained. The cumulative standard coal consumption rate and integral absolute error were selected as the economic evaluation index of the thermal system. The results show that the maximum power ramp rate of the unit in each region during loading up is smaller than that during loading down. In the loading down process, the standard coal consumption rate of the unit firstly decreases with the reduction of output power. When the load of the unit drops to the target load, the standard coal consumption rate rises to a stabler stage. The process of loading up is the contrary. The cumulative standard coal consumption rate of coal-fired power plant is the lowest, only 275 g•(kW•h)-1 during the loading down process of 100%-75% THA. However, the cumulative standard coal consumption rate is the highest, reaching 311 g(•kW•h)-1 during the loading up process of 30%-50% THA. Due to the output power of the unit fluctuating greatly, the integral absolute error of the unit is much higher than that of other load cycling transient processes during the loading up process of 50%-75% THA and 30%-50% THA. It can be concluded that the loading up process of the unit is more difficult to control stably. This study can provide sufficient theoretical and data guidance for improving the operational flexibility of the coal-fired power plants.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128598143","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}
With the development of power IoT, all kinds of sensing devices in the sensing layer have increased, and a large amount of time-series data is collected every moment. However, the data will inevitably be abnormal due to the external environment or equipment, etc. To ensure that the anomalous data collected by the sensing terminal in the power IoT can be detected, a BiLSTM-based anomaly detection model for time series data of the sensing terminal in the power IoT is proposed. Firstly, the Bi-LSTM can capture bi-directional timing information to build a prediction model. Secondly, multiple thresholds are set up, and the predicted value and the data collected by the sensing terminal are calculated as residuals and then compared with multiple thresholds, and the majority result is taken to determine whether the data is abnormal or not, avoiding the misjudgment of a single threshold.
{"title":"A BiLSTM-Based Method for Detecting Time Series Data Anomalies in Power IoT Sensing Terminals","authors":"Yiying Zhang, Lei Zhang, Hao Wang, Yeshen He, Xueliang Wang, Shengpeng Zhang","doi":"10.1109/AEEES56888.2023.10114073","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114073","url":null,"abstract":"With the development of power IoT, all kinds of sensing devices in the sensing layer have increased, and a large amount of time-series data is collected every moment. However, the data will inevitably be abnormal due to the external environment or equipment, etc. To ensure that the anomalous data collected by the sensing terminal in the power IoT can be detected, a BiLSTM-based anomaly detection model for time series data of the sensing terminal in the power IoT is proposed. Firstly, the Bi-LSTM can capture bi-directional timing information to build a prediction model. Secondly, multiple thresholds are set up, and the predicted value and the data collected by the sensing terminal are calculated as residuals and then compared with multiple thresholds, and the majority result is taken to determine whether the data is abnormal or not, avoiding the misjudgment of a single threshold.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128994098","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114298
Zi-Na Huang, Shuning Pan, S. Bu, S. Niu
This paper presents an optimal communication topology restoration method under denial-of-service (DoS) attacks. Firstly, a general hierarchical control framework for AC microgrids, including primary, secondary, and tertiary control, is introduced. Secondly, the attack effect is investigated by considering a non-ideal communication network where DoS attacks occur among the information exchanges of secondary control. Then, to eliminate the DoS attacks effect, an optimal communication topology restoration method is proposed. Notably, two indexes, including communication cost and convergence speed, are considered, and the optimal problem is formulated as a mixed integer semidefinite program. Lastly, the effectiveness of the proposed method is validated by time-domain simulations.
{"title":"Optimal Communication Topology Restoration for Islanded AC Microgrids under Denial-of-Service Attacks","authors":"Zi-Na Huang, Shuning Pan, S. Bu, S. Niu","doi":"10.1109/AEEES56888.2023.10114298","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114298","url":null,"abstract":"This paper presents an optimal communication topology restoration method under denial-of-service (DoS) attacks. Firstly, a general hierarchical control framework for AC microgrids, including primary, secondary, and tertiary control, is introduced. Secondly, the attack effect is investigated by considering a non-ideal communication network where DoS attacks occur among the information exchanges of secondary control. Then, to eliminate the DoS attacks effect, an optimal communication topology restoration method is proposed. Notably, two indexes, including communication cost and convergence speed, are considered, and the optimal problem is formulated as a mixed integer semidefinite program. Lastly, the effectiveness of the proposed method is validated by time-domain simulations.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129251217","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114082
Xue Hongtao, Z. Jianying, Du Yingqian, Yao Yongqi, Wang Zhijun, Liu Chaofeng, Leng Longmao, Wang Xiaolei
In this paper, the measured value of the surface vibration acceleration of the 750kV GIS bus cylinder was analyzed, the excitation principle of the electric field force on the bus was studied, the multi physical field coupling simulation model of bus was established, and the accuracy of the vibration acceleration simulation value was analyzed; The simplified equivalent circuit model of transformer and bus in power station was established, and the influence of circuit resonance on structure vibration and noise is analyzed.
{"title":"Vibration Analysis of 750kV GIS Power Station Bus","authors":"Xue Hongtao, Z. Jianying, Du Yingqian, Yao Yongqi, Wang Zhijun, Liu Chaofeng, Leng Longmao, Wang Xiaolei","doi":"10.1109/AEEES56888.2023.10114082","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114082","url":null,"abstract":"In this paper, the measured value of the surface vibration acceleration of the 750kV GIS bus cylinder was analyzed, the excitation principle of the electric field force on the bus was studied, the multi physical field coupling simulation model of bus was established, and the accuracy of the vibration acceleration simulation value was analyzed; The simplified equivalent circuit model of transformer and bus in power station was established, and the influence of circuit resonance on structure vibration and noise is analyzed.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123885573","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114319
Qin-ye Yu, Wei Xu, J. Lv, Y. Wang, Kaifeng Zhang
Microgrid provides an effective way to integrate renewable energy into power grid. However, the uncertainty of renewable energy and load demand bring great challenges to the energy management of microgrid. Therefore, this paper proposes a double-layer optimization method based on deep reinforcement learning (DRL) to solve this problem. The upper DRL agent takes Soft actor-critic algorithm to fully explore the regulation ability of the energy storage system. The lower nonlinear programming solver optimizes the output of other controllable equipment based on the output of the upper layer, and constantly revises the network parameters of the upper layer according to the optimization results. By combining DRL with traditional nonlinear programming, the convergence speed of the algorithm can be improved and the design difficulty of the DRL reward function can be reduced. Case studies show that the double-layer collaborative optimization method can provide real-time highquality solutions for energy management of the microgrid only based on the immediate information of the microgrid and can effectively accelerate the convergence speed of the model.
{"title":"Deep Reinforcement Learning Based Double-layer Optimization Method for Energy Management of Microgrid","authors":"Qin-ye Yu, Wei Xu, J. Lv, Y. Wang, Kaifeng Zhang","doi":"10.1109/AEEES56888.2023.10114319","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114319","url":null,"abstract":"Microgrid provides an effective way to integrate renewable energy into power grid. However, the uncertainty of renewable energy and load demand bring great challenges to the energy management of microgrid. Therefore, this paper proposes a double-layer optimization method based on deep reinforcement learning (DRL) to solve this problem. The upper DRL agent takes Soft actor-critic algorithm to fully explore the regulation ability of the energy storage system. The lower nonlinear programming solver optimizes the output of other controllable equipment based on the output of the upper layer, and constantly revises the network parameters of the upper layer according to the optimization results. By combining DRL with traditional nonlinear programming, the convergence speed of the algorithm can be improved and the design difficulty of the DRL reward function can be reduced. Case studies show that the double-layer collaborative optimization method can provide real-time highquality solutions for energy management of the microgrid only based on the immediate information of the microgrid and can effectively accelerate the convergence speed of the model.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127722437","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114311
Weile Kong, Hongxing Ye, Nan Wei, Dong Xing, Siwei Liu, Wei Chen
The HVDC tie-line has been successfully transfer- ring the renewable from rural areas to load centers. In the meantime, it is able to provide inter-regional flexibilities as well when flexible resources are located at different regions, maximizing the utilization of flexibility. This paper proposes a two-stage model for the medium-term flexibility planning, which considers the HVDC tie-line schedule, storage deployment, and unit retrofit. In the first stage, storage capacity, fossil-fired unit flexibility retrofit, and scheduled power for tie-line are determined. In the second stage, uncertainties are modeled and accommodated by inter-regional storages and fossil-fired units. Chance-constrained load shedding and renewable curtailment are modeled, avoiding the redundancy of energy storage investment. Simulation results show the efficiency and effectiveness of the proposed approach on promoting the inter-regional renewable accommodation.
{"title":"Optimization of Inter-Regional Flexible Resources for Renewable Accommodation","authors":"Weile Kong, Hongxing Ye, Nan Wei, Dong Xing, Siwei Liu, Wei Chen","doi":"10.1109/AEEES56888.2023.10114311","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114311","url":null,"abstract":"The HVDC tie-line has been successfully transfer- ring the renewable from rural areas to load centers. In the meantime, it is able to provide inter-regional flexibilities as well when flexible resources are located at different regions, maximizing the utilization of flexibility. This paper proposes a two-stage model for the medium-term flexibility planning, which considers the HVDC tie-line schedule, storage deployment, and unit retrofit. In the first stage, storage capacity, fossil-fired unit flexibility retrofit, and scheduled power for tie-line are determined. In the second stage, uncertainties are modeled and accommodated by inter-regional storages and fossil-fired units. Chance-constrained load shedding and renewable curtailment are modeled, avoiding the redundancy of energy storage investment. Simulation results show the efficiency and effectiveness of the proposed approach on promoting the inter-regional renewable accommodation.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126259866","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114257
Shi-da Zheng, Rongwu Zhu, Xiaoxiao Qi
With the increasing penetration of power electronics inverter (PEI)-interfaced renewable energy sources(RESs), the inertia of the power system is reduced, consequently resulting in the future electricity grid experiencing stability issues, due to the PEI-interfaced RESs working as a constant current/power source. To increase the system inertia, the grid-forming (GFM) operation of PEIs, which can emulate the conventional synchronous generator behavior in terms of inertia and damping support, is used instead of the constant current/power mode. However, the parallel operation of GFM inverters results in interactive oscillation issues in low and high-frequency bands, degrading the grid performances. The low-frequency interaction is caused by the power control loop, while the high-frequency interaction is caused by the voltage and current control loop. This paper models and analyzes the coupling mechanism in the high frequency of multiple-parallel GFM inverters based on their equivalent impedance models, and studies the stability based on the impedance stability criterion. The correctness and accuracy of theoretical analyses are clearly verified by the simulation results carried out in MATLAB/Simulink.
{"title":"Coupling Mechanism and Stability Analysis of Parallel Grid-Forming Inverters in High-Frequency Band","authors":"Shi-da Zheng, Rongwu Zhu, Xiaoxiao Qi","doi":"10.1109/AEEES56888.2023.10114257","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114257","url":null,"abstract":"With the increasing penetration of power electronics inverter (PEI)-interfaced renewable energy sources(RESs), the inertia of the power system is reduced, consequently resulting in the future electricity grid experiencing stability issues, due to the PEI-interfaced RESs working as a constant current/power source. To increase the system inertia, the grid-forming (GFM) operation of PEIs, which can emulate the conventional synchronous generator behavior in terms of inertia and damping support, is used instead of the constant current/power mode. However, the parallel operation of GFM inverters results in interactive oscillation issues in low and high-frequency bands, degrading the grid performances. The low-frequency interaction is caused by the power control loop, while the high-frequency interaction is caused by the voltage and current control loop. This paper models and analyzes the coupling mechanism in the high frequency of multiple-parallel GFM inverters based on their equivalent impedance models, and studies the stability based on the impedance stability criterion. The correctness and accuracy of theoretical analyses are clearly verified by the simulation results carried out in MATLAB/Simulink.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126402541","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114229
Jian Kang, Yuewei Xu, Bo Ding, Mukun Li, Wei Tang
Aiming at the problem that traditional power flow coordination and optimization methods are difficult to apply to the situation that a large number of Distributed Generations (DG) are connected and can′t effectively control power flow, a power flow Coordination and Optimization Control(COC) method based on Deep Reinforcement Learning (DRL) for Power Grid (PG) with DGs is proposed. Firstly, the influence of DG grid connection on the Distribution Network node voltage distribution is analyzed, and the JFNG algorithm is used to calculate the distributed power flow considering the connection of DG. Then, by introducing the DRL algorithm DQN into the COC of power flow with DG, a power flow COC strategy based on DRL is proposed. Finally, the proposed method is compared with the other two methods under the same conditions through simulation experiments. The results show that the average optimization success rate of the proposed method is the highest, reaching 95.64%, and the voltage deviation of each node of the Distribution Network is the smallest, with the amplitude of 1.032. The overall time consumption and maximum frequency fluctuation are also the lowest, which are 2.33s and 0.002Hz respectively. The algorithm performance is better than the other two comparison algorithms.
{"title":"Power Flow Coordination Optimization Control Method for Power System with DG Based on DRL","authors":"Jian Kang, Yuewei Xu, Bo Ding, Mukun Li, Wei Tang","doi":"10.1109/AEEES56888.2023.10114229","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114229","url":null,"abstract":"Aiming at the problem that traditional power flow coordination and optimization methods are difficult to apply to the situation that a large number of Distributed Generations (DG) are connected and can′t effectively control power flow, a power flow Coordination and Optimization Control(COC) method based on Deep Reinforcement Learning (DRL) for Power Grid (PG) with DGs is proposed. Firstly, the influence of DG grid connection on the Distribution Network node voltage distribution is analyzed, and the JFNG algorithm is used to calculate the distributed power flow considering the connection of DG. Then, by introducing the DRL algorithm DQN into the COC of power flow with DG, a power flow COC strategy based on DRL is proposed. Finally, the proposed method is compared with the other two methods under the same conditions through simulation experiments. The results show that the average optimization success rate of the proposed method is the highest, reaching 95.64%, and the voltage deviation of each node of the Distribution Network is the smallest, with the amplitude of 1.032. The overall time consumption and maximum frequency fluctuation are also the lowest, which are 2.33s and 0.002Hz respectively. The algorithm performance is better than the other two comparison algorithms.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128176562","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 : 2023-03-23DOI: 10.1109/AEEES56888.2023.10114316
Shan Li, Yangjun Zhou, Yubo Zhang, Rongrong Wu, Jie Tang
Accurate load forecasting can help the power sector to formulate a reasonable power generation scheme, which can ensure the reliability of power supply while minimizing resource waste. However, most of the existing prediction methods based on deep learning only regard the minimum loss function of the training dataset under laboratory conditions as the optimal model, resulting in low generalization of the model and poor performance of the model in solving practical engineering problems with universality. To solve the above problems, this paper proposes a load forecasting model based on multi loss function collaborative optimization, considering the constraint relationship between the variance, deviation and model generalization error of forecasting results. Considering the different physical meanings of different loss functions, the model calculates the weighted sum of multiple loss functions, and then optimizes the weight value of combined loss functions by using genetic algorithm. The results show that the prediction error of combined loss function is smaller than that of single loss function under the premise of selecting appropriate weight parameters.
{"title":"Load Forecasting Method based on Multi Loss Function Collaborative Optimization","authors":"Shan Li, Yangjun Zhou, Yubo Zhang, Rongrong Wu, Jie Tang","doi":"10.1109/AEEES56888.2023.10114316","DOIUrl":"https://doi.org/10.1109/AEEES56888.2023.10114316","url":null,"abstract":"Accurate load forecasting can help the power sector to formulate a reasonable power generation scheme, which can ensure the reliability of power supply while minimizing resource waste. However, most of the existing prediction methods based on deep learning only regard the minimum loss function of the training dataset under laboratory conditions as the optimal model, resulting in low generalization of the model and poor performance of the model in solving practical engineering problems with universality. To solve the above problems, this paper proposes a load forecasting model based on multi loss function collaborative optimization, considering the constraint relationship between the variance, deviation and model generalization error of forecasting results. Considering the different physical meanings of different loss functions, the model calculates the weighted sum of multiple loss functions, and then optimizes the weight value of combined loss functions by using genetic algorithm. The results show that the prediction error of combined loss function is smaller than that of single loss function under the premise of selecting appropriate weight parameters.","PeriodicalId":272114,"journal":{"name":"2023 5th Asia Energy and Electrical Engineering Symposium (AEEES)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132369813","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：