North China Electric Power Research Institution Co., Ltd (State Grid Jibei Electric Power Research Institute), China
YW
Yaohan Wang
North China Electric Power Research Institution Co., Ltd (State Grid Jibei Electric Power Research Institute), China
ZW
Zhengyu Wang
North China Electric Power Research Institution Co., Ltd (State Grid Jibei Electric Power Research Institute), China
XW
Xiaoxiao Wang
North China Electric Power Research Institution Co., Ltd (State Grid Jibei Electric Power Research Institute), China
This paper proposes a short-term wind and photovoltaic power forecasting framework considering time-frequency decomposition based on bidirectional long short-term memory networks. First, the seasonal and trend decomposition using loess is applied to the original wind and photovoltaic data for time domain decomposition, obtaining trend, seasonal, and residual components. Then, the residual component undergoes variational mode decomposition to further extract features of different frequencies. Next, the maximum information coefficient is used to select features, which is highly correlated with wind and photovoltaic power as input features to the prediction model. Finally, the selected features are input into bidirectional long short-term memory networks for training and prediction. Experimental validation using actual data from a photovoltaic station and a wind power station in Hebei Province, China from July to August 2023, which shows that the proposed method achieves high accuracy and reliability in photovoltaic and wind power output prediction. The proposed time-frequency decomposition with the smallest root mean square error of 0.92 and mean absolute error of 0.58 in photovoltaic prediction, at the same time, the smallest root mean square error of 67.5 and mean absolute error of 48.16 in wind power prediction, significantly outperforming other power prediction methods.
State Grid Sichuan Electric Power Research Institute, China
The flexibility of power systems needs to be enhanced to address the volatility of renewable energy sources, and industrial loads on the demand side have significant potential to improve this flexibility. As a typical high-energy-consuming industrial load, electrolytic aluminum has advantages such as large individual capacity and strong power controllability. However, its participation in various types of grid interaction faces challenges such as unclear power control methods, impacts on normal production, and associated cost burdens. To address these issues, this paper analyzes the production process characteristics of electrolytic aluminum loads and establishes a power control model for these loads. It examines the effects of power regulation on normal production, explores the mechanisms behind regulation costs, and proposes a method for calculating these costs. Considering production safety constraints and power regulation constraints during grid interaction, the paper proposes a control strategy for electrolytic aluminum loads that incorporates process flow and regulation costs. In the case study, the above power control strategy is verified based on the production data of an electrolytic aluminum load. The results show that the above strategy can take into account the response requirements of the power grid and the safety requirements of the load production. A feasible scheme is proposed for the electrolytic aluminum industrial load to participate in the interactive control of the power grid.
Extreme weather causes an increase in power system failure. Previous studies on system resilience have often overlooked the user-side of a microgrid. This study proposes composite resilience indices (RI) based on the power supply of a large-scale manufacturing campus microgrid to quantify its ability to withstand extreme events. The proposed RI consider load priority, minimum supply load, total energy supplied, and the performance recovery-to-degradation slope ratio in an islanding microgrid. Accordingly, this study presents a three-stage resilience optimal dispatch and reconfiguration strategy, including energy-level scheduling, grid-level reconfiguration, and dynamic-level verification. A multi-objective optimization approach is used for energy scheduling, followed by system reconfiguration via DIgSILENT system modeling to meet the grid code and maximize load supply. Value at Risk methods are used to verify microgrid stability and load shedding requirements, supported by the virtual synchronous generator control of the energy storage system. The test results from a practical large-scale manufacturing campus microgrid validate the proposed approaches for enhancing system resilience considering load values.
The increasing penetration of distributed generations (e.g., solar power and wind power) causes unpredictable disturbances in power systems, which accelerates the application of intelligent control in automatic generation control (AGC), such as reinforcement learning (RL). However, traditional RL cannot ensure constraint safety during training and frequently violates constraint (e.g., frequency limitations), further threatening the grid operation's safety and stability. To address the safety issue, this paper proposes a novel safe RL framework to combine expert experiences with the RL controller, forming imitation-based RL. This method obtains an initialized safe policy by imitating the expert experience, which prevents random explorations at the beginning. Specifically, we first formulate the AGC problem as a Markov decision process in a mathematical way. Then, the imitation mechanism is developed on top of soft actor-critic RL algorithm. Finally, numerical studies on an IEEE 39-bus network show that, the proposed method can better satisfy the frequency control performance standard, with improving the RL training efficiency.
Economic and Technological Research Institute of State Grid Qinghai Electric Power Company, China
HW
Haiting Wang
Economic and Technological Research Institute of State Grid Qinghai Electric Power Company, China
XG
Xiaokan Gou
Economic and Technological Research Institute of State Grid Qinghai Electric Power Company, China
FL
Fei Liu
Economic and Technological Research Institute of State Grid Qinghai Electric Power Company, China
LL
Lixin Li
China Electric Power Research Institute, China
QW
Qian Wang
China Electric Power Research Institute, China
XZ
Xin Zhang
China Electric Power Research Institute, China
YL
Yuanyuan Li
China Electric Power Research Institute, China
Under the dual-carbon target, hydrogen energy, as a zero-carbon secondary energy source, has great scope for replacing fossil feedstocks in the fields of energy, transportation and industry. However, the current research on the competitiveness of hydrogen energy in various fields is not sufficiently addressed. In this paper, we use the LEAP model to predict the future scale of hydrogen use and the two-factor learning curve to predict the trend of hydrogen price change from 2025 to 2050, using Qinghai Province as the research background. At the same time, considering the carbon emission reduction benefits and raw material costs, the competitiveness of hydrogen energy in energy, transportation, industry and other fields in the future is compared. The results show that: (1) The hydrogen load scale in Qinghai Province will grow fast from 2025 to 2030. From 2030 to 2040, it slows under the steady and basic scenarios but remains high under the accelerated one. By 2040, the consumption scales are 1.057 million, 649,000 and 442,000 tons respectively. (2) The price of hydrogen energy will drop rapidly from the current 28 CNY/kg to about 20 CNY/kg in the next five years. By 2040, the price of hydrogen energy will be reduced to about 17 CNY/kg. (3) In terms of hydrogen energy competitiveness, when carbon emissions are not taken into account, hydrogen energy is currently competitive in the transportation field. During 2032-2038, it will be competitive in the field of methanol synthesis. By 2040, hydrogen energy will not be competitive in the fields of ammonia synthesis and power/heating. When considering carbon emissions, the competitiveness of hydrogen energy in the transportation field will become greater. The competitive year in the field of methanol synthesis will be 1-2 years ahead. By 2040, it will not be competitive in the field of synthetic ammonia and power/heating, but the gap will be significantly reduced due to the consideration of carbon emissions.
Electric Power Research Institute of Guangdong Power Grid Co., Ltd., China
JL
Jianming Liu
Guangdong Power Grid Co., Ltd., China
YL
Yuan La
China Southern Power Grid Co., Ltd., China
WS
Wenxing Sun
Guangdong Power Grid Co., Ltd., China
MS
Mo Shi
Electric Power Research Institute of Guangdong Power Grid Co., Ltd., China
To achieve load management optimization and timely failure warning for power transformers, as well as improve the reliability of the power network, this paper proposes a multiple time scale prediction method for top oil temperature (TOT) based on an adaptive extended Kalman filter (AEKF) algorithm. This method combines the Kalman filter (KF) algorithm and the D.Susa thermal model. The TOT, oil exponent and oil time constant are taken as state variables, while the ambient temperature and load current are used as input variables. The iterative optimization of the oil exponent and oil time constant is realized by comparing the estimated and observed TOT values. Moreover, the proposed method utilizes an adaptive noise estimator to correct the noise statistics parameters, which simplifies the initial noise setting and thus further improves the TOT prediction accuracy. A case study is conducted with two 110kV transformers. The results show that comparing the thermal equivalent circuit model and the extended KF algorithm, the proposed method has a higher accuracy in the intraday ultra-short-term prediction on a 15-minute time scale and day-ahead short-term prediction on a 24-hour time scale for the TOT.
With the increase in the number of microgrids in the same distribution area, they belong to dif-ferent subjects of interest, forming a multi-subject game pattern. Considering the interests of distribution networks and microgrids, a distribution network-multi-microgrid master-slave game model is established by selecting distribution networks as game masters and microgrids as game slaves. A master-slave game equilibrium algorithm based on Kriging metamodel is proposed. The method replaces the microgrid energy internal management model with a proposed Kriging metamodel. In the iterative optimization process, the particle swarm optimization algorithm is used to generate new sampling points and modify the model in a targeted way, so as to quickly and accurately obtain the transaction price and output plan of each microgrid. The algorithm does not need all the parameters of the microgrid, which not only achieves the purpose of protecting the privacy of the microgrid but also avoids a large number of calls to the lower optimization model, effectively reducing the amount of computation, and improves the efficiency of the solution. The results show that the overall operating costs of the microgrids are reduced by 1.4%, 4.6%, and 1.6%, respectively, which effectively balances the interests of multiple parties in the microgrid system; the revenue of the distribution network is increased by 50.6%.