Table of Content

28 December 2025, Volume 58 Issue 12

Previous Issue    Next Issue

Special Contribution

Interactions Between Global Energy Transition and Geopolitical Environment

WANG Yaohua, LI Xiaochun, YANG Yu, MIAO Zhongquan, MAO Jikang, JIAN Yongfang

2025, 58(12):  1-13.  DOI: 10.11930/j.issn.1004-9649.202509037

Asbtract ( )   PDF (4445KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Global energy transition has become a core issue in the sustainable development strategies of countries worldwide. Focusing on the interactions between global energy transition and the geopolitical, economic, and social dimensions, this study constructs a multidimensional indicator system for global energy transition and the geopolitical environment based on data from 146 countries over the period 2008—2021. It systematically evaluates the evolutionary paths and interaction characteristics of two systems—global energy transition and the global geopolitical environment—from two spatial scales (global and regional). The results show that the global energy transition level has maintained a steady upward trend overall, with significant progress in renewable energy development and energy efficiency, while per capita energy access capacity remains relatively insufficient. The geopolitical environment level has exhibited an overall downward trend since 2018, as the superimposed effects of political unrest, social inequality, and economic vulnerability have become increasingly prominent. At the global scale, the two systems show a significant positive correlation, reflecting a trend of coordinated evolution. At the regional scale, the interaction structure between energy transition and the geopolitical environment demonstrates obvious spatial heterogeneity. North America and South Asia are dominated by trade-off relationships, while Europe boasts the strongest synergistic effect.

Key Technologies for Resilient Urban Energy Systems Integrating Massive Distributed Flexible Resources

Heterogeneity Modeling and Low-Dimensional Broadcast Cooperative Control of Air Conditioning Load for Power Grid Demand Response

YU Junyi, LIAO Siyang, KE Deping, ZHANG Jie

2025, 58(12):  14-26.  DOI: 10.11930/j.issn.1004-9649.202504010

Asbtract ( )   PDF (2832KB) ( )  

Figures and Tables | References | Related Articles | Metrics

To address the demand for refined regulation of large-scale heterogeneous air-conditioning loads in power grid demand response, a second-order equivalent thermal parameter model is firstly used to derive the steady-state operating power of air conditioners, and the heterogeneity of air-conditioning parameters is characterized by a Gaussian mixture model to accurately calculate the regulation capacity. Then, an innovative low-dimensional broadcast signal control strategy of "central guidance + local autonomy" is designed. A Markov chain model combined with an augmented Lagrangian function is used to find the optimal temperature rise probability in different temperature intervals, balancing the power grid demand and economic cost constraints. Finally, the proposed method is verified through simulation analysis. The results show that the calculation accuracy of regulation capacity reaches 97.5% in a heterogeneous sample set of 50000 air conditioners, which can stably provide a maximum load reduction of 11188.12 kW. Moreover, the broadcast strategy can dynamically track the grid demand power and effectively alleviate the summer grid peak pressure. The proposed method provides theoretical support and a practical solution for large-scale thermostatically controlled loads to participate in demand response.

An Assessment Method for Power Grid Structural Resilience Based on Topological Feature Extraction

LIU Qixing, LEI Aoyu, ZHAI Zhe, LI Jialu, CHEN Yiping, WU Weimin, LEI Shunbo

2025, 58(12):  27-36, 49.  DOI: 10.11930/j.issn.1004-9649.202506001

Asbtract ( )   PDF (3448KB) ( )  

Figures and Tables | References | Related Articles | Metrics

With the frequent occurrence of extreme weather events and the large-scale integration of distributed energy resources, the operating environment of power grids has become increasingly complex, making enhancing power grid resilience a key to ensure the normal operation of society. To address the issues of traditional resilience assessment methods, such as strong dependence on source-load parameters and difficulty in coping with data uncertainty in practical engineering, this paper proposes a power grid resilience evaluation method integrating topological structure parameters and an adaptive analytic hierarchy process (A-AHP). The indicator of "maximum root node coverage radius" is introduced, combined with structural characteristics like betweenness centrality and closeness centrality, to comprehensively reflect the vulnerable links and overall collaborative capacity of the power grid under multi-unit black start recovery conditions. Through data-driven weight optimization and joint adjustment of classification thresholds, the transparency, interpretability, and classification accuracy of the model are improved. Simulation results show that the proposed method can effectively evaluate the impact of different black start resource allocation schemes on power grid resilience in the absence of detailed source-load parameters.

A Non-iterative Decentralized Collaborative Scheduling Method for Hydrogen-Electricity-Heat Integrated Energy Microgrid Clusters Based on Polyhedral Equivalent Aggregation

SHEN Yichun, PENG Hongyi, ZHANG Zhaocheng, YAN Mingyu

2025, 58(12):  37-49.  DOI: 10.11930/j.issn.1004-9649.202508019

Asbtract ( )   PDF (2297KB) ( )  

Figures and Tables | References | Related Articles | Metrics

To address the problems of high computational burden and difficulty in privacy protection in the cross-regional collaborative operation of hydrogen-electricity-heat integrated energy microgrid clusters, this paper proposes a non-iterative decentralized collaborative scheduling method based on polyhedral equivalent aggregation. Firstly, an aggregated equivalent model of the hydrogen-electricity-heat integrated energy microgrid is established, where each subsystem is represented by an equivalent generator and an equivalent energy storage device. Secondly, a polyhedral contraction method is proposed to map the feasible regions of the equivalent generator and storage device to a lower dimension. Finally, the contracted feasible regions are transformed into linear constraints that can be directly processed by solvers, which are embedded into the collaborative optimal scheduling problem of the hydrogen-electricity-heat integrated energy microgrid clusters. The optimal scheduling decisions of each microgrid can be obtained through solution. The proposed polyhedral equivalent aggregation approach avoids the exchange of private information such as user demands between microgrids and eliminates the time-consuming iterative processes existing in traditional distributed algorithms. The proposed model and method are tested on a two-area 6-6-8 bus hydrogen-electricity-heat integrated energy interconnected microgrid system and a two-area 40-33-13 bus hydrogen-electricity-heat integrated energy interconnection microgrid system, verifying its advantages such as computational efficiency and privacy protection.

DLMP Signal-Driven Orientated Inner Approximation Aggregation Scheduling Method for Distributed Resources in Distribution Networks

QIAO Li, MO Shi, GUO Mingyu, CUI Shichang, ZHANG Zitong, WANG Bo, AI Xiaomeng, FANG Jiakun, CAO Yuancheng, YAO Wei, WEN Jinyu

2025, 58(12):  50-62.  DOI: 10.11930/j.issn.1004-9649.202508045

Asbtract ( )   PDF (2515KB) ( )  

Figures and Tables | References | Related Articles | Metrics

To fully tap the flexible scheduling potential of distributed resources and address the volatility challenges brought by the large-scale integration of new energy into the distribution networks, this study proposes a distribution locational marginal price (DLMP) signal-driven orientated inner approximation (OIA)-based aggregation optimization scheduling method for distributed resources in distribution networks. Firstly, the aggregation direction is guided by minimizing the operating cost, and a small amount of feasible region space is selectively sacrificed during the OIA aggregation process to obtain higher-quality solutions, which overcomes the limitation of existing maximum inner approximation-based aggregation methods that pursue the maximization of feasible regions while ignoring the optimality of scheduling objectives. Meanwhile, a bi-level cooperative operation optimization framework between distribution network operators and load aggregators is proposed: the upper-level distribution network operator conducts the optimization scheduling with the goal of minimizing the total operation cost, including the power purchase cost of the distribution network and the wind/solar curtailment costs, and calculates the DLMP through Lagrange relaxation, which reflects the supply-demand balance status and operating costs of the distribution network from both spatial and temporal dimensions; the lower-level load aggregator receives the DLMP signal and updates the power consumption plans of distributed resources such as electric vehicles, variable-frequency air conditioners, and user-side energy storage within the cluster feasible region obtained by OIA aggregation, aiming to minimize the power consumption cost of agent users. The two parties use the DLMP signals as the medium for cooperative optimization, ultimately achieving the dual goals of secure and economic operation of distribution network and tapping the flexible potential of distributed resources. Case studies based on the IEEE 33-bus system show that the proposed method effectively guides the balances load distribution of the distribution network, reduces the power consumption cost of users represented by load aggregators, and lowers the network loss, node voltage deviation, and operating cost of the distribution network.

Ultra-Short-Term Load Forecasting Method for Aggregated Users Considering the Impact of Temperature-Controlled Load Characteristics

MENG Hao, XU Fei, FU Shuai, SUN Peng, HAO Ling, LIU Boyu, LIU Zhiwei

2025, 58(12):  63-72, 85.  DOI: 10.11930/j.issn.1004-9649.202502075

Asbtract ( )   PDF (3978KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The load of aggregated users with a high proportion of temperature-controlled loads is prone to abrupt in characteristics due to factors such as temperature variations, leading to temporal distribution shifts in load characteristics across different historical periods. This results in poor performance of existing load forecasting modeling methods for aggregated users due to insufficient generalization capability. Drawing on the concept of transfer learning for extracting domain-invariant features in the spatial dimension, an ultra-short-term load forecasting method for aggregated users based on time-domain invariant feature modeling is proposed. Since the cycles of temporal distribution shifts in load data and the boundaries of these cycles are typically unknown, firstly, the temporal distribution shift is quantified, and the load is segmented into sequences with significant distribution differences to support the subsequent extraction of time-domain common features among samples. Then, a Transformer-based time-domain invariant feature extraction algorithm is proposed, which minimizes the temporal distribution differences among data samples with varying distributions to extract time-domain invariant features, thereby optimizing load forecasting modeling and improving prediction accuracy under scenarios of abrupt load characteristic changes. Finally, the superiority of the proposed method is validated using real residential load data.

Collaborative Configuration of Multi-temporal and Spatial Flexible Resources in New Distribution Systems Considering Operational Risks

JIA Dongli, LIU Jiajing, ZHAN Huiyu, WANG Huanchang, BU Qiangsheng

2025, 58(12):  73-85.  DOI: 10.11930/j.issn.1004-9649.202502060

Asbtract ( )   PDF (2807KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The high proportion of renewable energy integration and diversified load connections have exacerbated the operational risks of high operating costs and high voltage deviations in new distribution systems. To address the issue that the current new distribution system does not consider the operational risks and the collaborative configuration of multi-temporal and spatial flexible resources across multiple time scales during the configuration phases, this paper proposes a collaborative configuration model for multi-temporal and spatial flexible resources considering operational risks. Firstly, a source-load scenario set is generated using Monte Carlo sampling and the K-means clustering algorithm, and the multi-scale morphological algorithm is employed to decompose the source-load curve waveforms at multiple scales. Then, based on the conditional value at risk (CVaR) theory, a quantitative assessment of the multi-temporal operational risks in the distribution system is conducted. On this basis, a bi-level configuration model for multi-temporal and spatial flexible resources considering operational risks is established. In this model, the upper level aims to minimize the annual total cost of the distribution system for collaborative configuration of the multi-temporal and spatial flexible resources, while the lower level focuses on minimizing the expected loss value and the CVaR-based operational risk value for system optimization. Finally, an improved IEEE 33-node system is used for case study, validating the proposed method can effectively reduce the operational risks related to high operating costs and voltage deviations in the distribution systems.

Key Technologies for Carbon Monitoring, Accounting, Carbon Footprint, and Carbon Management in New Power Systems

User-Side Dynamic Carbon Responsibility Accounting Method Considering Marginal Carbon Emissions and Demand Response

YU Wanshui, YI Jun, YANG Wenli, MIAO Bo, ZHANG Haotian, CHEN Wenjing, BAO Jixiu, JIN Xianglong

2025, 58(12):  86-95.  DOI: 10.11930/j.issn.1004-9649.202507009

Asbtract ( )   PDF (3228KB) ( )  

Figures and Tables | References | Related Articles | Metrics

In the pursuit of the dual carbon goals, electricity is accounting for a growing share in final energy consumption. Fairly assigning carbon responsibilities to end-users and incentivizing their participation in emission reduction are crucial for the low-carbon transition of power systems. To address the issues of regional inequity and insufficient incentives for demand response in carbon responsibility accounting on the user side of the power system, this paper proposes a user-side dynamic carbon responsibility accounting method incorporating marginal carbon emissions and demand response. The method introduces the concept of marginal carbon emission intensity and establishes a bilevel optimization model. The upper-level model optimizes electricity consumption behavior via a user demand response model, while the lower-level model determines the system carbon emissions based on an operator pre-dispatch model. The solution is achieved using the iterative decomposition and coordination method. Case study results demonstrate that the proposed method effectively and dynamically accounts for carbon responsibility on the user side, and using marginal carbon emission intensity as an incentive signal yields superior emission reduction outcomes. This approach provides a fair accounting basis for fulfilling user-side carbon responsibilities and allocating carbon quotas, thereby promoting coordinated carbon reduction across both generation and consumption sides.

Drivers of the Electricity Carbon Emission Factor: An LMDI-based Analysis and International Comparison

ZHANG Shining, HOU Fangxin, WEN Ya, LIU Yifang, YANG Fang

2025, 58(12):  96-106.  DOI: 10.11930/j.issn.1004-9649.202504015

Asbtract ( )   PDF (4025KB) ( )  

Figures and Tables | References | Related Articles | Metrics

The electricity carbon emission factor, as a core indicator for measuring the low-carbon progress of power systems, has become an important basis for the international carbon accounting system and green trade rules. Firstly, the core influencing factors for the electricity carbon emission factor were identified, and a method suitable for quantitatively assessing the contribution of the electricity carbon emission factor was proposed. Secondly, the contributions of different factors driving the decline of China's electricity carbon emission factor were quantitatively assessed, and the differences and reasons for the electricity carbon emission factors between China, Japan, and Germany were compared and analyzed. Finally, the potential for the decline of China's electricity carbon emission factor and the characteristics of the contributions of the main driving factors were quantitatively assessed. The study shows that from 2005 to 2022, China's electricity carbon emission factor decreased by 35%, with the increase in the grid cleanliness and the improvement of power generation efficiency being the two main influencing factors, contributing 19% and 14% respectively. China's electricity carbon emission factor is 22.5% higher than Japan's, primarily due to its heavy reliance on coal for thermal power. It is 53.5% higher than Germany's, mainly because of a lower clean energy share. Enhancing the grid cleanliness is the key driver for reducing China's electricity carbon emission factor, contributing over 80% to the decline. After new energy expansion reaches a certain scale, optimizing the structure of thermal power will also play a significant role.

Calculation and Analysis of the Electricity Carbon Footprint of Pumped Storage Power Stations in Power Systems

XU Sanmin, ZHANG Gong, ZHANG Yiwen, LIU Yanzhen, ZHANG Binliang, TANG Jin

2025, 58(12):  107-118.  DOI: 10.11930/j.issn.1004-9649.202507056

Asbtract ( )   PDF (4010KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Quantitative analysis of the carbon footprint of pumped storage power stations (PSPS) is conductive to the quantitative study on the green and clean attributes of pumped storage stations and improve the national average electricity carbon footprint factors. Based on the life cycle assessment (LCA) approach, this study analyzes carbon emission sources and their quantities of five typical PSPSs across all stages: including pre-construction preparation, construction, operation & maintenance, and decommissioning. Furthermore, it proposes an emission calculation method tailored to application scenarios for the pumping-generation phase. When used to improve the average carbon footprint factor for national power industry, the Jilin Dunhua PSPS (hereafter referred to as Dunhua Station) records a total lifecycle carbon emission of 1.7486 million tonnes. The daily operational emissions during its operation phase constitutes the primary source, accounting for more than 41%. At a 95% confidence level, the electricity carbon footprint from Dunhua Station ranges within [0.0235, 0.0266] kg/(kW·h). The average carbon footprint across the five studied PSPSs is 0.0238 kg/(kW·h), with operational maintenance costs and non-traceable construction costs identified as significant influencing factors for Dunhua Station. After incorporating the impact of pumped storage, the 2023 national average carbon footprint for electricity is calculated at 0.6206 kg/(kW·h). The influence of pumped storage on this national average is found to be less than one-thousandth (＜ 0.1%).

New-Type Power Grid

Analysis of Key Issues and Implications of the April 28 Great Power Blackout in Spain and Portugal

ZHANG Chen, GE Rui, JIN Xiaoling, SONG Pengcheng, SHENG Tongtian, XIONG Yuwei, DAI Xianzhong, ZHANG Xing

2025, 58(12):  119-127.  DOI: 10.11930/j.issn.1004-9649.202508030

Asbtract ( )   PDF (1997KB) ( )  

Figures and Tables | References | Related Articles | Metrics

On April 28, 2025, a nationwide power blackout occurred in Spain and Portugal. Prior to the incident, renewable energy output accounted for over 70% of the power generation, highlighting the distinct characteristics of a high-penetration renewable energy power system. Unlike previous blackouts triggered by external environmental factors or critical equipment failures, this incident stemmed from a systemic collapse of stability resulting from profound changes in source-grid structure, operational dynamics, and stability mechanisms. Through an in-depth analysis of the phenomena and causes of the Spain-Portugal blackout, this study deepens the case analysis from the perspectives of preventing accident escalation and recurrence, identifying seven critical challenges for maintaining stability in high-penetration renewable energy systems. As the renewable-energy-dominated power system progressing fast, high-penetration renewable energy output will become the norm in China's power production, implying that similar security risks will persist long-term. Drawing critical lessons from the Spain-Portugal blackout and considering current trends in the new-type power system, this paper assesses four major risks of large-scale blackouts in China's power grid and proposes targeted countermeasures.

Optimization Operation of Integrated Energy Station Coordinating the Interests of Distributed Generation Provider and Load Users

JIA Dongli, REN Zhaoying, LIU Keyan, WANG Zezhou, XIE Yifeng, YANG Kaitong, YIN Zhongdong

2025, 58(12):  128-136.  DOI: 10.11930/j.issn.1004-9649.202503005

Asbtract ( )   PDF (4390KB) ( )  

Figures and Tables | References | Related Articles | Metrics

To promote the utilization of renewable resources and improve economic interests of different subjects, here proposes a method for optimization operation strategy of Integrated Energy Station that coordinating different subject interests. Firstly, some demand response models for various loads were established by considering load uncertainty and time series electricity prices, based on the analysis of energy flow relationship of different subjects and the impact of demand response. Secondly, through introduction of carbon trading, her constructed three objective functions with the maximum net profit of Integrated Energy Station, the new energy consumption rate and the user's satisfaction. Thirdly, according to high dimensional and nonlinear features of the proposed objective, an improved multi-objective decomposition evolution algorithm is proposed by employing penalty function and polynomial mutation strategy. The simulation experiment shows that the proposed optimization model can increase the net revenue of the integrated energy station by 9.39%, the consumption rate of new energy by 8.18%and user satisfaction by 5.81%.

Entity Extraction Method of Power System Based on Semantic Enhanced Graph Convolutional Neural Network

JI Xin, WU Tongxin, WANG Honggang, LI Jianfang, CHEN Yiting

2025, 58(12):  137-146.  DOI: 10.11930/j.issn.1004-9649.202505026

Asbtract ( )   PDF (3294KB) ( )  

Figures and Tables | References | Related Articles | Metrics

With the deepening development of new-type power systems, there is an urgent demand for processing unstructured texts in areas such as equipment condition monitoring, intelligent fault diagnosis and dispatch instruction parsing. To address the heterogeneity of entity types such as equipment, operations and faults, along with the complex semantic relations in power texts, a semantic enhanced graph convolutional network method for entity extraction in power systems is proposed. First, data augmentation strategies such as synonym replacement, random insertion, swapping, and deletion are applied to reduce the expression differences in manually recorded texts. Then, a robustly optimized bidirectional encoder representation from transformers pretraining approach (RoBERTa) with a dynamic masking mechanism and whole-word masking is used for fine-grained semantic encoding of domain-specific terms. Next, a semantic relation graph is constructed, including nodes such as equipment, operations, and system states. Finally, a relation extraction framework based on large language models is introduced. This framework incorporates a part-of-speech filtering matrix to enhance graph convolution feature aggregation. Experiments show that the proposed model achieves score of 88.79% on dispatch logs, outperforming existing models by 2.3%~3.1%. The model can accurately identify power operation and status entities such as "closing the tie switch" and "grounding disappears", providing strong support for updating dispatch knowledge bases and enabling intelligent decision-making under fault conditions.

Transformer Excitation Inrush Current Suppression Method Based on Dynamically Adjustable Resistors

SONG Yujie, LI Zhenxing

2025, 58(12):  147-154.  DOI: 10.11930/j.issn.1004-9649.202503084

Asbtract ( )   PDF (4071KB) ( )  

Figures and Tables | References | Related Articles | Metrics

When the transformer is closed under light load or no load, the magnetizing inrush current caused by the saturation of the core magnetic flux can easily differential protection maloperation, transient voltage dip and other issues, which seriously threatens the safe operation of the power system. The generation mechanism of transformer magnetizing inrush current is analyzed and a transformer magnetizing inrush current suppression method based on dynamic adjustable resistance is proposed. By controlling the size of the adjustable resistance connected to the voltage division circuit, the on the primary side of the transformer is smoothly increased, avoiding the magnetizing inrush current caused by the rapid rise of voltage and the saturation of the core magnetic flux. This alleviates the problem of large magnetizing inrush current generated when the transformer is closed under light load or no load.

Similar-Day Electricity Consumption Prediction Considering Meteorological Sensitivity

ZHENG Kuanyun, CHEN Yiping, SHAN Shuaijie, LIU Xiu, YU Yixiao

2025, 58(12):  155-164.  DOI: 10.11930/j.issn.1004-9649.202507060

Asbtract ( )   PDF (4491KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Accurate electricity consumption prediction can provide technical support for the precise supply of electricity. To achieve high-precision prediction, a method based on meteorologically similar days for electricity consumption prediction is proposed. By deeply exploring the temporal fluctuation characteristics of electricity consumption and the meteorological sensitivity, and combining key meteorological factors, the matching degree of similar days is improved, thereby enhancing the prediction accuracy. Firstly, the linear correlation degree between each meteorological factor and electricity consumption is analyzed through Pearson correlation coefficient. Then, through P-value test, the key meteorological factors affecting electricity consumption are screened out by integrating the results of both. And the temperature data that exhibits the highest correlation with electricity consumption is classified using the K-means clustering algorithm. Historical days are categorized into two types: high-temperature days and moderate-temperature days, thereby establishing a similar-day typology framework. Secondly, within the same temperature category, the grey relational analysis is used to calculate the grey correlation degree between the meteorological factors of the target day and those of the historical days. By integrating Euclidean distance, the historical days most similar to the target day are screened out to deeply explore the temporal fluctuations and meteorological sensitivity characteristics of electricity consumption. Subsequently, a combined prediction model is constructed based on the similar-day data, comprehensively considering both meteorological factors and time-series characteristics, thereby enhancing prediction accuracy. Finally, the proposed method is validated using the electricity consumption data from a residential district in a certain area of Shandong Province. The results demonstrate that the proposed method has significant advantages in addressing the challenges of electricity consumption prediction.

Measurement Error Identification of Protection Secondary Circuit by Fusing Chebyshev Distance and Modified Cosine Similarity Algorithms

WANG Haiming, WANG Jianfeng, WANG Zhongyuan, JIANG Jiali, WENG Hanli

2025, 58(12):  165-177.  DOI: 10.11930/j.issn.1004-9649.202508007

Asbtract ( )   PDF (5624KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Correctly identifying the measurement error of the protection secondary circuit is crucial to improving protection reliability and ensuring stable operation of the power system.Therefore, a method using the entropy weight-based fusion of Chebyshev distance and modified cosine similarity algorithm (EBW-CDMCS) is proposed to accurately identify measurement errors of the protection secondary circuits. Firstly, in order to ensure the accuracy and consistency of data analysis, two sets of measured data undergo standardized preprocessing, and the distance values between the two sets of sampled data are obtained using the Chebyshev distance and the modified cosine similarity algorithm, respectively, to characterize the dispersion of the data. Then, the entropy-based weight method is used to calculate the weights of the two sets of distance values, which are then fused to obtain the EBW-CDMCS distance index. Finally, by analyzing the measured current data of the secondary circuit under normal operating conditions, an appropriate threshold is set. The calculated EBW-CDMCS distance index is compared with this threshold to identify potential measurement errors in the secondary circuit. A 220kV substation model was built based on PSCAD, and simulation results verified the effectiveness and accuracy of the proposed method.

A KAN-BiLSTM-based Power Load Forecasting Method Utilizing Composite Factor Construction

CHEN Jingwen, HUANG Yuqian, LIU Yaoxian, CHEN Songsong, QIAN Xiaorui, ZHOU Ying, ZHAN Xiangpeng

2025, 58(12):  178-189, 198.  DOI: 10.11930/j.issn.1004-9649.202504033

Asbtract ( )   PDF (4809KB) ( )  

Figures and Tables | References | Related Articles | Metrics

To address such problems as the insufficient consideration of the interaction of meteorological factors, and the limitations of model's nonlinear expression ability, a power load forecasting method based on Kolmogorov-Arnold network (KAN) and bidirectional long short-term memory (BiLSTM) network utilizing composite factor construction is proposed. Firstly, the power load curves with similar characteristics are classified through the Gaussian mixture model. Secondly, a composite factor construction strategy is proposed: the linear correlation degree between meteorological factors and loads is quantified through Pearson correlation analysis; key meteorological variables are screened out and interaction terms are constructed to fully explore the potential interaction among meteorological factors, and nonlinear dependent features are further extracted with maximum information coefficient. Finally, aiming at the problem that the fully connected layer of the traditional BiLSTM model has limited ability to learn high-dimensional nonlinear features, the KAN is introduced to replace the fully connected layer, and a KAN-BILSTM hybrid prediction model is constructed using its nonlinear mapping ability. The experimental results demonstrate that the proposed method has high prediction accuracy and universality under four different load modes. It can provide a feasible solution for the precise prediction of power load in multi-meteorological coupling scenarios.

New Energy and Energy Storage

Carbon Reduction Potential Assessment of Recycling Retired NCM Batteries from a Life Cycle Perspective

DENG Xiangyan, DONG Xinzhao, DING Xu, WU Zhuhui, HAI Bao, BAI Zhenming, YAO Zheng, QIAN Zihao, LI Jianxi

2025, 58(12):  190-198.  DOI: 10.11930/j.issn.1004-9649.202410041

Asbtract ( )   PDF (3550KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Proper disposal of of electric vehicles retired batteries is crucial for improving resource utilization efficiency and reducing potential carbon emissions. This study, based on the life cycle assessment method, analyzes the carbon footprint of retired batteries across various stages, including recycling, disassembly, cascade utilization and regeneration, and explores the optimization of recycling paths for retired ternary lithium batteries from this perspective. The results indicate that the overall recycling of retired ternary lithium batteries provides positive emission reduction benefits, with the carbon footprint of recycling 1 kW·h of retired NCM batteries being approximately –51.2 kg CO₂-eq and –30.1 kg CO₂-eq for hydrometallurgical and pyrometallurgical paths, respectively. The hydrometallurgical path shows the greatest emission reduction during the regeneration stage, while the pyrometallurgical path shows the greatest reduction during cascade utilization. Compared to the pyrometallurgical process, the hydrometallurgical process has greater emission reduction advantages due to its higher material recovery rates and lower energy consumption. Developing renewable energy to optimize the power structure, increasing the proportion of cascading utilization, and applying the hydrometallurgical recycling can effectively enhance the carbon reduction benefits of recycling retired NCM batteries.

Multi-dimensional Value Assessment and Attribution Analysis of Electrochemical Energy Storage for Multi-service Coupling

WU Bingqing, YUE Hao, GUAN Haowen, MENG Zijian, SHAN Tihua, YANG Linyan, WU Zhaoyuan

2025, 58(12):  199-210.  DOI: 10.11930/j.issn.1004-9649.202507004

Asbtract ( )   PDF (4108KB) ( )  

Figures and Tables | References | Related Articles | Metrics

Driven by the "dual-carbon" goals, electrochemical energy storage (EES) has emerged as an indispensable flexibility resource in new-type power systems due to its high efficiency and operational flexibility. However, no systematic and interpretable framework is in place to comprehensively quantify the multidimensional values of EES across multiple coupled services and to identify the underlying drivers of value formation. Therefore, this paper proposes an integrated framework for multidimensional value and attribution analysis of EES under multi-service coupling scenarios. Firstly, a multidimensional value indicator system covering economic, technological, and environmental aspects is established to systematically reflect the comprehensive contribution of energy storage in multiple services such as peak shaving and frequency regulation. Furthermore, based on the random forest model, the nonlinear driving relationship between system operational characteristics and energy storage value is identified, revealing the value formation mechanism under complex operational conditions. The case study results demonstrate that energy storage providing multi-service coupling can increase system inertia by 46.09%, improve the minimum frequency by 0.21 Hz, reduce carbon emission intensity by 18.9%, and exhibit significant synergistic effects across multiple value dimensions. The proposed method offers a mechanism-driven and interpretable quantitative basis for value mining, functional positioning, and configuration optimization of energy storage, supporting the strategic goals of enhancing flexibility and promoting low-carbon transition in new-type power systems.

Day-Ahead Photovoltaic Power Forecasting Based on Large Language Model with Meteorological Covariate Attention

GE Yanshuo, ZHOU Yanzhen, GUO Qinglai, XIAO Dajun, XU Xialing, LI Xin, LIU Tao

2025, 58(12):  211-222.  DOI: 10.11930/j.issn.1004-9649.202504060

Asbtract ( )   PDF (6011KB) ( )  

Figures and Tables | References | Related Articles | Metrics

To improve the accuracy of day-ahead photovoltaic (PV) power forecasting and mitigate the challenge of data scarcity in newly built photovoltaic power stations, leveraging both the reasoning capabilities of large language models (LLMs) and the correlation information across various dimensions of sequences, this paper proposes a meteorological covariate attention-enhanced Time-LLM model for day-ahead PV power prediction. Firstly, the input sequence is constructed by padding and concatenating historical PV power series with meteorological covariate series. Then, a novel covariate attention module is introduced to capture the cross-dimensional dependencies between meteorological variables and PV power sequences. Finally, the Time-LLM architecture is employed to achieve modality alignment between time-series and text sequences, effectively leveraging the textual analysis capabilities of LLMs for accurate PV power forecasting. Experimental results on public PV datasets demonstrate that the proposed model achieves superior forecasting performance and exhibits remarkable zero-shot learning capability. The proposed method not only improves the accuracy of day-ahead photovoltaic power forecasting, but also provides a promising solution for newly built PV plants with limited historical data, where traditional deep learning models often fail due to data scarcity.