With the rapid development of the power grid, the high penetration of new energy sources and the diversity of loads have further aggravated the uncertainty of “source-load”, which has brought huge challenges to the peak shaving of the power grid. In order to ensure the reliable operation of the power system during the electricity peak, this paper combines the IoT technology to propose a flexible control strategy for controllable large industrial loads that considers the interrupt priority. Firstly, the perception and fusion framework of large industrial load information is constructed based on the IoT technology. After that, the improved TOPSIS method is adopted to establish the evaluation model of the adjustable potentials of large industrial loads and the load interruption priority is further divided. Finally, a three-stage rolling regulation model for controllable large industrial loads to participate in the deep peak shaving of the power grid is constructed to achieve the goal of bidirectional peak shaving on the power generation side and the demand side. The case uses an improved IEEE 30-node system for simulation. As the simulation results show, the method proposed in this paper can not only take the cost of peak load regulation into account, but also effectively achieve the goal of ‘peak shaving and valley filling’.

In order to reduce the influence of abnormal data on load forecasting effects and further improve the training efficiency of forecasting models when adding new samples to historical data set, an ultra-short-term load dynamic forecasting method considering abnormal data reconstruction based on model incremental training is proposed in this paper. Firstly, aiming at the abnormal data in ultra-short-term load forecasting, a load abnormal data processing method based on isolation forests and conditional adversarial generative network (IF-CGAN) is proposed. The isolation forest algorithm is used to accurately eliminate the abnormal data points, and a conditional generative adversarial network (CGAN) is constructed to interpolate the abnormal points. The load-influencing factors are taken as the condition constraints of the CGAN, and the weighted loss function is introduced to improve the reconstruction accuracy of abnormal data. Secondly, aiming at the problem of low model training efficiency caused by the new samples in the historical data set, a model incremental training method based on a bidirectional long short-term memory network (Bi-LSTM) is proposed. The historical data are used to train the Bi-LSTM, and the transfer learning is introduced to process the incremental data set to realize the adaptive and rapid adjustment of the model weight and improve the model training efficiency. Finally, the real power grid load data of a region in eastern China are used for simulation analysis. The calculation results show that the proposed method can reconstruct the abnormal data more accurately and improve the accuracy and efficiency of ultra-short-term load forecasting.

An improved differential evolution (DE) method based on the dynamic search strategy (IDEBDSS) is proposed to solve dynamic economic dispatch problem with valve-point effects in this paper. The proposed method combines the DE algorithm with the dynamic search strategy, which improves the performance of the algorithm. DE is the main optimizer in the method proposed. While chaotic sequences are applied to obtain the dynamic parameter settings in DE, dynamic search strategy which consists of two steps, global search strategy and local search strategy, is used to improve algorithm efficiency. To accelerate convergence, a new infeasible solution handing method is adopted in the local search strategy; meanwhile, an orthogonal crossover (OX) operator is added to the global search strategy to enhance the optimization search ability. Finally, the feasibility and effectiveness of the proposed methods are demonstrated by three test systems, and the simulation results reveal that the IDEBDSS method can obtain better solutions with higher efficiency than the standard DE and other methods reported in the recent literature.

Existing electric vehicle (EV) charging scheduling strategies commonly rely on weak temporal coupling to approximate cross-period travel behaviors, which makes it challenging to achieve orderly and precise charging coordination under distribution network operational constraints. To address this issue, this paper proposes a coordinated cross-temporal charging scheduling optimization method for EVs considering distribution network security constraints.First, a time-varying adaptive path optimization approach oriented to cross-temporal mobility characteristics is developed. By introducing time-segmentation and dynamic weight updating mechanisms, a cross-temporal recursive Dijkstra algorithm is constructed to accurately evaluate travel times across multiple time intervals.Second, a time-of-use (TOU) pricing regulation model based on distribution network demand response is established. Taking charging stations and EV users as interactive stakeholders, a Stackelberg game framework is formulated to derive dynamic pricing strategies. The proposed mechanism guides charging loads toward spatiotemporally optimal self-adaptive distribution, achieving an effective balance between distribution network operational security and user satisfaction.Finally, simulations conducted on a realistic regional road network demonstrate that the proposed method reduces user charging costs, improves voltage profiles of the distribution network, and enhances charging station revenues. The results validate the effectiveness of the proposed framework and provide practical insights for large-scale coordinated operation of coupled transportation–distribution networks with high EV penetration.