Power load combination prediction method and system, electronic equipment and medium

CN119315528APending Publication Date: 2025-01-14SHANXI CITY POWER NEW ENERGY CO LTD +1
View PDF 0 Cites 6 Cited by

Patent Information

Application Number
CN202411344873.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-25
Publication Date
2025-01-14

AI Technical Summary

Technical Problem

The existing power load prediction methods have problems such as low prediction accuracy, easy to fall into local optimization and slow convergence speed.

Method used

The improved parrot algorithm and BiTCN-BiGRU-Attention combined model are adopted to process power load data through empirical modal decomposition of fully adaptive noise ensembles, and the improved parrot optimization algorithm is used to optimize the model hyperparameters, combining bidirectional time convolution network and bidirectional gating cycle Unit and attention mechanisms perform power load prediction.

Benefits of technology

The accuracy and efficiency of power load prediction are improved, local optimal traps are avoided, and the global search capability and convergence speed of the model are enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119315528A_ABST
    Figure CN119315528A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of control in new energy power generation, and particularly relates to a power load combination prediction method and system, electronic equipment and a medium. The method comprises the following steps: firstly, decomposing power load data by utilizing ICEEMDAN to obtain a plurality of subsequences, and combining the subsequences with relevant characteristics such as meteorological data to form an input sequence of a prediction model; secondly, optimizing internal hyper-parameters of the neural network by using an IPO algorithm; then, a BiTCN-BiGRU-Attention model is established for the input sequence for training and prediction, the model carries out preliminary feature extraction on historical load and meteorological data through BiTCN, then, BiGRU is used for further mining time sequence association among feature data, and an attention mechanism is introduced, so that the model can efficiently capture complex time sequence features in power load prediction; finally, the predicted values of all the subsequences are accumulated, and a final load predicted value is output.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Temperature and humidity prediction method for solution dehumidification air conditioning system and application of temperature and humidity prediction method

    CN119514389A

  • Electrocardiogram classification method for optimizing residual network based on improved parrot algorithm

    CN120694656A

  • An electrocardiogram classification method based on improved parrot algorithm optimization residual network

    CN120694656B

  • Distributed power supply configuration detection method and system based on deep learning fusion model

    CN121579985A

  • Short-term bus load prediction method, device, equipment and medium

    CN122159183A