Short-term load prediction method

Through the method of fusion of variational modal decomposition and multi-model, combined with time convolution network and long and short-term memory network, the problem of insufficient accuracy of the existing load prediction method under nonlinear and dynamic changes is solved, and more accurate short-term load prediction is achieved, providing guarantees for the stable operation of the power system.

CN119849963APending Publication Date: 2025-04-18SHENZHEN POWER SUPPLY BUREAU
View PDF 5 Cites 4 Cited by

Patent Information

Application Number
CN202411823721.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-12
Publication Date
2025-04-18

AI Technical Summary

Technical Problem

The existing load prediction methods face challenges such as nonlinearity, complexity and dynamic changes, resulting in insufficient prediction accuracy and difficult to meet the stable operation needs of the power system.

Method used

The short-term load prediction method of fusion of variational modal decomposition and multi-model is adopted to collect and preprocess multivariate data, and use a model combining time convolutional network and long and short-term memory network to predict, and improve prediction accuracy through hyperparameter optimization and historical data training.

Benefits of technology

Effectively extract the nonlinear features in the power load data, and achieve more accurate load prediction through the complementary advantages of multiple models, providing solid guarantees for the stable operation of the power system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119849963A_ABST
    Figure CN119849963A_ABST
Patent Text Reader

Abstract

The invention provides a short-term load prediction method, which comprises the following steps: collecting multivariable data of a power system, and preprocessing the collected multivariable data; wherein the multivariable data at least comprises power load data, weather data, holiday and festival information and date data; inputting the preprocessed multivariable data into a pre-trained prediction model to obtain a corresponding prediction result; wherein the prediction model is a model combining a time convolution network and a long short-term memory network, hyper-parameter optimization is carried out through a preset search algorithm, training is carried out through historical multivariable data, and a plurality of intrinsic mode component prediction results are combined to be output as the model; and determining the short-term load corresponding to the power system according to the prediction result. According to the method, the nonlinear features in the power load data are effectively extracted, more accurate load prediction is realized through advantage complementation of multiple models, and a solid guarantee is provided for stable operation of a power system.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Aero-engine residual service life prediction algorithm based on convolutional memory residual self-attention mechanism

    CN113158445A

  • TCN-LSTM-based short-term load prediction method

    CN113902207A

  • Short-term power load prediction method combining VMD decomposition and time convolution network

    CN114358389A

  • Short-term building load prediction method

    CN117526294A

  • Short-term power load prediction system and method

    CN117791570A

Cited By

  • Network fault prediction method and system based on deep learning

    CN121441775A

  • Short-term load prediction method and system based on load decomposition and parameter optimization

    CN121529530A

  • Load prediction method and device for optimizing gray neural fusion model through genetic algorithm

    CN121710164A

  • A method and apparatus for short-term power load forecasting based on feature engineering

    CN122677944A