Load prediction method based on decomposition representation and attention encoder network

By employing a load prediction algorithm based on trend-seasonal decomposition representation and a hybrid attention encoder network, the problem of ineffective utilization of complex temporal features of load in oil and gas development is addressed, achieving higher prediction accuracy.

CN121836020APending Publication Date: 2026-04-10SICHUAN UNIV
0 Cites 1 Cited by

Patent Information

Application Number
CN202512037870.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-31
Publication Date
2026-04-10

AI Technical Summary

Technical Problem

Existing load forecasting models fail to effectively decompose and utilize the complex characteristics of load time series in oil and gas development, resulting in insufficient forecast accuracy.

Method used

A load prediction algorithm based on trend-seasonal decomposition representation and hybrid attention encoder network is adopted. By contrastive learning, trend and seasonal components are extracted, and the hybrid attention encoder network is combined to deeply explore the dependencies between features, thereby improving the model's ability to extract long-range dependency information.

Benefits of technology

It improves the accuracy of load forecasting and reduces forecasting error. Experiments show that the forecasting error is reduced by 8.51% compared with the best baseline algorithm in the task of forecasting electricity load for oil and gas development.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to a load prediction method TSR-MAEN based on decomposition representation and an attention encoder network. Firstly, the algorithm provides a trend-seasonal decomposition representation method based on comparative learning so as to enhance the decoupling performance of a model on a trend component and a seasonal component of a load sequence and the representation capability of decomposition characteristics. Secondly, in order to deeply mine and decompose time sequence information contained in a feature hidden space and dependency between features, the algorithm proposes a mixed attention encoder network; through segmented embedding, a trend-season dynamic projection mixed self-attention mechanism and a multi-head mixed output mechanism, trend and season multi-dimensional features after load sequence decomposition are effectively utilized to express and enhance the long-range dependence information extraction capability of the model. Experiments show that the TSR-MAEN algorithm provided by the invention has excellent performance in an electrical load prediction task of oil and gas development, and the prediction error is reduced by 8.51% compared with that of an optimal baseline algorithm.
Need to check novelty before this filing date? Find Prior Art