Power transmission and transformation project progress recognition and analysis method and system based on deep learning

By using deep learning technology, a recursive transformer fusion network model was constructed. Combined with multimodal data analysis, the accuracy and efficiency problems of traditional power transmission and transformation project progress monitoring were solved. This enabled efficient and accurate project progress identification and trend prediction, supporting modern project management.

CN119514844BActive Publication Date: 2026-05-29ECONOMIC & TECH RES INST OF HUBEI ELECTRIC POWER COMPANY SGCC
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Patents(China)
Current Assignee / Owner
ECONOMIC & TECH RES INST OF HUBEI ELECTRIC POWER COMPANY SGCC
Filing Date
2024-10-09
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Traditional methods for monitoring the progress of power transmission and transformation projects rely on manual inspection, which makes it difficult to process and analyze large amounts of data in real time. This results in poor accuracy in progress identification, and existing data analysis methods have low prediction accuracy in complex projects, failing to meet the needs of modern project management.

Method used

Using a deep learning-based approach, a recursive transformer fusion network model is constructed by acquiring site plans, images, and video data of power transmission and transformation projects. Multimodal feature extraction and fusion are performed, and dynamic self-attention adjustment is combined to identify project progress and predict future trends, and analyze key factors and differences.

Benefits of technology

It achieves high-precision, real-time project progress identification, reduces human interference, improves identification efficiency and accuracy, provides scientific trend analysis support, and reduces project management risks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119514844B_ABST
    Figure CN119514844B_ABST
Patent Text Reader

Abstract

A kind of power transmission and transformation engineering progress identification and analysis method and system based on deep learning, method first acquires and pre-processes power transmission field multimedia data, constructs engineering progress data set, then utilizes plan and progress data set to train deep learning model, then input model with new progress data to identify current engineering state, and compare planned progress, analyze deviation, finally based on model and key influencing factors are comprehensively trend forecasted;The design in application, through the recursive transformer fusion network model fused with multiple modules and the acquisition of key factors of various difference data, and combining the difference of different voltages realizes comprehensive trend analysis, so that it can effectively extract deep features from multiple data sources, realize high-precision identification of engineering progress, compared with traditional manual inspection and recording method, greatly improve the identification efficiency and accuracy, reduce the interference of human factors, ensure the real-time and reliability of progress identification.
Need to check novelty before this filing date? Find Prior Art