Online SVR Neural Network Based Ultra High Voltage Transmission Early Warning of Construction Progress

Ultra high-voltage (UHV) transmission construction involves live work, requires crossing long distances, and must overcome terrain and climate challenges. Therefore, to enhance the accuracy of UHV transmission construction progress prediction and address the issue of online early warning, an early warning method for UHV transmission construction progress based on the online support vector regression (Online SVR) algorithm is proposed. Utilizing the digital twin model of UHV transmission construction, principal component analysis is employed to extract and analyze key factors influencing construction progress, the Online SVR algorithm dynamically learns and predicts construction progress. The prediction results are compared with the planned construction warning line to provide early warnings of potential construction progress risks. Finally, the method is applied to an actual UHV transmission construction site to verify its effectiveness.