Smart cost estimation: Empirical case for extra-high voltage transmission towers

Area of Science:

• Engineering
• Computer Science

Background:

• Electricity is essential for modern life, requiring robust infrastructure like Extra-High Voltage (EHV) transmission towers.
• Manual cost estimation for EHV towers is labor-intensive and time-consuming.
• An automated, precise cost prediction tool is needed to improve efficiency.

Purpose of the Study:

• To develop a Support Vector Regression (SVR) model for accurate prediction of EHV transmission tower construction costs.
• To automate and streamline the cost estimation process for EHV transmission projects.
• To enhance the resilience and robustness of transmission network systems through efficient cost management.

Main Methods:

• Literature review to identify cost-influencing attributes for EHV transmission towers.
• Data collection from 238 EHV transmission tower construction projects in Taiwan (2009-2019).
• Development and validation of an SVR model with optimized parameters (C=0.2, γ=0.1) using 5-fold cross-validation.

Main Results:

• The SVR model achieved an average prediction accuracy of 97.91%.
• The developed model effectively predicts construction expenses for EHV transmission tower projects.
• Significant reduction in time required for cost estimation compared to manual methods.

Conclusions:

• The proposed SVR-based model offers an accurate and efficient solution for EHV transmission tower construction cost estimation.
• Automating this process contributes to improved planning and resource allocation in the power transmission sector.
• Enhanced cost prediction supports the development of more resilient and robust electricity transmission networks.