Research Article
Deep Learning-Based Hazard Evaluation for Resource Network Setting Up via Horizontal Directional Drilling
@INPROCEEDINGS{10.1007/978-3-031-77075-3_26, author={Rathish Manivannan and Manivannan Sethuraman}, title={Deep Learning-Based Hazard Evaluation for Resource Network Setting Up via Horizontal Directional Drilling}, proceedings={Cognitive Computing and Cyber Physical Systems. 5th EAI International Conference, IC4S 2024, Bhimavaram, India, April 5--7, 2024, Proceedings, Part-I}, proceedings_a={IC4S}, year={2025}, month={2}, keywords={Hazard evaluation Drilling input horizontal directional drilling (HDD) resource network tunnel construction}, doi={10.1007/978-3-031-77075-3_26} }- Rathish Manivannan
Manivannan Sethuraman
Year: 2025
Deep Learning-Based Hazard Evaluation for Resource Network Setting Up via Horizontal Directional Drilling
IC4S
Springer
DOI: 10.1007/978-3-031-77075-3_26
Abstract
A pipeline and another utility are transported through the underground tunnel construct utilizing horizontal directional drilling (HDD). This method includes a drilling tunnel under a river or other specified location. It is crucial to ensure the safety and dependability of HDD operations, particularly while drilling in areas with significant resource networks. Ensuring data accuracy and high quality in dynamic drilling circumstances that could impact the efficiency of the system was challenging. The study presented the Red Deer Optimized Bidirectional RNN algorithm (RDO-Bi-DRNN) method to examine and predict the possible hazards related to HDD operations in the resource network. The RDO-Bi-DRNN method assesses hazard evaluation by developing resource networks through HDD. The research collected a dataset of 3,082 data of different drilling input parameters and pre-processed the data using the min-max normalization to generate a high-quality dataset for analysis. To evaluate the performance of the proposed method by using the Python tool. This study examines the significant performance indicators, such as Accuracy (95%), Recall (97%), Precision (94%), and F1-Score (98%). This method provides professionals and planners of projects with the knowledge and skills that are necessary to ensure the security, dependability, and sustainability of resource network development.
