Performance Analysis of Artificial Gorilla Troop Optimization Algorithm Features based BAT Algorithm for Ground Vibrations Detection using Different Geophone Sensors

• Published in: 2025 International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI) pp. 1 - 4
• Main Authors: Manivannan, Gowri Shankar, Deb, Sanjoy, J, Guruprasath, M, Gunasekaran, G, Vasundhara M, Talawar, Satish V
• Format: Conference Proceeding
• Language: English
• Published: IEEE 28.03.2025
• Subjects: Accuracy; AGTOA; BAT; Classification algorithms; Feature extraction; Geophone Sensor; Ground Vibration; Intelligent sensors; LDA; Monitoring; Optimization; Performance analysis; Sensor phenomena and characterization; Sensors; Vibrations
• DOI: 10.1109/ICDSAAI65575.2025.11011823

Ground vibration sensors, including geophone sensors, are utilized in several applications that involve monitoring seismic activities and structural health. In order to ensure effective monitoring, it is very important to be able to classify and analyze the signals from geophone sensors with precision. In this approach, the first part deals with extracting features in which Linear Discriminant Analysis (LDA) is applied, which then becomes the feature selection Artificial Gorilla Troop Optimization Algorithm (AGTOA). The categorization is done through the BAT algorithm for signals from a geophone sensor with 10Hz and 30Hz frequencies. The AGTOA-based BAT classifier performed better for the 10Hz sensor, with an average accuracy of 93.88% as compared to 91.86% for the 30Hz sensor. This proves the efficacy of the method proposed; even with inferior quality of data in the lower frequencies, the classifier yields a better performance.