Acoustic networks facilitate productive work for enterprises and researchers. However, communication and network life span require large energy storage batteries. So, enhancing IoT network life in an acoustic environment is major issue in this research domain. This paper has proposed a model Acoustic IOT Optimization by Prediction and Clustering (AIOTOPC) that works on IoT communication to reduce the energy losses. The biogeographic optimization method was utilized to group nodes in an underwater Internet of things network, which was optimized because the dynamic environment required a self-adaptive system. Additional cluster nodes were examined in order to gather sensing data. In this paper, a learning and prediction system using node sensing data was proposed. Experiment was done on different sets of IoT nodes and volume. The results indicate that the suggested model extends the network life span, increases the packet count, and requires less time for clustering.