A Bio-Inspired Met heuristic Framework for IoT Botnet Mitigation and Renewable Energy Forecasting Using Tree Growth and Rooster Optimization

Abstract

The exponential rise in Internet of Things (IoT) devices has significantly expanded the attack surface of global cyber infrastructures, leading to the emergence of sophisticated botnet-based threats that compromise system integrity and data confidentiality. Simultaneously, the growing integration of renewable energy systems into IoT-driven smart grids necessitates accurate forecasting models to ensure grid stability and energy optimization. This study introduces a bio-inspired metaheuristic framework that synergistically employs Tree Growth Optimization (TGO) and Rooster Optimization Algorithm (ROA) to address two critical challenges: IoT botnet mitigation and renewable energy forecasting. The proposed dual-domain framework utilizes TGO for optimal feature selection in intrusion detection systems (IDS) and ROA for hyperparameter optimization in deep neural models tailored for renewable energy prediction. The hybrid architecture demonstrates superior convergence behavior, high detection accuracy, and efficient energy forecasting performance. Experimental validation using benchmark datasets such as BoT-IoT and Solar Power Data confirms that the framework achieves a 98.7% detection accuracy with a 12% lower computational cost compared to traditional optimization approaches. These findings establish the proposed model as a versatile and adaptive solution for intelligent IoT ecosystem management.