In this paper, it addresses the challenge of optimizing feature selection using Adaptive Genetic Algorithms (AGA) integrated with Crowding Distance Selection (CDS) to enhance the effectiveness of intrusion detection. The primary objective of this research is to develop a method that automatically selects the most pertinent network traffic features for intrusion detection. In this paper, we propose a novel AGA-CDS that adapts genetic algorithms to efficiently explore the search space and the effectiveness of crowding distance selection in maintaining diversity and enable convergence to a Pareto-optimal front. Experimental results show the efficacy of the proposed approach, showing an improvement in feature selection performance compared to traditional methods. The proposed AGA with CDS achieves a feature subset that outperforms existing techniques by 2% effectively reducing dimensionality while preserving or enhancing classification accuracy in intrusion detection tasks.