The rise of Electric Vehicles (EVs) and their integration into modern infrastructure have created a growing need for secure and user friendly charging systems. Electric Vehicle Charging Stations (EVCS), enhanced with Internet of Things (IoT) technology, allow for seamless data exchange and remote management, making charging more convenient for users. However, these connected systems face increasing risks from cyber threats, calling for stronger security measures. This paper introduces a machine learning-based Intrusion Detection System (IDS) combined with homomorphic encryption to address these challenges and protect EVCS networks in real time. The system uses traffic data from EVCS networks to detect and respond to cyberthreats. Machine learning classifiers analyze and classify this data into binary or multiclass categories, identifying potential attacks like Distributed Denial of Service (DDoS) or spoofing. Homomorphic encryption adds an extra layer of security by ensuring sensitive data remains encrypted even during analysis, protecting user information and system operations. Together, these technologies create a robust solution that minimizes disruptions and enhances system reliability. This research sheds light on critical security challenges in EVCS networks, uncovering vulnerabilities and attack vectors that could compromise operations. It demonstrates how machine learning and privacy focused cryptography can work together to detect anomalies and strengthen system defenses. By integrating advanced security measures like homomorphic encryption, the proposed solution ensures a safer, more reliable EV charging experience, paving the way for wider adoption of electric vehicles in a secure and trusted environment.