Traditional machine learning (ML) techniques have transformed modern day computing with predictive analytics and automated decision-making. However, these struggle with real-time, high-accuracy, and scalable systems. Quantum Machine Learning (QML) provides a solution that incorporates ML with quantum computing principles by leveraging quantum phenomena like superposition and entanglement to achieve exponential speed-ups and improved accuracy. This work investigates the application of QML to forecast electricity demand, addressing the limitations of classical machine learning algorithms when dealing with high-dimensional, nonlinear data. Two QML models, Quantum Support Vector Classification (QSVC) and Quantum k-Nearest Neighbors (Quantum k-NN), were implemented using the UK electricity demand dataset. The models leverage quantum kernels, superposition, and entanglement to improve prediction accuracy. Results show that QML models outperform classical ML algorithms, achieving classification accuracies of 94.00% (QSVC) and 95.00% (Quantum k-NN). This study highlights the potential of QML for real-time, high accuracy applications in energy grid management, while also discussing the challenges posed by current quantum hardware limitations, noise, and algorithm complexity (Acknowledgement: This publication has emanated from research conducted with the financial support of Taighde Éireann – Research Ireland under Grant number 13/RC/2077_P2 at CONNECT: the Research Ireland Centre for Future Networks.).