Hybrid Deep Learning Framework for Sustainable Infrastructure Protection and Resilience Enhancement in Smart Cities

Smart cities increasingly be contingent on interconnected digital infrastructures to support transportation, healthcare, energy management, industrial automation, and public communication systems. However, the fast expansion of intelligent infrastructures disclosures urban environments to cyber threats, network failures, data breaches, and operational unsteadiness. Traditional security and monitoring systems often lack adaptability, scalability, and resilience against growing attack patterns. This Paper suggests a Hybrid Deep Learning Framework for Sustainable Infrastructure Protection and Resilience Enhancement in Smart Cities. The proposed framework combines the Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) and Federated Learning mechanisms to empower the adaptive threat detection, infrastructure resilience analysis, and privacy-preserving distributed intelligence. The model executes real-time anomaly detection, predictive resilience valuation and sustainable infrastructure optimization across distributed smart city environments. Experimental evaluation demonstrates improved detection accuracy, reduced computational overhead, enhanced resilience response and sustainable operational efficiency compared with conventional deep learning approaches. The proposed framework strongly aligns with Sustainable Development by promoting resilient infrastructure, sustainable industrialization and intelligent innovation for future smart city ecosystems.