This paper presents a novel approach to cybersecurity by integrating blockchain technology, RSA hashing, and deep learning techniques to enhance data security and intrusion detection. We employ Differential Evolution (DE) for intelligent data selection from blockchain, ensuring that sensitive information is securely managed. The data is then partitioned into training and testing sets through a Money-grubbing Simulated Annealing algorithm, optimizing the dataset for model performance. A Deep Belief Network (DBN) is utilized to predict and classify potential intrusions with high accuracy, leveraging its ability to detect complex patterns in large datasets. Our method ensures robust data integrity and security through blockchain, while simultaneously achieving superior classification precision in intrusion detection tasks. Experimental evaluations and simulations validate the effectiveness of the proposed system, demonstrating significant improvements in both security and performance. This work highlights the potential of combining blockchain- based data security with advanced machine learning models, offering a scalable and efficient solution to the growing challenges in cybersecurity.