Designing a Model for Earthquake Timing and Magnitude Prediction based on Neural Networks and Particle Swarm Optimization (PSO) Algorithm

The present study offers a hybrid predictive model integrating Artificial Neural Networks (ANN) and Particle Swarm Optimization (PSO) as a way of forecasting earthquake timing and magnitude in Saman, Iran, having a strong focus on vibration signal analysis and dynamic measurement. The offered model implements 12 vibration-based input features, including peak ground acceleration (PGA), shear wave velocity, and spectral intensity, all of which are derived from seismotectonic and accelerometer data. PSO optimizes ANN weight initialization, and this can enhance the ability of the model to capture seismic wave dynamics for applications relevant to vibration engineering. The dataset, made up of historical seismic records, was split into 80% for training and 20% for testing. The ANN-PSO model showed better performance compared to the conventional ANN and Support Vector Machine (SVM) methods and achieved an average accuracy of 94.1% for magnitude and 91.7% for timing, with a mean squared error (MSE) determined at 0.023. Precision and recall rates were determined at 92.8% and 93.4%, respectively, and the training time decreased by 26% compared to standard ANN implementations. The model, validated over 20 independent runs and using dynamic measurement experiments, showed consistent performance. Thus, it was identified as a robust tool in the field of vibration-based seismic forecasting, structural health monitoring, and mechanical reliability analysis in regions that are tectonically active.