Data-Driven Modelling And Computational Intelligence For Predictive Engineering And Smart Systems

The fast development of smart systems of engineering has augmented pressure on precise and robust predictive modelling structures that have the capacity to manage intricate and information-intensive settings. This paper suggests a hybrid data-driven computational intelligence framework that combines ensemble learning with optimization-based parameter optimization to improve the predictive performance. The framework was tested in comparison to the traditional machine learning models, i.e., Artificial Neural Networks, Support Vector Machines, and Random Forest, with the statistical validation measures, i.e. Accuracy, RMSE, Precision, Recall, and F1-Score. The results of the empirical study indicate that the hybrid model proposed has better accuracy, convergence, and robustness in the presence of noise. The reliability and generalization possibility of the approach are verified with the help of cross-validation and significance testing. Moreover, the framework is extended to the intelligent engineering fields like predictive maintenance and energy forecasting, which denotes its usefulness in practice. The results prove the hybrid model of computational intelligence to be a scalable and effective solution to the next-generation predictive engineering and intelligent decision-support systems.