Predictive analytics of disc brake deformation using machine learning

Abstract

This paper explores the application of machine learning techniques to predict disc brake deformation, a critical aspect in ensuring the safety and reliability of braking systems. The study employs a dataset comprising 50 data points, with input parameters such as pressure and an output parameter of deformation. The focus is on developing predictive models that can accurately estimate disc brake deformation under various operating conditions. To achieve this, four machine learning approaches are investigated and compared: Decision Tree, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). Each model is trained and evaluated based on its ability to predict deformation, with performance assessed using R-squared metrics. Results indicate notable variations in the predictive capabilities of the models. The Random Forest model emerges as a top performer, demonstrating robustness in capturing complex relationships within the dataset. The Decision Tree model exhibits competitive performance, showcasing its suitability for interpretable predictions. Meanwhile, the SVM model, while effective, exhibits sensitivity to the choice of kernel function. The KNN model, with its simplicity and flexibility, also offers promising results. This research provides valuable insights into the effectiveness of different machine learning approaches in predicting disc brake deformation. It is found that the Random Forest model achieved an accuracy of 99%. These results suggest that Random Forest model are more effective at predicting disc brake deformation than SVMs. The findings contribute to the advancement of intelligent braking systems, enhancing safety and reliability in automotive applications.