Highway interchange section state discrimination method based on improved fuzzy C-means clustering algorithm

As an important transportation channel and traffic hub, the highway is frequently congested due to the rapid growth of traffic demand. Therefore, this paper adopts the Mahalanobis distance, which is more sensitive to the change relationship and discreteness between different dimensional data, to replace the traditional Euclidean distance to improve the algorithm, and constructs a traffic congestion discrimination model based on the improved fuzzy C-means clustering algorithm. The model takes the measured data of the highway interchange section as the input, and discriminates the highway operation condition according to the clustering results. Through empirical verification, the discrimination effect of the model is significantly better than the traditional clustering algorithm and the current speed threshold discrimination method, and has good applicability and accuracy.