Cold-rolled strip steel fault diagnosis optimization method based on PCA-Kmeans algorithm

Abstract

The invention discloses a cold-rolled strip steel fault diagnosis optimization method based on a PCA-Kmeans algorithm. According to the method, a mathematical model is optimized by combining a PCA dimension reduction algorithm with a K-means clustering algorithm, Barkhausen signal characteristic data are optimized by utilizing the PCA algorithm, and then optimized low-dimensional data are processed by combining a K-means unsupervised learning algorithm, so that prediction and diagnosis of cold-rolled strip steel faults are completed. The mathematical model is optimized through PCA, and the prediction precision can reach 95% or above.

Description

technical field [0001] The invention relates to an improved clustering algorithm, uses PCA-optimized data features to establish a prediction model, and completes cold-rolled strip fault diagnosis and prediction, in particular to a cold-rolled strip fault diagnosis optimization method based on the PCA-Kmeans algorithm. Background technique [0002] With the continuous development of my country's industry, industries such as railway transportation, aerospace, shipbuilding, automobiles, energy, and construction have developed rapidly. Ferromagnetic materials are functional materials, and they have become key industrial raw materials in the industrial field. . Ferromagnetic materials such as steel are one of the most widely used materials and have become an important foundation for industrial prosperity and economic construction. Therefore, strictly controlling the quality of ferromagnetic materials such as steel has become an important task, and it has broad application prospec...

AI Technical Summary

Problems solved by technology

Exploring the same dataset in a high-dimensional vector space is more difficult than exploring the same sparse dataset

[0005] Multi-dimensional feature data will undoubtedly provide rich information for cold-rolled strip fault prediction and diagnosis models, but it also affects data processing and model prediction accuracy to a certain extent

If each indicator is analyzed separately, the analysis is often isolated and the information in the data cannot be fully utilized, so blindly reducing indicators will lose a lot of useful information, resulting in wrong conclusions

Images