HRB400E steel mechanical property prediction method based on data preprocessing and bidirectional recurrent neural network

CN119227520APending Publication Date: 2024-12-31CISDI ENGINEERING CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411255480.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-12-31

AI Technical Summary

Technical Problem

The prior art has problems such as high cost, time-consuming, poor accuracy, insufficient robustness and poor versatility in the prediction of steel mechanical properties, especially when relying on traditional physical metallurgy models and existing artificial intelligence algorithms.

Method used

Using a method based on data preprocessing and bidirectional recursive neural network (BRNN), the HRB400E steel mechanical performance prediction model is constructed through data collection, preprocessing, hierarchical clustering, outlier culling, data equalization, feature screening and model training. Bayesian optimization is used to determine the model hyperparameters to improve prediction accuracy and robustness.

Benefits of technology

It significantly improves data quality and prediction accuracy, enhances the robustness and applicability of the model, reduces the dependence of laboratory tests, saves time and costs, and has a wide range of application prospects and industrial practicality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119227520A_ABST
    Figure CN119227520A_ABST
Patent Text Reader

Abstract

The invention relates to an HRB400E steel mechanical property prediction method based on data preprocessing and a bidirectional recurrent neural network, and belongs to the technical field of steel mechanical property prediction. The method comprises the steps of data collection, data preprocessing, data mining, feature screening, construction of a bidirectional recurrent neural network model and the like. The data quality is improved through methods of clustering analysis, abnormal value deletion, data equalization processing and the like, and time sequence features and complex nonlinear relationships in the data are better captured by using a BRNN algorithm. In addition, optimal hyper-parameters of the model are determined by adopting a Bayesian optimization method, so that a high-precision mechanical property prediction model is obtained. According to the method, dependence on laboratory tests is greatly reduced, time and cost are remarkably saved, the adaptive capacity of the model in actual production application is enhanced, and efficient and accurate prediction driven by data is achieved.
Need to check novelty before this filing date? Find Prior Art