abstract |
A mechanical property prediction method of a container rolled plate is characterized in that a mechanical property prediction model is established through an improved BP neural network model, the mechanical properties of seven steel grades of the container rolled plate are predicted, and a predicted value is output. The prediction model can be self-monitored and self-updated, and when the prediction precision does not meet the requirement, the prediction model is automatically updated. A mechanical property prediction method of a container rolled plate enables prediction accuracy of tensile strength, yield strength and elongation of the container rolled plate to reach 97.84%, 96.79% and 92.06%, so that a traditional manual inspection mode is replaced, mechanical property of the container rolled plate is predicted more conveniently and accurately, process parameters and component parameters are optimized through the prediction values, blindness in a design process is reduced, and time and cost are saved. |