Innovative Application of Gradient Descent to Optimize Strip Process Parameters

Cold-rolled steel strip, renowned for its high strength, excellent toughness, and other superior properties, is extensively utilized across various industries. However, coupling parameters in the continuous annealing process poses significant challenges for quality control. To address this issue, this study employs the gradient descent algorithm to optimize the process parameters. By defining clear objectives, identifying key parameters, establishing a loss function, as well as iteratively updating the parameters, an optimal parameter combination is identified, thereby enhancing product quality and production efficiency. Experimental results demonstrate that the algorithm exhibits outstanding performance in optimizing hardness errors, with a notably low MSE value. Looking ahead, ‌research will focus on developing adaptive or real-time optimization systems to propel the intelligent development of the steel industry.