Prediction of matte grade in the oxygen-rich top blown smelting based on WD-SSA-SVM algorithm

In view of the defects of noise in the industrial data of the oxygen-rich top blown smelting and time lag in off-line detection of matte grade, it is difficult to accurately establish the prediction model of process parameters. A process parameter prediction model based on wavelet denoising sparrow search algorithm optimized support vector machine (WD-SSA-SVM) was proposed. Firstly, wavelet denoising is used to improve the quality of the data. Secondly, the sparrow search algorithm was used to optimize the support vector machine to establish the nonlinear relationship model between the melting process and the process index, and the model parameters were identified by the sample data after denoising, so as to effectively predict the process index of frosted sand grade in the process of the oxygen-rich top blown smelting. The experimental results on the actual production data of a smelter show that the WD-SSA-SVM model proposed in this paper has high accuracy, meets the accuracy requirements in the actual industrial production, and can effectively guide the optimization and adjustment of the operating parameters of the oxygen-rich top blown smelting process.