Research on Lssvm Sintering Main Suction Inlet Pressure Prediction Based on Error Correction
In order to improve the accuracy of predicting the inlet pressure of the sintering plant,the least squares support vector machine is applied for prediction.The prediction model is improved by adding complementary set empirical mode decomposition and Markov chain method.The main inlet pressure time series is decomposed into different components through complementary set empirical mode decomposition.These components are used as model inputs and trained using LSSVM models to obtain initial predicted values.Then,based on Markov chain theory,an error state transition matrix is established to obtain the corrected error,which is combined with the initial predicted values to obtain the final predicted value.The improved algorithm model has higher prediction accuracy than the LSSVM algorithm,which verifies the effectiveness of the model.
万方数据
维普
