Prediction for Silicon Content in Molten Iron Using a Combined Fuzzy-Associative-Rules Bank

摘要

A general method is developed to generate fuzzy rules from numerical data that collected online from No.1 BF at Laiwu Iron and Steel Group Co.. Using such rules and linguistic rules of human experts, a new algorithm is established to predicting silicon content in molten iron. This new algorithm consists of six steps: step 1 selects some key variables which affecting silicon content in molten iron as input variables, and time lag of each of them is gotten; step 2 divides the input and output spaces of the given numerical data into fuzzy regions; step 3 generates fuzzy rules from the given data; step 4 assigns a degree to each of the generated rules for the purpose of resolving conflicts among the generated rules; step 5 creates a combined Fuzzy-Associative-Rules Bank; step 6 determines a fuzzy system model from input space to output space based on such bank. The rate of hit shot of silicon content is more than 86% in [Si] ± 0.1% range using such new algorithm.