Blast furnace bosh gas volume prediction method and program for multi-clustering prototype-based T-S model

Abstract

The invention discloses a blast furnace bosh gas volume prediction method and program for a multi-clustering prototype-based T-S model. The method comprises the steps of performing data preprocessingon blast furnace data, and deleting abnormal values; performing variable screening by utilizing a Spearman rank correlation coefficient; selecting a hyperspherical clustering prototype-based FCM fuzzyclustering algorithm and a hyperplane clustering prototype-based NFCRMA fuzzy clustering algorithm to perform fuzzy clustering on the blast furnace data, performing calculation by using membership functions corresponding to the two fuzzy clustering algorithms to obtain two types of different T-S model antecedent rule fitness, sorting the two types of the rule fitness from high to low, and finallycalculating a weighted sum of the two types of the rule fitness to obtain weighted rule fitness; and calculating consequent parameters of the T-S model by using a least square method, and finally adjusting a weighted coefficient to improve the prediction precision of the model. The method can accurately predict a value of a blast furnace bosh gas volume index at a next moment.

Description

technical field [0001] The invention relates to the field of blast furnace ironmaking, in particular to a method and program for predicting bosh gas volume based on a multi-clustering prototype T-S model. Background technique [0002] The iron and steel industry is a pillar industry of the national economy and a symbol of the level of technological development of a country. Blast furnace ironmaking is the upstream process of the steel production process, and it is a major energy-intensive and high-polluting household. How to keep the blast furnace running smoothly and achieve the purpose of saving energy and reducing consumption is of great significance to the improvement of economic benefits of iron and steel enterprises and the sustainable development of the national economy. The blast furnace gas flow participates in the reduction reaction and heat transfer in the furnace, which is the most direct reflection of the energy utilization and temperature field distribution of...

AI Technical Summary

Problems solved by technology

However, the hypergeometric shape of high-dimensional data is uncertain, and it is impossible to find a clustering prototype suitable for all kinds of high-dimensional data.

In previous studies, the T-S model’s antecedent identification process only fuzzy clustered the data based on a clustering prototype, and could not fully extract the feature information contained in the data.

Images