Optimized-extreme-learning-machine-based detection system and method for blast furnace fault

Abstract

The invention, which belongs to the technical field of blast furnace fault detection, provides an optimized-extreme-learning-machine-based detection system and method for a blast furnace fault. On the basis of an artificial bee colony (ABC) algorithm, an input weight and an implicit strata threshold value in an extreme learning machine (ELM) are optimized, thereby establishing a reasonable and effective classification model; on the basis of Tent mapping, an initial colony with uniform distribution is generated, thereby improving the quality of an initial solution and enhancing stability of the method; good balancing between global searching and local searching is realized by using an adaptive searching strategy; and jumping out of a local optimal solution is realized by using a Tent chaotic local searching strategy. According to the invention, the convergence speed and the optimizing searching precision of the ABC algorithm are improved; the classification precision of the blast furnace fault detection is enhanced; the probability of generating a new solution is increased; involvement in a local optimal solution is avoided; the rapid evolution direction of the algorithm is guaranteed; and the global searching capability is guaranteed and the local searching capability of the algorithm is also improved to a certain extent.

Description

technical field [0001] The invention belongs to the technical field of blast furnace fault detection, and in particular relates to a blast furnace fault detection system and method based on an optimized extreme learning machine. Background technique [0002] In iron and steel enterprises, the blast furnace ironmaking production is in the leading and important position, and the energy consumption accounts for about 60%. Once a failure occurs, it will cause huge losses of personnel and property; therefore, realize the intelligent, automatic, timely and accurate fault detection process Accurately detect and predict the abnormal state of the blast furnace ironmaking production process, which can not only reduce accidents, ensure stable and reliable production operation, but also reduce production management costs and improve product quality; Enterprise competitiveness is of great significance; blast furnace smelting is a complex continuous production process. Parameters such as...

AI Technical Summary

Problems solved by technology

The ABC algorithm has a strong global search ability, but the local optimization ability is weak, and the convergence speed is slow in the later stage of evolution.

Images