文献类型：专利

英文题名：Ethylene cracking reaction alarming method for cracking furnace, involves judging working conditions of cracking furnace according to output of neural network alarming model of ethylene cracking

作者：LIU M;LI S;HUANG H;ZHANG J

机构：[1]UNIV EAST CHINA SCI & TECHNOLOGY

申请号：CN101551663-A

申请日：2009-05-08

公开日：2009-10-07

语种：英文

收录：DERWENT

摘要：NOVELTY - The method involves selecting operation conditions of a cracking furnace as an input variable of a neural network alarming model in an ethylene cracking. Marker bits of working conditions are selected as an output variable of the neural network alarming model. Real production data of an industrial cracking is utilized as a training sample to build up the neural network alarming model of the ethylene cracking. Working conditions of the cracking furnace are judged according to an output of the neural network alarming model. USE - Method for realizing alarming for an ethylene cracking reaction in a cracking furnace on the basis of a distributed control system and a spot control system (all claimed). ADVANTAGE - The method effectively judges the working conditions whether the cracking is executed normally, so that abnormal tendency can be promptly found and alarm the working staff, which enables the working staff to perform corresponding adjustment, thus eliminating the abnormal conditions as fast as workers to reduce occurrence of accidents and enhances the producing interests. The method can be implemented easily at low cost without maintenance, and ensures real-time alarming. DESCRIPTION OF DRAWING(S) - The drawing shows a flowchart illustrating a method for alarming a cracking reaction.'(Drawing includes non-English language text)'