Strip steel deviation monitoring and diagnosing method and system for soaking pit of continuous annealing unit

Abstract

The invention provides a strip deviation monitoring and diagnosis method for a soaking furnace of a continuous annealing unit, comprising: establishing a monitoring model of a strip deviation; The strip steel of specification builds PCA model, and calculates the control limit of two statistics T2 and SPE, establishes LSSVM model for the strip steel of each specification; Monitoring: Select the online monitoring PCA model and the online monitoring LSSVM model, and use the online monitoring PCA model and the online monitoring LSSVM model to monitor according to the data samples of the current production process collected at regular intervals; When the LSSVM alarm is carried out again, it is judged that the strip steel in current production deviates. The invention also provides a strip deviation monitoring and diagnosis system for the soaking furnace of the continuous annealing unit. The invention improves the accuracy rate of early warning and the accuracy rate of fault diagnosis for strip deviation.

Description

technical field [0001] The invention relates to a method and system for monitoring and diagnosing deviation of strip steel, in particular to a method and system for monitoring and diagnosing deviation of strip steel used in a soaking furnace of a continuous annealing unit. Background technique [0002] The continuous annealing unit is an important process in the cold rolling mill. The strip steel after cold rolling needs to travel through the continuous annealing unit at high speed, and then undergoes the processes of heating, soaking, slow cooling, rapid cooling, overaging, water quenching, and leveling. To eliminate the internal stress of the strip. [0003] After the heated thin steel strip enters the soaking furnace, the tension of the strip steel often fluctuates on the drive roller due to temperature changes, and then deviation or slippage occurs in the soaking furnace. If the degree of deviation is large, there will be production accidents such as broken belts. At p...

AI Technical Summary

Problems solved by technology

Although this method can realize the automatic monitoring of strip deviation, it also has the following disadvantages: first, this method can only provide early warning for strip deviation, but because it is impossible to install a camera monitoring system for each drive roller, the This method cannot be used for fault diagnosis, that is, it is impossible to determine where the strip deviation begins; secondly, for the sections with high temperature and large number of drive rollers such as heating furnace and soaking furnace, it is difficult to install image monitoring in these sections. devices, so that the method cannot be applied to these sections

[0007] Patent ZL201010192542.5 proposes a strip tension monitoring and fault diagnosis method based on principal component analysis (PCA), but since the PCA method cannot reflect the nonlinear relationship between the control variables, the strip steel The accuracy rate of tension fluctuation early warning and the accuracy rate of fault diagnosis based on the contribution rate of control variables are not high, and false early warning and inaccurate fault source diagnosis are prone to occur

Images