Research And Application On Fault Monitoring And Quality Prediction Against The Fermentation Process

Fermentation processes have been utilized to produce products in the chemical, vintage, biological industries. It is an nonlinear, dynamic and multistage batch process. It is hard to describe by certainty mathematical model because of process mechanism's complexity and bad repeatability of data. Therefore, data-driven technology is a good solution of this complicate and noliner biochemical reaction system, and become hot research of process control in recent years. Online monitoring and quality prediction for fermentation process is build statistical model based on process history data to detect the fault, process upsets and other abnormal events promptly, locating and removing the factors causing such events, and the safety of production process will be assured and the quality of product will be improved.This article fully utilize the fermentation process data to build statistical model, and making a systematic study on the problems of online monitoring, fault diagnosis and quality prediction. Some new monitoring algorithms are also proposed, the main contents are as follows:1) An orthogonal signal correction (OSC) method is proposed as a preprocessing procedure of MPLS method to remove from X the variation information which is not correlated to Y. The resulted model has reduced complexity and improved interpretation ability.2) For multistage, time-variant, nonlinear characteristic and unavailable on-line product qualities of fermentation process, a multi-stage OSC-MPLS method is proposed. Using ISODATA dynamic clustering algorithm, process data was automatically divided into several operation stages according to relevance. Multi-stage statistical analysis strategy can increase the accuracy of fault detecting and online evaluation effectively.3) In multistage process, both the whole batch duration and each phase duration can be different from run to run, due to disturbances in operating conditions or different process settings. This article using recursive DTW algorithm to synchronize these unequal Sub-phase.4) Fermentation fault monitoring software development:A set of software for modeling and monitoring of the fermentation process has been designed by VC++ 6.0 in Windows system. The software can read current and historical data through OPC and ADO to establish the multistage OSC-MPLS online local model of actual fermentation process, at the same time the dynamic real-time curve of model predictive output and the major measurable variables can be draw when the new data arriving. The development of the software is helpful to realize the optimal control of the fermentation process.