Study On Fault Diagnosis For Glutamic Acid Fermentation Process

The glutanic acid is the first big amino acid product in the world. The fermentative process product quality fluctuates in a big way. The mistake and the breakdown are not easy in early time to discover and easy to create raw material the waste and equipment's idle operation. This article takes the glutanic acid fermentation as the object of study, to fermentative process's failure detection and the diagnosis development research, has certain theory significance and the practical application value.The present paper has given process breakdown identification mechanism and the strategy based on the multi-direction principal element analysis, and has established the multi-direction principal element anatomic model glutanic acid fermentation failure diagnosis system which based on the variable. Through inspects the SPE statistics and Hotlling T2 statistics discovered that the normal fermentative process #4 SPE statistics and Hotlling T2 statistics have not surpassed the respective confidence of 95% control limitses throughout. But fermenting unusually in #6, SPE statistics when the 12th hour has surpassed its confidence of 95% control limitses, from this might judge in the glutanic acid fermentative process had not the normal situation. Then through exceptionally ferments the pot to approve #6 the procedure variable to SPE contribution chart analysis knowing, when the 12th hour, the rotational speed was biggest to the SPE statistics's contribution value, showed the problem in the 12th hour on agitation rotational speed.The present paper has used the auto-association neural network (AANN) to carry on the online failure diagnosis to the glutanic acid fermentative process the validity and the practicability research. The auto-association neural network is one kind of belt bottleneck level, 5 structure special neural network. Through carried on suitable screening to fermentative process's apparent variable, we have selected OD620, RPM, OUR, CER and the ammonia water consumption finally in 5-6-2-6-5 AANN networks carrying on the training data 5 entries of variables to the structure. The different performance's fermentative process may obtain the cluster and may carry on the failure diagnosis using evaluating indicator J to the glutanic acid fermentation. Raid of #4 and the #5 J value have not achieved or close"the control limit"from beginning to end and may recognize that for is the normal fermentation. But the raid of #6 J value, surpasses"the control limit"at about the 12th hour, and continues about 5-6 hours only then to start to recede. According to the above judge during this period of time, the glutanic acid fermentation presented the breakdown. The raid of #7 J value has surpassed"the control limit"in the 12.5th hour, after testing the personnel receive the alarm, after the analysis and to the breakdown has carried on processing, when the 16th h makes the glutanic acid fermentation to restore normally, and enabled the glutanic acid density to achieve 56.3 g/L levels finally, before this ratio trouble shooting, enhanced 20 g/L.