Fault diagnosis in a distillation column using a support vector machine based classifier

Fault diagnosis has always been an essential aspect of control system design. This is necessary due to the growing demand for increased performance and safety of industrial systems is discussed. Support vector machine classifier is a new technique based on statistical learning theory and is designed to reduce structural bias. Support vector machine classification in many applications in various fields of machine learning has been successful and appears to be effective for fault diagnosis in industrial systems. This project is to design a support vector machine fault diagnosis system for a distillation tower as a key component of the process. The study included 41 stage distillation condenser and boiler theory is that a combination of two partial products of 99% purity breaks Based on the calculations, modeling and simulation is a tray to tray. Considering the variety of different origins faults in the system under study, a multi-class classification problem can be achieved two techniques commonly used to solve multi-class classification for support vector machine as "one to one" and "one against all" is used. The classifier models designed to detect faults in the systems studied were evaluated as successful results were obtained for all types of faults. The model was designed based on the speed in detecting various faults were compared on the basis of support vector machine model based on a technique called "One on One" have delivered a better performance.