Modified fuzzy c-means combined with neural network based fault diagnosis approach for a distillation column

This work aims to propose a reliable method that can be used in the steady-state regime of a nonlinear procedure. Such method should be able to distinguish between normal and abnormal operations. Moreover, we propose a modified procedure of the fuzzy c-means clustering method (MFCM) that can be used in place of classical features extraction and selection step. MFCM calculates the percent variation between two clustered classes i.e., normal and faulty modes. This step is followed by the application of artificial neural network (ANN) classifier. Then, this methodology is tested on real experimental data obtained from the distillation column, in order to capture and diagnose the faults that may occur during the automated continuous distillation process. The aim of using MFCM is to decrease the calculation time and increase the performance of the classifier. The results of the proposed method confirm the ability to classify between normal and eight abnormal classes of faults.