ORIGINAL ARTICLE Fault detection in distillation column using NARX neural network Syed A. Taqvi 1 • Lemma Dendana Tufa 1 • Haslinda Zabiri 1 • Abdulhalim Shah Maulud 1 • Fahim Uddin 1 Received: 23 January 2018 / Accepted: 24 July 2018 Ó The Natural Computing Applications Forum 2018 Abstract Fault detection in the process industries is one of the most challenging tasks. It requires timely detection of anomalies which are present with noisy measurements of a large number of variable, highly correlated data with complex interactions and fault symptoms. This study proposes the robust fault detection method for the distillation column. Fault detection and diagnosis (FDD) for process monitoring and control has been an effective field of research for two decades. This area has been used widely in sophisticated engineering design applications to ensure the proper functionality and performance diagnosis of advanced and complex technologies. Robust fault detection of the realistic faults in distillation column in dynamic condition has been considered in this study. For early detection of faults, the model is based on nonlinear autoregressive with exogenous input (NARX) network. Tapped delays lines (TDLs) have been used for the input and output sequences. A case study was carried out with three different fault scenarios, i.e., valve sticking at reflux and reboiler, and tray upset. These faults would cause the product degradation. The normal data (no fault) is used for the training of neural network in all three cases. It is shown that the proposed algorithm can be used for the detection of both internal and external faults in the distillation column for dynamic system monitoring and to predict the probability of failure. Keywords Aspen plus Ò simulation  Distillation column  Fault detection  NARX neural network  Nonlinear process  Process monitoring 1 Introduction Multivariate statistical process control (MSPC) has been extensively used for the monitoring of large-scale process plants [1–5]. However, the monitoring task in process industries is difficult due to various reasons. These reasons can be complexities in the process itself, high nonlineari- ties, dynamic operations, automation [6] and lack of online sensors for product composition measurements. Failure in one of the instruments could be a reason for variation from steady-state operation unless it is detected promptly and a corrective action is taken on time. Therefore, the best information to improve product quality as well as minimize the risk of accidents can be achieved by fault detection and diagnosis (FDD) system. Accidents caused by anomalies in the system occur every year, and they keep on increasing every year in chemical process plants around the world [7, 8]. The pro- cess faults affect the performance of the plant as well as system components; hence, early detection and diagnosis of the faults are needed in order to prevent process plant from major challenges such as safety issues and plant shutdown. There is a need of intelligent technique for the detection and identification of the fault in complex chem- ical processes. Fault detection is essential to monitor the continuity of operation under normal operating conditions. Fault detection in any chemical process will help to min- imize the disturbance and to keep the system safe and reliable [9, 10]. A fault can be defined as the anomalous behavior that causes the system to deviate unacceptably from its normal operating regime [7, 11, 12]. In chemical process plants, there are different types of faults which can be categorized according to their sources, i.e., sensor faults affect the process measurements, and actuator faults lead to an error in the plant’s operation. These faults can arise abruptly, such as sudden failure of any equipment at the plant, or evolve over time with gradual wear and tear & Lemma Dendana Tufa lemma_dendena@utp.edu.my 1 Chemical Engineering Department, Universiti Teknologi PETRONAS, 32610 Seri Iskandar, Perak Darul Ridzuan, Malaysia 123 Neural Computing and Applications https://doi.org/10.1007/s00521-018-3658-z