Development of non-linear fault detection system using Cusum-based technique applied to a pilot distillation column

摘要

Early detection of faults (FD) is important in chemical industry since a lot of damage and loss can result before a fault present in the system is detected. In this paper, a real-time system for detecting changes in dynamic systems is designed. The main contribution consists in the design of a simplified procedure of the incident detection scheme through a combination of the optimisation property of cumulative sum and a neural adaptive black-box identification. The simplicity of the developed neural model, under all regimes (i.e. steady-state and unsteady state), used in this case is realised by means of a non-linear auto-regressive with exogenous input model and by an experimental design. The performance of the proposed FD system has been tested on a real plant as a distillation column. Then, the FD system has been tested under real anomalous conditions. The experimental results demonstrate the robustness of the FD method.