Power of process and sensor fault detection methods for stationary mineral processing plants

摘要

The power of a statistical test for fault detection in chemical or mineral processing plants is an indicator of the efficiency of the detection method. The paper proposes a model-based method which is able to detect fault in the process model as well as in the measurements. The model consists of stationary mass conservation equations, allowing the calculations of the covariance matrix of the model uncertainties. The test is a X{sup}2 test, the power of which can be analytically calculated. The method is illustrated for a flotation circuit using either instantaneous data or measurements over an observation window of varying width.