Diagnosis of faulty measurements by using fuzzy relational matrix

The process data are the foundation upon which all process control and evaluation of process performance is based. Inaccurate data generate erroneous interpretations of process behavior and they create faulty chains of reasoning leading to unsustainable decisions. In this paper, a fuzzy-relational-matrix-based scheme for diagnosis of faulty measurements from process data is proposed. A new approach for solving the problems of data reconciliation and diagnosis of faulty measurements by using intelligent techniques is presented. Simulation on test examples was used to evaluate the performance of the proposed method. Numerical results showed that the problem of diagnosing the faulty measurements has been effectively solved by the developed algorithm. The obtained results demonstrate the ability of the suggested fault diagnosing procedure for implementation into real complex technological systems.