An improved method for detecting and identifying gross errors simultaneously

A linear combination technology is applied to identify gross errors (bias and leaks) and estimate their magnitude in a linear steady state process. A linear combination scheme is used firstly to isolate the candidate subset of gross errors and then simultaneous identification/estimation of gross errors is accomplished. Comparative studies of performance is performed on a recycle network and a steam system simulation results show that the proposed approach has higher performance for identifying gross errors as well as to estimate their magnitude.