Abstract: An efficient approach to identify the structural condition of free spanning pipeline based on nonlinear discriminant analysis (KDA) is presented. The main characteristics of condition identification for free span (CIFS) are discussed firstly. Then, condition vectors consisting of natural frequencies, normalized frequencies and frequency change ratios are constructed based on internal relationship of frequency construction. Afterwards, an effective algorithm for CIFS is established based on the statistical pattern recognition of KDA. Finally, the validity of the proposed approach is evaluated by a case study. The effects of selection of condition vectors on the identification results are also studied. The results show that the proposed approach can identify effectively the structural condition of free span even with the presence of the measurement error.