The singular value truncation method is studied in this work for structural damage identification. The main advantage of the proposed method is that it can obtain the robust and accurate results by the ill-posed linear equations due to measurement noise. The influences of the truncation threshold value and the error level on the calculation result are discussed in detail. In addition, the second calculation using the singular value truncation method by another loading state is proposed in order to further improve the accuracy of the calculation. The numerical experiments show that the proposed method is reasonable and feasible, and has better anti-noise performance than the least squares method. When the noise level is less than 10%, the accurate enough results can be obtained. It was found that the proposed method is promising for structural damage identification.