Abstract
In this paper, a parameter identification procedure based on the Kalman filter finite element technique for damage mechanics model of jointed rock mass is proposed. In addition that the sensitivity of parameters to be identified is increased with transforming the variables, the convergence of parameter identification process is improved by means of introducing the system noise. An example of jack test by use of generated data has been carried out to demonstrate the applicability and effectiveness of the proposed procedure. It is found that the performance of the proposed procedure is very well even if the value of system noise is chosen in a large range except zero.
