A progressive computing method based on random sampling survey in seismic buildings damage evaluation

There is an urgent requirement to evaluate local area buildings damage due to a severe earthquake within a shortest time and a sufficient accurately results, so as to assess and distribute rescuing resources. The paper proposed an evaluating method based on progressive computing and random sampling survey to improve the normalized large-scope uniform sampling method in reducing the number of sampling points, shorting survey time, and being convenient to operate. Then a partition random sampling method was put forward furthermore to solve the problem of local dominant characters. Some experiments had been done to numerically simulate the random sampling procedure with the real data from Yutian-Celee earthquake in the year of 2003. The results showed that the random sampling method could achieve the required precision within a relative little time, and could be used in early stage evaluation after earthquake occurred.