Density model solving method based on dynamic updating weighting matrix

The invention discloses a density model solving method based on a dynamic updating weighting matrix, which solves the problems of inaccurate solving and low resolution of a density model in the prior art, and comprises the following specific steps of: distributing an initial weight for each unit body by using a cross correlation coefficient, namely, constructing an initial weighting matrix; solving a first least square problem to obtain the ratio of the density to the weight of each unit body; solving a density vector of the k-th iteration according to the ratio of the density to the weight of each unit body; obtaining a new weight of each unit body by solving a second least square problem, and constructing a new weighting matrix; and replacing the initial weighting matrix with the new weighting matrix, and circularly executing iteration. In the iteration process, monitoring the 2 norm of the difference between the density vectors of two adjacent iterations in real time, if the norm is smaller than a preset t