Abstract: To solve parameter inversion problems, differential evolution Markov chain Monte Carlo algorithm is presented. The method is involved through differential evolution algorithm into Markov Chain Monte Carlo algorithm based on Bayesian inference algorithm. It reduces the number of iterations of parameter inversion and has changed the traditional perturbation methods to improve the computational accuracy and stability.

摘要： 针对参数反演问题，提出了微分进化蒙特卡洛算法。此方法是在贝叶斯推理的蒙特卡洛算法的基础上融入微分进化思想。与传统的蒙特卡洛算法相比，此方法有效地缩减了迭代次数，提高了反演精度并具有一定的稳定性。