Quantum path integral algorithm and its application in magneto-telluric inversion

摘要

It being the fact that quantum algorithms play a good role in extremely fast convergence and excellent global optimization to the nonlinear inversion problems, we present a new inversion method-Quantum Path Integral Algorithm (QPIA) in this paper. Quantum conceptions, such as transverse field and propagator in quantum mechanics, are introduced into this algorithm, and quantum-tunneling effect is applied fully. By composing the objective function and regarding Feynman propagator as the acceptable probability, we accomplish the algorithm. The models tested in one-dimensional magneto telluric inversion showed that the method is more precise, stable convergent and practical method comparing with traditional simulated annealing method.