Power grid reliability evaluation method based on maximum entropy principle and cross entropy important sampling

The invention discloses a power grid reliability evaluation method based on a maximum entropy principle and cross entropy important sampling. According to the method, the maximum entropy principle is introduced, an optimization model with the maximum information entropy of an important sampling probability density function under the condition that a prior information constraint condition is met is established, an optimization model with the minimum cross entropy of the important sampling probability density function is coupled, and a multi-target planning model for continuous random variable important sampling distribution estimation is constructed; and multiple inferior solutions of the problem,namely a general expression of an important sampling probability density function is obtained through an optimization theory. The continuous random variable important sampling probability density function established by the method is completely based on each moment constraint condition of prior information, and the di