Solution method for maximum expectation value algorithm system of nonlinear equations under Weibull distribution

The invention discloses a solution method for maximum expectation value algorithm system of nonlinear equations under Weibull distribution. The method comprises the following steps: (1) deducing the maximum expectation value algorithm system of nonlinear equations under the Weibull distribution; (2) conducting transformation to nonlinear equations, and forming a function relational expression of F and ; (3) according to the characteristics of site data, finding the actual change rule of F changing with , and determining value range of ; (4) changing a shape parameter directionally within the value range of , and iterating the shape parameter inversely to an F function. Meanwhile, determining a shape parameter value which conforms to accuracy requirements by increasing the accuracy continuously; if F is greater than 0, then continuing to increase ; if F is less than 0, then increasing the accuracy which increases every time, until finding a zero point which meets certain accuracy. By means of the method, the result of parameter estimation under an EM algorithm can be determined fast and accurately; compared with a traditional newton iteration method, the operating speed of a computer increases by ten times or so.