Input data anomaly identification and evaluation method and system of power prediction model

The invention provides an input data anomaly identification evaluation method and system for a power prediction model, and the method comprises the steps: obtaining historical measured data of meteorological elements of a new energy station, and classifying each meteorological element; solving a probability density function and a cumulative probability density function corresponding to historical measured data of each meteorological element at any moment under each type; setting a confidence interval of the cumulative probability density function of each meteorological element at any moment under each type, and obtaining upper and lower bounds of the normal data range of each meteorological element at any moment under each type; determining envelope lines of normal data corresponding to different moments under various types of meteorological elements, and performing real-time input data online anomaly identification; and based on an anomaly identification result, designing evaluation indexes of an input data