Power plant statistical data prediction verification method based on QR-GRU

The invention discloses a power plant statistical data prediction and verification method based on QR-GRU, and the method comprises the steps: carrying out the correlation detection of a data type to be predicted and verified, and obtaining a correlation coefficient; a QR-GRU neural network is constructed; according to the correlation coefficient, obtaining data types to be predicted and verified in the historical power plant statistical data and data corresponding to the data types with strong correlation with the data types as training data, and obtaining two network parameters; acquiring statistical data of a power plant in the last month, taking data corresponding to the data type with strong correlation obtained during training as an input value, and inputting the input value into the two QR-GRU neural networks to obtain two predicted values for the data type to be predicted and verified; if the actual value of the data type to be predicted and verified in the current month is located between the two pre