Method for predicting power generation of thermal power plant based on multi-variable modeling

Abstract

The invention provides a method for predicting power generation of a thermal power plant based on multi-variable modeling, comprising the steps of: acquiring multiple factor variables and characteristic variables of the thermal power plant on line; performing noise reduction processing; determining main factor variables affecting the characteristic variables by means of gray correlation analysis;establishing a gray model using the characteristic variables and principal component modeling variables satisfying a modeling condition, and correcting the gray model using a neural network; and obtaining a power generation prediction result. By filtering interference information such as noise in data of the thermal power plant, removing unnecessary factor sequence data using gray correlation andthen reducing the dimensionality of the remaining principal factor variables, the dimensionality of the data is reduced, the utilization rate of the factor variables is improved, and the influence ofthe factor variables on the characteristic variables is also considered; the modeling error data of the characteristic variables of the gray GM(1, N) model is trained using the neural network, and then the prediction error of the characteristic variables of the gray GM(1, N) model is corrected, thereby fully utilizing the non-linear function approximation ability of the neural network and the characteristic of being unable to fall into local optimal solution.

Description

technical field [0001] The invention belongs to the technical field of thermal power generation, and in particular relates to a method for predicting the power generation of a thermal power plant based on multivariable modeling. Background technique [0002] The real-time data of the thermal power unit records the operation of the thermal power plant equipment and the operation process of the operator, which can provide decision-making basis for unit operation, maintenance and accident handling, and has positive guidance for improving the production efficiency, fault diagnosis and condition-based maintenance technology of the thermal power plant significance. Due to the influence of various factors such as unit load, ambient temperature, fuel composition and operation mode in the thermal system, there is a coupling relationship between the operating parameters of the equipment, which brings a lot of inconvenience to the actual adjustment. The invention carries out knowledge...

AI Technical Summary

Problems solved by technology

When existing schemes use neural networks for modeling and prediction, all factor variables are generally input to the neural network, and the complexity of the model is high.

Principal component analysis has the function of parameter dimensionality reduction and model simplification, but only considers the interaction between variables, that is, correlation and coupling, and does not consider the influence and contribution of factor variables on feature variables, so it is not enough to rely solely on principal component analysis

Images