Intelligent generator set control model online updating method based on working condition discrimination

Abstract

The invention discloses an intelligent generator set control model online updating method based on working condition discrimination. The method comprises the following steps of intelligent generator set data acquisition, full working condition sample set establishment, generalized predictive control model construction, working condition discrimination and updating decision making. According to the invention, a deep residual shrinkage network is adopted to establish a full-working-condition sample set which can comprise various known operation working conditions of the intelligent generator set; stepped control is adopted for system input increment in a generalized predictive control model, rolling optimization is performed by using a genetic-particle swarm algorithm, the prediction precision is effectively improved, the calculation amount for solving the control model is reduced, and the generalized predictive control can be applied to a rapid sampling control process; the Fisher discrimination method is used for working condition discrimination and decision updating, the current working condition can be accurately judged, online updating work of the control model can be completed conveniently, and the prediction performance and stability of the control model are improved.

Description

technical field [0001] The invention belongs to the technical field of intelligent generator set control, and in particular relates to an online updating method for an intelligent generator set control model based on working condition discrimination. Background technique [0002] The thermal power generation unit undertakes the main task in my country's electric power production, and the safe, stable and efficient operation of thermal power plants is an important issue that needs to be studied and solved in electric power production. Modern thermal power generation units are increasingly developing towards large capacity and high parameters. The increase in power of a single unit and the improvement of initial steam parameters will inevitably lead to further complexity of the automatic adjustment and control system of the steam turbine and boiler, requiring the system to have higher reliability and Automation level, thermal automatic control is becoming more and more importa...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to deal with the problem that it is difficult to ensure safe and stable operation in the application of intelligent control technology in the DCS of thermal power plants in the prior art, and propose an online update method for intelligent generator set control models based on working condition discrimination, which can accurately predict the The future change trend of the control quantity can effectively improve the prediction accuracy and reduce the calculation amount of solving the control model, so that the generalized predictive control can be applied to the rapid sampling control process, and the Fisher discriminant method is used to judge the working condition of the intelligent generator set, which can accurately judge The current working conditions facilitate the online update of the control model, greatly improving the predictive performance and stability of the control model

Images