Hydroelectric generating set degradation degree prediction method based on EEMD and LSTM

Abstract

The invention discloses a hydroelectric generating set degradation degree prediction method based on EEMD and LSTM. The method comprises the steps: constructing a characteristic parameter health standard model about working condition parameters; inputting the collected current working condition parameters of the hydroelectric generating set into a health standard model to obtain corresponding current characteristic parameter health values; calculating a degradation degree time sequence; decomposing the degradation degree time sequence into a plurality of components by using ensemble empiricalmode decomposition; reconstructing each component according to a preset time step length to obtain a plurality of sequence samples; constructing a long-term and short-term memory network degradation trend prediction model for each sequence sample to obtain a degradation trend prediction component corresponding to each component; and superposing all the degradation trend prediction components to obtain the degradation trend of the hydroelectric generating set. The prediction method can improve the prediction accuracy of the degradation degree of the hydroelectric generating set.

Description

technical field [0001] The invention relates to the maintenance of a hydroelectric unit, in particular to a method for predicting the deterioration degree of a hydroelectric unit based on EEMD and LSTM. Background technique [0002] At present, most hydropower stations still adopt the combination of "after-event maintenance" and "planned maintenance". This maintenance method is to formulate maintenance plans according to the maintenance cycle instead of considering whether the unit is abnormal, which is likely to cause great waste of manpower, material and financial resources during maintenance work, and untimely maintenance. With the development and application of condition monitoring technology for hydropower stations, many hydropower stations have begun to try to switch from planned maintenance to condition-based maintenance. That is to use the operating status of the equipment itself to judge whether the unit needs to be overhauled, and to formulate the overhaul process...

AI Technical Summary

Problems solved by technology

[0003] Purpose of the invention: The purpose of this application is to provide a method for predicting the deterioration degree of hydroelectric units based on EEMD and LSTM, so as to solve the defect of inaccurate prediction of the operating state of hydroelectric units in the prior art

Images