Research of intelligent adjustment method for technical water supply system of pumped storage power station
[Objective]During the actual operation of the pumped storage power station′s water supply system,the regulation of cooling water volume is relatively crude.Historical data cannot be directly used for research.Therefore,considering the unit tem-perature as an intermediate variable,the intelligent regulation of the technical water supply system is achieved by studying the mapping relationship between operating conditions,temperature,and flow rate.[Methods]To address this issue,an improved Long Short-Term Memory(LSTM)model for predicting the temperature of the technical water supply system is first established,with load and other operating condition parameters as input to predict the temperature of the technical water supply object.Subse-quently,the fluent simulation is used to establish the mapping relationship between temperature and flow rate.Finally,the tem-perature threshold obtained by the Pauta criterion is used to determine whether an alarm should be raised and to provide the rec-ommended flow rate.[Results]The result of the case study show that the proposed improved LSTM method improves the predic-tion accuracy by at least 58.45%compared to other method.At the same time,the recommended flow rate obtained from the sim-ulation basically matches the actual operating data of the power station,achieving a water saving of at least 2.33%while ensuring safe operation.[Conclusion]The result indicate that the proposed method can achieve a relatively accurate prediction of the tar-get temperature and,based on this,realize a more rational allocation of water through the built-in simulation model.This has a certain guiding significance for the intelligent regulation of the pumped storage power station′s technical water supply system.
pumped storage power plantstechnical water supply systemimprove Long Short-Term Memory network(LSTM)fluent simulationthreshold alarm
万方数据
维普
