Water hammer damage caused by abnormal valve closure in long-distance large flow pipeline water conveyance project is one of the most serious hydraulic phenomena.In this paper,an experimental study on the hydraulic characteristics of pipe flow closing valve is carried out by using the water conveyance pipeline experimental platform,and the influence of flow parameters on the water pressure fluctuation and pipe vibration in the pipeline is analyzed.The main contents are as follows:(1)In view of the influence of water pipeline velocity on the hydraulic fluctuation and pipeline vibration at the moment of valve closing,the method of graphic comparison and extracting the signal peak value is used for analysis.The results show that the higher the velocity is,the higher the hydraulic fluctuation peak value caused by valve closing is,and the fluctuation period is prolonged.At the same time,the pressure peak value increases exponentially with the velocity.(2)In view of the correlation between the hydraulic fluctuation in the pipe caused by valve closing and the pipeline vibration,the method of extracting the normalized peak value of the signal and the symbol sequence entropy is used for analysis.It is concluded that with the increase of the flow velocity in the pipeline system,the hydraulic fluctuation peak value and the pipeline vibration in X direction and Y direction both increase first and then decrease.(3)In order to explore the influence of non velocity causes on pressure fluctuation under the condition of emergency valve closing,the orthogonal test method is used to compare and analyze different schemes of valve closing by sections.The results show that the order of the influence of each valve closing control factor on water pressure fluctuation from large to small is the angle after the first section of valve closing,the second section of valve closing time,and the first section of valve closing time.(4)In view of the experimental conditions exceeding the safety limit,the neural network algorithm is used to predict the maximum value and time series signal of hydraulic fluctuation caused by valve closing.It is found that the prediction accuracy of RNN cycle neural network for hydraulic fluctuation time series during valve closing can reach 99%,which shows that it is feasible and has good prediction accuracy to predict the pressure value of hydraulic fluctuation by using cycle neural network. |