Research And Implementation Of The State Evaluation And Parameter Prediction Method Of Pump Unit

Large pumping stations are responsible for flood control,drought relief,water diversion and other important tasks.The interruption of water supply will result in significant economic losses and serious social impacts due to the failure of the pump station or the shutdown of the accident.Therefore,higher requirements for operation monitoring,condition monitoring and maintenance mode of pumping stations are put forward.Compared with traditional troubleshooting and regular maintenance,state maintenance has the advantages of improving the safety of equipment,reducing maintenance time and maintenance cost,and improving the life of equipment.The key to state maintenance in the pumping station is to evaluate the state of the pump set in time and objectively and predict the changing trend of the key parameters of the equipment so as to give reasonable maintenance advice in time.Therefore,combined with the needs of operation and maintenance of Jiangsu pumping station in the eastern route of the South-to-North water diversion project,the state monitoring and remote cooperative diagnosis system of pumping station is studied.This paper is focused on two aspects of comprehensive evaluation and parameter prediction and analysis of pump units,including the state evaluation of the pump unit based on the extension comprehensive evaluation method and the prediction and analysis of the parameters of the pump unit based on the multi algorithm.In chapter 1,the research status and development trend of state maintenance,state prediction,condition monitoring and diagnosis and comprehensive evaluation technology of pumping station are introduced.In chapter 2,combined with the operation,maintenance and management requirements of the pumping station unit in the Jiangsu section of the eastern route of the South-to-North water diversion project,the functional requirements of the state monitoring and analysis of the pump station unit and the remote cooperative diagnosis system are analyzed,and the system architecture,hardware composition and software composition are given.In chapter 3,the comprehensive evaluation method of pump set based on extension comprehensive evaluation is introduced in detail.Firstly,the basic principle and mathematical model of extension comprehensive evaluation method are introduced.Then,the evaluation basis of the pump unit of the South-to-North water diversion project,the division of the evaluation equipment and components,and the selection method of the evaluation parameters are introduced.AHP and entropy weight method are used to calculate the comprehensive weight of evaluation parameters of pump unit.Finally,Matlab is used to verify the extension comprehensive evaluation method of pump unit.In chapter 4,the prediction method of state parameters of pump unit based on Grey GM(1,1)prediction model and BP neural network prediction model is introduced in detail.Firstly,the concept and mathematical model of GM(1,1)and the basic principles of BP neural network prediction are introduced.Then,GM(1,1)prediction model and BP neural network prediction model are applied to predict and analyze the trend of pump unit's state parameters.The parameter prediction algorithm is verified by Matlab simulation.In chapter 5,the function modules of the remote state evaluation and the state parameter prediction analysis of the pump unit are designed and developed by the software of Eclipse,Mysql,Matlab,Tomcat and Navicat.The comprehensive evaluation function of the pump unit is realized,and the trend of the vibration and temperature of the pump unit is predicted.