Research On On-line Monitoring Technology Of Vibration And Sound Of Bulb Tubular Turbine

As a kind of renewable clean energy,hydroelectric energy occupies an extremely important posit ion in the history o f energy development in China.The bulb tubular turbines are widely used in hydropower stat ions with mid-low heads and relat ively large flow due to their low invest ment,short construct ion period and small scale o f civil works.However,due to its unique structure,it is easy to make the turbine vibrate during operat ion.The vibrat ion wit hin a certain range will not affect the stable operat ion of the hydropower turbine.However,if a serious failure occurs,it will have important influence on the power supply qualit y,the life o f the turbine and even the who le power system.Therefore,it is o f great significance to carry out the research on the vibrat ion and sound online monitoring device of the bulb tubular turbine.In recent years,small and medium-sized hydropower stat ions are developing towards the trend o f "unattended",which requires the improvement o f the integrated intelligence level o f hydropower units and the realizat ion o f condit io nal m aintenance.Condit io n mo nitoring is the basis o f condit ion maintenance.As a kind o f co mplex nonlinear dynamic system,the operat ion process o f hydropower turbine is affected by the co mplex coupling factors of water,machinery and electricit y.The signals o f the turbine are usually non-stat ionary and non-linear.How to extract the informatio n closely related to the operation state o f the turbine from the co mplex signals of the hydroelectric turbine is the key to realize online monitoring.The monitoring parameters o f the exist ing online monitoring devices for hydropower turbines include swing,vibrat ion,power,vo lt age,temperature,torque and so on.80% o f the hydropower turbine faults can be reflected by vibration signals,sound is another for m of vibrat ion,which contains a lot of informatio n related to the operat ing state of the equipment.When the operating state of the equipment changes,its acoust ic characterist ics will also change.Therefore,this thesis co llects the vibrat ion and sound signals generated during the operat ion of the hydropower turbine,then extracts the characterist ic values of the signals and fuses them to obtain more accurate status informat ion to realize online mo nitoring of hydropower turbines.The on-line monitoring system for vibr at ion and sound o f the bulb tubular turbine designed in this thesis can effect ively mo nitor the running status of the hydropower turbine and has the advantages of simple operation and high accuracy.The main work includes the fo llowing aspects:First ly,the great significance o f condit ion mo nitoring and fault diagnosis for the safe and stable operat ion o f hydroelectric generat ing unit s is briefly described and the significance o f designing an on-line mo nitoring system for vibrat ion and sound o f bulb tubular units is clarified.Then the structure and working principle of the bulb tubular turbine are introduced in detail.According to the different causes o f vibrat ion and sound,the fault mechanism of the hydropower turbine is analyzed and summarized.The unsuitabilit y o f the exist ing large and medium-sized online mo nitoring devices for bulb tubular turbines is analyzed and the ideas of har dware opt imizat ion are proposed.Secondly,in view o f the difficult y o f extract ing weak features of early faults o f hydropower generators,a feature extract ion method based on variat ional modal deco mposit ion(VMD)and singular value deco mposit ion(SVD)is proposed in this thesis.The signal is deco mposed by VMD to obtain several modal co mponents and then the singular value o f the matrix co mposed of modal co mponents is calculated by singular value deco mposit ion,which is regarded as the eigenvalue of the signal.On the basis o f the results o f the experimental data analysis,each modal component obtained by VMD has a center frequency and a narrow band.The original sig nal can be reconstructed by the co mponents obtained fro m VMD and the singular value o f each state has certain data characterist ics,which essent ially reflect t he energy of the maximum frequency.The singular value is used as the eigenvalue to classify and the classificat ion results show that the method proposed in this thesis can accurately extract the weak characteristic signals in the early fault stage,which is o f certain pract icalit y.Thirdly,the so ftware and hardware design of the on-line vibrat ion and sound mo nitoring device for bulb tubular turbine are co mpleted.In terms o f hardware,an online co llect ion device for vibrat ion and sound signals o f the turbine is designed,including the select ion o f sensors,the arrangement of measuring po ints and the design of signal condit io ning circuits.The on-line acquisit io n devices are installed in various parts o f the unit to realize remote co llect ion of sound and vibrat ion signals.In terms of so ftware,the C# programming language is used to develop online monitoring background so ftware,which realizes the Ethernet co mmunication wit h the data acquisit ion system of the lower co mputer,the received vibrat ion and sound signals are analyzed and processed.The software has funct ions such as real-t ime display o f mo nitoring results,database storage and historical data query.Finally,the vibrat ion and sound mo nitoring device o f bulb tubular turbine studied in this thesis is applied to a hydropower stat ion in Hunan Province.Installing the online monitoring system in turbine 3# to co llect a total o f eight sound signals and four vibration signals at the bulb head,bulb body,water distribut ion ring and transformer.The frequency spectrum distribution o f vibrat ion and sound signals o f hydropower units under normal operat ing condit io ns are analyzed and summarized,the fault is simulated by knocking the unit with iron bar.The effect iveness o f the mo nitoring system designed in this thesis is verified through pract ical applicat ion.