Research And Improved Design Of Dynamic Characteristics Of Umps And Valves Based On Cvclostationaritv And Numerical Simulation

In the fields of shipbuilding industry and process control,pumps and valves are extremely important power equipment and control devices.Their dynamic characteristics are directly related to the reliability of the system.Centrifugal pumps and control valves are one of the most important categories,but there are also difficulties in the analysis of the dynamic characteristics of the two.For the control valve,the control reliability is directly affected by the dynamic characteristics of the flow field of the valve.Therefore,accurate acquisition methods and analysis of the dynamic characteristics of the flow field need further study.For the centrifugal pump,the unsteady nature of its internal flow has a crucial influence on the vibration of the pump foot.The way to analyze the influence of the flow-induced vibration excitation source from the vibration signal causes difficulties in pump vibration research,and the traditional spectrum analysis is difficult to reveal the flow mechanism.Therefore,this paper proposes a method based on cyclostationarity and numerical simulation to obtain and analyze the dynamic characteristic signal of the pumps and valves.On this basis,an improved design for the vibration characteristics of the pump valve is proposed.The main contents of the thesis include the following aspects:(1)The different focuses of the acquisition and analysis of the dynamic characteristics of the control valve and the centrifugal pump are introduced.The flow field analysis and signal processing methods are proposed to reveal the internal flow characteristics of the two important pipeline devices.The dynamic characteristics of the flow field are obtained based on numerical simulation,and analyzed with cyclostationarity.(2)For the characteristics of the first-order periodic signal of the control valve flow field,the time-varying mean value is introduced as a first-order cyclostationary statistic to study this type of signal.Aiming at the flow-induced vibration problem of centrifugal pump,the modulation model of the flow-induced vibration was deduced.The second-order cyclostationarity of the vibration signal of the centrifugal pump foot was proved theoretically.On this basis,the second-order cyclostationarity statistics,such as the spectral correlation density and the cyclic coherence,were introduced to analyze the centrifugal pump vibration signals.(3)The differences between the methods for obtaining the dynamic characteristics of the driven valves and automatic valves are analyzed.Taking the Pulse-Width Modulation(PWM)control valve as an example,the dynamic flow field characteristics acquisition method of the driven valve based on the dynamic mesh technology is proposed.At the same time,taking the spring-loaded pressure regulating valve as an example,the method of acquiring the dynamic characteristics of the flow field of the automatic valve combined with numerical simulation and one-dimensional transient simulation was proposed.The simulation results were compared with the experiments and the validity of the method was verified.The first-order cyclostationarity in the automatic valve signal was analyzed.The effects of damping,inlet pressure variation,and inlet pressure pulsation on the dynamic characteristics of the system were also studied.It is concluded that the system damping and inlet pressure change rate are two important factors that affect the system convergence.The relationship between the inlet pressure pulsation frequency and the natural frequency of the valve system can produce different mechanisms for the valve dynamic characteristics.(4)The second-order cyclostationary statistics are used to analyze the vibration acceleration signal of the centrifugal pump.The results show that the centrifugal pump vibration signal has second-order cyclostationary characteristics,which verifies the reliability of the modulation model of fluid-induced vibration.It is pointed out that the excitation source of centrifugal flow-induced vibration,which is dominated by the blade passing frequency,can increase the amplitudes of the vibration signal of the pump foot at the full-range frequency spectrum.Through the cyclostationary analysis of the vibration signal in the process of changing the rotation speed of the centrifugal pump,the effects of time sampling and angular sampling on the second-order cyclostationarity of the centrifugal pump vibration signal were studied.(5)In order to suppress the excitation force caused by the unsteady flow in the centrifugal pump,the blade thickness distribution of the centrifugal pump is studied through the design concept of the suction surface and pressure surface coupling design.An airfoil blade was designed and compared with the conventional blade and the thickened blade model in pressure pulsation and radial force fluctuation.The unsteady simulation of three kinds of blades in volute and guide vane centrifugal pumps shows that the airfoil blades can effectively reduce the pressure pulsation and radial force fluctuation amplitude in the volute and guide vanes.In this paper,based on unsteady numerical simulation,the methods of obtaining the unsteady flow characteristics in the pumps and valves are proposed.According to the differences of the dynamic characteristics of the control valve and the centrifugal pump,the first-order and second-order cyclostationarity statistics are introduced.The methods are verified by theory and experiments.On this basis,the structures of the centrifugal pump and the control valve are optimized with the aim of suppressing the flow-induced vibration excitation source,and the excitation force was controlled effectively.The results show that the proposed methods for acquisition and analysis of dynamic characteristics of pumps and valves,based on numerical simulation and cyclostationarity,can effectively reveal the unsteady flow mechanism of the flow field,and guide the low vibration design of centrifugal pump and control valve in industrial applications.