基于数值仿真的泵阀变刚度弹簧设计与优化

    Design and optimization of variable stiffness spring for pump valve based on numerical simulation

    • 摘要: 目前往复泵锥形阀普遍采用定刚度系数弹簧,但是对于定刚度系数弹簧,单纯的通过调节刚度与预紧力并不能实现提高吸入性能的同时降低阀盘的落座速度与滞后高度,此外传统泵阀设计中,弹簧预紧力与刚度的确定往往是根据经验设计公式,这些公式是在特定条件下近似得到,简化过程中存在很大的计算误差。针对上述不足,该文建立了弹簧变刚度系数条件下阀盘运动规律的数学模型,论证了刚度渐减型弹簧可以提高往复泵的吸入性能,并基于阀盘运动规律的数值仿真模型,建立了以阀盘无冲击为约束条件,阀隙间最大水头损失最小为目标函数的优化模型,采用遗传算法与模式搜索相结合的优化方法对变刚度弹簧参数进行了优化求解,优化结果与传统设计方法确定的定刚度系数弹簧泵阀相比最大水头损失降低了20.85%。该优化设计方法对于指导泵阀弹簧参数设计具有一定的指导意义。

       

      Abstract: Abstract: Automatic poppet valve is a key component of reciprocating pump, which directly influences its performance and lifespan by controlling flow direction of fluid. Many researchers focus mainly on constant stiffness coefficient springs, due to its widespread application in the automatic poppet valve. Previous research already proves that spring plays an important role in the dynamic properties of poppet valve. The main function of spring is to balance inertia force. Studies also suggest that impact velocity and lag height gradually decreases with the increases of stiffness coefficient and preload. However, the increases of stiffness and preload makes head loss rise. Therefore, only by adjusting different combinations of spring stiffness and preload, it is difficult to reduce impact velocity and lag height while reducing head loss. This contradiction restricts the further improvement of reciprocating pump performance. In addition, the determination of the preload and stiffness of spring is often based on empirical formulas in the design process about constant stiffness coefficient spring. To solve these problems, application of variable stiffness coefficient spring on the pump valve was proposed in this paper. In the meantime, a new optimization design method for it was established. The main research contents of this paper included: (i) A mathematical model on the motion properties of valve disc was established under the condition which spring stiffness coefficient was variable. In this model, dynamics differential equation of valve disc motion, continuous flow equation of fluid between valve gaps and initial condition were all considered about. Based on above mathematical model, a computer simulation program was compiled to solve this simulation model by using of Runge-Kutta method. (ii) In order to verify accuracy of simulation model, two contrast tests were carried out. Firstly, lift of valve disc was compared between simulation results and measured results when stroke frequency was 74 min-1. In the process of measurement, the lift of valve disc was measured by lift sensor and pressure in the liquid cylinder was measured by semiconductor pressure transducers. The contrast result showed that the error was less than ±8%. Secondly, by changing parameter of stroke frequency, maximum lift was compared under different stroke frequencies, the error was less than ±10%. In summary, the above comparison showed that simulation model met the requirement of precision. (iii) According to above simulation model, we selected many different combinations of stiffness and preload for the constant stiffness coefficient spring for simulation comparison. After calculation, there was a mutual contradiction between impact velocity, lag height and head loss. In other words, impact velocity and lag height would increase with head loss reduced. Subsequently, in response to above deficiencies, three common types of spring: stiffness decreasing type, stiffness increasing type and stiffness constant type were chosen in the simulation. By comparing with curves of lift, velocity and head loss, it was demonstrated that stiffness decreasing type can improve the suction performance of reciprocating pump, while impact velocity and lag height were not significantly increased. (iv) Based on the numerical simulation model of valve disc motion properties, optimization simulation model of variable stiffness spring parameters was established. In this optimization model, the objective function was that maximum head loss was minimum. The constraint condition was that impact velocity satisfied the theory of no impact. Genetic algorithm and pattern search were used to solve this optimization model. In the end, compared with the constant stiffness coefficient spring, the maximum head loss for variable stiffness spring was reduced by 20.85%. In conclusion, variable stiffness coefficient spring can be used for improving suction performance of reciprocating pump and alleviating the occurrence of cavitation. Besides, this optimization method has a certain guiding significance for guiding design process of spring parameters on the pump valve.

       

    /

    返回文章
    返回