This paper deals with the utilization of the dynamic characteristics of laminar flow in circular pipes for the indirect measurement of flow rates. A discrete-time state space realization of the transmission line dynamics is computed via inverse Laplace transform and an identification and model reduction method based on the singular value decomposition. This dynamic system is used for the computation of the flow rate at one end of a pipe section. Special attention is paid to the identification of the speed of sound and the dimensionless dissipation number of the pipe section, since exact knowledge of these parameters is crucial for the reliability of the measurement results. First, experimental validation results are given in a limited range of operating frequencies between 100 Hz and 2000 Hz. Flow rate variations within have been measured with an uncertainty of at the 95% confidence level. The test fluid was mineral oil.
31 March 2020 in Journal of the Brazilian Society of Mechanical Sciences and Engineering
Theoretical and experimental modal analyses are treated for hydraulic systems modeled by discrete capacities, inductances, resistances, and fluid lines with dynamic laminar flow. Based on an approximate multi-degrees-of-freedom description, it is shown how hydraulic natural frequencies, damping ratios, and mode shapes can be identified from measured frequency response functions between flow rate excitation and pressure response. Experiments are presented for a pipeline system that includes three side branches and an accumulator. In view of practical applications, two different types of servovalve excitation as well as impact hammer excitation are considered. Pressure is measured by 19 sensors throughout the system. Results are compared in terms of frequency response functions between 50 and 850 Hz, the first five hydraulic modes, and weighted auto modal assurance criteria of experimental mode shapes. Out of the tested excitation devices, the servovalve is clearly preferred; if valves cannot be used, the impact hammer offers a reasonable workaround. For a reduced number of five sensors, different sensor arrangements are assessed by the respective weighted auto modal assurance criteria of experimental mode shapes. A theoretical hydraulic modal model provides a similar assessment. The quality of the theoretical model is confirmed by the weighted modal assurance criterion of theoretical and experimental mode shapes from servovalve excitation.
1 May 2006 in Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering
The Galerkin finite method for the simulation of transient, laminar flow in hydraulic pipelines with an interlacing grid has gained considerable popularity because of the relatively good accuracy achieved with a moderate model order. However, the implementation used in a number of software packages has recently been shown not to guarantee model passivity. This paper reviews the concept of using an interlacing grid. Two modifications of the previously published interlacing grid method are proposed and the resulting model is shown to fulfil the passivity requirement in an example where the previously published method fails. The new model also improves the approximation error in terms of the deviation of the frequency response from the theoretical frequency response given by the transcendental transfer functions of the well-known ‘frequency-dependent friction’ model. This improvement is bought at the expense of a slight increase in model order due to additional nodal values for the friction model at the boundary nodes. Furthermore, the new model is given in the form of a descriptor system in order to maintain the sparse structure of the system matrices.
1 December 2009 in Proceedings in Applied Mathematics and Mechanics
A coupled hydraulic-mechanical system has been set up, measured and described by a mathematical model in the frequency domain. This system can be excited hydraulically, by adjusting the flow rate, or mechanically by applying a force. The pressures in the hydraulic subsystem and the accelerations of the mechanical subsystem are measured. From this data, frequency responses between the excitations and the states can be calculated, which are used to validate the mathematical model. In order to identify unknown parameters of the hydraulic subsystem, a variation of certain parameter sets is carried out. They serve as degrees of freedom for a optimization task in model adaptation. The focus here is on quality, effort and suitability of this approach.