COMPUTATIONAL METHODS FOR DESIGNING NEW PASSIVE FLUID BORNE NOISE SOURCE REDUCTION STRATEGIES IN HYDRAULIC SYSTEMS
Hydraulic systems have many applications in the construction, transportation, and manufacturing sectors. Recent design trends involve systems with higher working pressures and more compact systems, which are advantageous because of power density increase. However, these trends imply higher forces and larger vibration amplitudes while having lesser mass and damping, leading to higher noise levels. Meanwhile, hydraulic machinery started prospecting new applications with tighter noise regulations, a trend which was also pushed by the electrification tendency in several fields of transportation and agriculture. One method to attain noise mitigation is passive-noise canceling techniques have the advantage of not introducing energy to the system. This approach arranges pressure ripple waves in a destructive pattern by projecting a hydraulic circuit's geometry, configuration, and features.
This dissertation aims to predict fluid-borne noise sources and investigate passive noise-canceling solutions for multiple operations conditions targeting to impact many hydraulic systems and a broad range of operating conditions. Primarily a coupled system model strategy that includes a one-dimensional line finite element model is developed. The line model predicts pressure wave generation and propagation. The model features versatility since parameters like line diameter and material can be discretized node by node. Simulations are compared to measured data in a realistic novel hydraulic hybrid transmission for validation.
Subsequently, an extensive numerical investigation is performed by setting fixed parameters along the hydraulic lines' length and comparing several isolated geometric properties in simulation. The developed line model is also used to study the influence of line features such as diameter and extent of the conduit. Cost-effective and simple passive solution solutions such as Quincke tubes (parallel lines), expansion chambers, and closed branches are selected and investigated on simulation. Four target pressure ripples are chosen as indicators for summarizing passive line elements behavior. The frequency-domain behavior of the pressure ripple peaks regarding the line's length is identified and isolated in simulation at the 50-5000Hz frequency spectrum. An experiment test rig is designed to implement these solutions and the experiments show three developed passive elements as practical and effective solutions for reducing fluid borne noise sources. The selected designs yielded noise source attenuation over most of the frequency spectrum measured with piezoelectric pressure variation sensors and accelerometers in different positions in the hydraulic circuit. Sound pressure measurements detected reductions over 3dB in the best cases.
Also, a passive interference approach based on the principle of secondary source flow ripple cancellation was conceptualized, modeled, and implemented in a tandem axial-piston unit. The strategy consists of setting the phase between the two synchronous units to accomplish destructive interference in targeted unit harmonics. Two indexing strategies are investigated first analytically and then on simulation. One of the indexing strategies was implemented in a pre-existent commercial axial-piston tandem unit. Experiment results confirmed effectiveness for the first and third unit’s harmonics, where reductions over 15dB on pressure ripple were measured.
Finally, a fluid-structure interaction based on the poison coupling principle is developed using the method of characteristics. Transfer functions of the pipeline accelerations versus the pressure ripples on lines calculated on simulation and later obtained experimentally to highlight ta critical vibration band from 2000Hz to 3000Hz with high acceleration response.