Network-synthesis-based identification methodology for passive vibration suppression system design

Abstract

Network synthesis theory originally deals with the problem of realising optimal electrical networks based on the desired system performance. The introduction of a novel mechanical element, the inerter, has completed the electrical-mechanical analogy, allowing network synthesis to be directly applicable to mechanical systems. This method, however, can lead to complex realisations requiring many elements. Compared to electrical systems, the minimal realisation of mechanical networks is much more critical due to practical considerations around alignment, linkages and size. This has motivated our research on developing a network-synthesis-based approach to identify beneficial networks with pre-constrained network complexity. It leads to the design of compact and cost-effective multi-domain vibration suppression systems for a broad range of applications.

First, a series of generic networks, which contain the complete topological information of Immittance-Function-Blocks (IF-Blocks), are proposed to characterise connection possibilities of different types of networks. Desired networks will then be systematically enumerated by assigning a certain number of elements into these IF-Blocks. Since IF-Blocks are two-terminal sub-networks, their immittance functions can be represented as the structural immittances. Thus, combined with their mathematical expressions, the obtained networks can be used to identify optimal vibration suppression systems.

The proposed generic networks are termed as 1PT1NG, IFL, 2PT1NG, GIFN networks. Three of them include one reaction mass, which can be represented through the use of a notional ground as the mass element does not have a direct electrical equivalent. The 1PT1NG network has one physical terminal (PT) and one notional ground (NG) where the reaction mass is connected to; the Immittance-Function Layout (IFL) network represents networks with 2 IF-Blocks and one reaction mass in between; the 2PT1NG network has two PTs and one NG; the Generic-Immittance-Function-Network (GIFN) is the generic network of all possible three-terminal networks. All these networks contain an arbitrary number of transformerless elements connected in series and parallel.

This systematic approach, with its inherent advantages, can be directly used to identify optimal vibration absorbers for various engineering systems. The 1PT1NG network is subjected to offshore wind turbines to suppress their structural vibrations. The IFL and 2PT1NG networks are applied to a three-storey building to reduce its inter-storey drift. The GIFN network is used to design hydraulically interlinked suspensions for vehicles. Simulation results show that the proposed approach is effective in identifying beneficial vibration suppression systems with constrained network complexity compared to the conventional network synthesis methods.

Date of Award	25 Jan 2022
Original language	English
Awarding Institution	University of Bristol
Supervisor	Jason Zheng Jiang (Supervisor) & Simon A Neild (Supervisor)