Vehicle drivers are exposed daily to harmful low-frequency vertical vibration over the frequency range of 1-20 Hz. This reduces ride comfort and safety as well as possibly causing long-term harmful effects on human health in the form of lower back pain and driver fatigue. Accordingly, intensive work has been undertaken in this field on active seat suspension systems that have superior performance over a wide frequency range compared with passive and semi-active systems. One of the main features of these systems is the control strategy that is used to generate the demand control force and whilst many control strategies have been investigated in this area; their practical implementation is challenging as they require unavailable or expensive system states. Hence, in this thesis, a novel and cost-effective strategy has been developed that uses measurable and inexpensive displacement and velocity preview information from the vehicle suspension. In addition to these practical advantages, employing a prior knowledge of the disturbance in the control strategy increases the ability of the active seat to react rapidly to disturbances and hence provides a supplementary improvement to the vibration attenuation performance.The potential application of this strategy for an active seat suspension is investigated through both simulation and experimental tests. Firstly, for simplicity, the control force is defined from this suspension preview information based upon a linear control approach, with optimum gains using an integrated simulation model of a linear quarter vehicle model (QvM) and one degree of freedom of seat suspension. These gains are obtained off-line by optimising ride comfort in terms of the vertical Seat Effective Amplitude Transmissibility (SEAT) factor using a genetic algorithm (GA) and considering the physical constraints on both the limited seat suspension travel and actuator force capacity. The experimental tests are performed using a prototype active seat suspension installed on a multi-axis simulation table (MAST), which has been developed to mimic the dynamic motion of the sprung mass of the (QvM) through the principle of hardware-in-loop (HIL) simulation. Moreover, the experimental test rig is used to estimate the characteristics of a passive seat suspension as well as the driver’s body model. The ‘preview’ control strategy is examined according to the ISO 2631-1 standard, in both the frequency and time domains, under a range of operating conditions, including different road profiles and vehicle speeds. Both simulation and experimental results reveal that, in comparison with a passive seat suspension, employing this strategy for the active seat system significantly improves ride comfort, especially over the HBSF range (4-8 Hz). Also, experimental tests demonstrate that combining both the preview information with the vehicle body and seat acceleration feedback states provides further improvement in the vibration attenuation level, achieving up to a 19.5 dB reduction over the HBSF range. The linear control approach cannot always satisfy the physical constraints over a range of operating conditions and thus, to overcome this fault, a fuzzy logic controller (FLC) is selected. Accordingly, two novel and cost-effective FLCs are designed and optimised using the Particle Swarming Optimisation (PSO) algorithm. The feedforward fuzzy logic controller (FF-FLC) uses similar preview information as in the linear control approach, while the feedforward/feedback controller (FFFB-FLC) utilises a combination of both the preview information with seat suspension deflection and velocity feedback states. Once again, the simulation and experimental results confirm the effectiveness of these strategies for attenuating the vertical vibration, especially over the HBSF range, in which the FFFB-FLC provides the best performance as well as the highest robustness level at a variety of different driver weights and vehicle speeds. The application of the preview enhanced controller for an active seat suspension in a full vehicle model has been investigated in the simulation. Accordingly, three FLCs strategies, namely, front-left suspension (FLS-FLC), front-axle (FA-FLC) and four wheels (4W-FLC), have been developed based upon which vehicle suspension or/ suspensions are used to acquire the preview information. The former involves utilising suspension displacement and velocity preview information from the vehicle suspension nearest to the driver’s seat. The FA-FLC uses similar preview information, but from the front-left and front-right suspensions, whilst the 4W-FLC controller employs similar preview information from all the vehicle suspensions. Numerical results show that the proposed controllers are very useful in attenuating the vertical acceleration at the driver’s seat compared with a passive alternative. The 4W-FLC provides the best vibration attenuation performance, independent of the vehicle speed. Finally, to reduce the implementation cost of this controller, a practical alternative has been developed that requires less measured preview information. In conclusion, using the preview information enhanced controller for an active seat suspension provides a practical and cost-effective system that improves ride comfort and reduces driver fatigue.
|Date of Award
|1 Jan 2018
|Jos Darling (Supervisor) & Andrew Hillis (Supervisor)