Investigation of The Effect of Energy Shaping Via Interconnection and Damping Assignment Passivity Based Control on The Performance of Active Suspension Systems

Abstract

Active suspension systems play a significant part in increasing the pas- senger ride comfort and vehicle ride stability. In addition to the spring and damping elements, active suspension systems have actuator that can inject additional force into the system to improve the performance char- acteristics. The performance of an active suspension system is determined by a feedback control law that governs the output of the actuator as well as the entire system. Design of a control law for active suspension system is a challenging control problem. This is owing to the fact that the perfor- mance objectives of increased ride comfort and stability while preserving suspension deflection limitations are incompatible. To address this issue, numerous control approaches have been developed and investigated in the literature. Interconnection and Damping Assignment Passivity Based Control (IDA- PBC) is a popular passivity based control technique. The control law is designed by shaping the closed-loop energy function of the system and modifying the damping characteristics to a desired level required to im- prove the performance of the system. Modelling of the system in Port- Hamiltonian framework has an advantage when designing a control law using IDA-PBC. This is because the system is represented by its physical attributes in the port-Hamiltonian framework. The system can be repre- sented by its inertia, stiffness, damping coefficients, and energy function, especially in mechanical systems. Therefore, while designing the control law using IDA-PBC, a Port Controlled Hamiltonian (PCH) model is typ- ically used, particularly in mechanical systems where intrinsic physical features can be employed in control design and performance analysis. Quarter-car active suspension system is a Two-Degree-Of-Freedom (2DOF) system representing a corner of a car. Although its capabilities are lim- ited to solely vertical dynamics analysis and control, it can be utilised as a basis for the design and analysis of active suspension system controllers. IDA-PBC control law is designed for a quarter-car model of active sus- pension system. Different cases of the controller emerge after designing a general control law based on the structure of the desired inertia matrix. The choice of desired inertia matrix has a big impact on the dynamics ivof the closed-loop system. As a result, a detailed analysis is carried out by examining the control structure with various inertia matrix scenarios. The results reveal that the design of the control law can be done using a choice of inertia matrices depending on the priority of the performance indices. The performance of a closed-loop system with various control law instances is analysed and compared using simulations and experiments on a bench scale prototype of an active suspension system. On the quarter-car active suspension system, an observer design is devel- oped and implemented. In general, a full-state feedback control law must be implemented to achieve better results in terms of various objectives. However, determining the unsprung mass states in the suspension system necessitates the deployment of many sensors. Furthermore, measuring ve- locities has been a long-standing issue in mechanical systems. A full-state observer is intended for quarter-car active suspension system to overcome these issues. The observer design is done to estimate the states of the PCH system for ease of implementation of IDA-PBC in PCH framework. On the experimental prototype, the performance of observer is evaluated. Furthermore, utilising the state estimates derived from the observer de- sign, full-state IDA-PBC is realised on the experimental configuration. When state estimates are employed to execute the control law, the results reveal that the performance of closed-loop system is comparable to the case with full-state feedback. Half-car active suspension is a Four-Degree-Of-Freedom (4DOF) system that represents one half of a four-wheeler system. It captures the vertical and pitch dynamics of the system. Complex analysis and the solution of several equations are required when designing a control law for a half-car active suspension system. The solution of partial differential equations is difficult to acquire, especially when designing an IDA-PBC control law. The IDA-PBC control law is designed using an algebraic method to over- come this complexity. Two controller scenarios are constructed based on the structure of the inertia matrix, and the performance of system is evaluated using performance indices in simulation in terms of their peak and RMS values, which show good improvement when compared to the uncontrolled system.