Advanced control strategies for mobile hydraulic applications

Davide Cristofori, Purdue University

Abstract

This dissertation presents a unique control method designed to be applicable to mobile hydraulic machines controlled by proportional directional valves. In particular, the proposed control method is targeted to hydraulic machines such as those used in the field including construction (wheel loaders, excavators, and backhoes, etc.), load handling (cranes, reach-stackers, and aerial lift, etc.), agriculture (harvesters, etc.), forestry, and aerospace. For these applications the proposed control method is designed to achieve the following goals: A. Improvement of the machine dynamics by reducing mechanical vibrations of mechanical arms, load, as well as operator seat; B. Reduction of the energy dissipation introduced by current vibration damping methods; C. Reduction of system slowdowns introduced by current vibration damping methods. ^ Two case studies are discussed in this work: 1. Hydraulic crane: the focus is on the vibrations of the mechanical arms and load (goals A and B). 2. Wheel loader: the focus is on the vibrations of the driver's seat and bucket (goals A and C). ^ The proposed controller structure is organized so that the control parameters are adapted with respect to the instantaneous operating point which is identified by means of feedback sensors. The Gain Scheduling technique is used to implement the controller whose set of parameters are function of the specific identified operating point. ^ The optimal set of control parameters for each operating point is determined through the non-model-based controller tuning. The technique determines the optimal set of controller parameters through the optimization of the experimental machine dynamics. The optimization is based on an innovative application of the Extremum Seeking algorithm. The optimal controller parameters are then indexed into the Gain Scheduler. ^ The proposed method does not require the modification of the standard valve controlled machine layout since it only needs for the addition of feedback sensors. The feedback signals are used by the control unit to modify the electric currents to the proportional directional valves and cancel the undesired dynamics of the machine by controlling the actuator motion. In order for the proposed method to be effective, the proportional valve bandwidth must be significantly higher than the frequency of the undesired dynamics. This condition, which is typically true for heavy machineries, is further investigated in the research. ^ The research mostly focuses on the use of pressure feedback. In fact, although the use of position, velocity, or acceleration sensors on the vibrating bodies of the machine would provide a more straightforward measurement of the vibration, they are extremely rare on mobile hydraulic machines where mechanical and environmental stress harm their integrity. A comparison between pressure feedback and acceleration feedback alternatives of the proposed method is investigated with the aim to outline the conditions making one alternative preferable over the other one (for those applications were both alternatives are technically viable in terms of sensors and wiring reliability). ^ A mid-sized hydraulic crane (case study 1) was instrumented at Maha Fluid Power Research Center to study the effectiveness of the proposed control method, its stability and its experimental validation. Up to 30% vibration damping and 40% energy savings were observed for a specific cycle over the standard vibration damping method for this application. The proposed control method was also applied to a wheel loader (case study 2), and up to 70% vibrations attenuation on the bucket and 30% on the driver's cab were found in simulations. These results also served to demonstrate the applicability of the control method to different hydraulic machines. (Abstract shortened by UMI.)^

Degree

Ph.D.

Advisors

Andrea Vacca, Purdue University.

Subject Area

Engineering, Agricultural

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS