Investigation of Separate Meter-In Separate Meter-Out Control Strategies: How does the separate meter-in separate meter-out control system work and what is the limits for separate control of the hydraulic cylinder? What are the consideration associated with chosen the most suitable control strategies for the hydraulic cylinder? What methods are sufficient for decoupling the separate meter-in separate meter-out hydraulic system?

Abstract

The objective for this Master’s Thesis is to investigate and design controllers for the concept of separate meter-in separate meter-out control for a hydraulic cylinder. The separate meter-in separate meter-out concept is investigated under various operation conditions where the velocity of the cylinder is positive and negative, an overrunning and resisitve load force is applied to the system and the result of the analysis is that the concept has some limitations regarding certain operating conditions. The limitations associated with the concept is that cavitation and excessive pressure build up could occur. Cavitation can occur when the velocity is positive, an overrunning load is present and meter-in flow rate control is being used, causing a restriction on the inlet flow, which results in insufficient supply flow to the piston chamber and eventually causes the hydraulic oil to cavitate. A similar scenario where cavitation can occur is when the velocity is negative, a resisitve load is present and meter-in flow rate control is being used, which causes the inlet flow to be restricted and causes insufficient flow to the rod chamber which can lead to caviation in the chamber. Another limit associated with the concept is excessive pressure build up which occur in the scenario where meter-out flow rate control is being used and the velocity of the cylinder is positive and an overrunning load is present and in the scenario where the velocity is negative and a resisitve load is present. In these critical operating conditions the caviation or excessive pressure can result in unwanted behavior for the system and to overcome these suitable control strategies are investigated. The possible control strategies are compared and two chosen control strategies are seen suitable to control the system. The control strategies investigated further are a slave function control where one input signal is dependent on the other input signal which results in the system only being able to control one state. The other control strategy is where the two proportional valves are controlled independently which allowing more flexibility and better performance for the system. Due to the limits for the slave function control the chosen control strategy that is further investigated is when the two proportional valves are controlled independently. When the proportional valves are controlled independently a coupling analysis is conducted to observe the input-output parring when choosing the primary control state as the velocity and the secondary control state as either the piston or the rod pressure. The coupling analysis includes two analysis, the relative gain array and a singular value decomposition. The results of the coupling analysis shows the least input-output parring between the velocity and the rod pressure. The control method where the velocity is controlled by input signal up and the rod pressure is controlled by the input signal ur is further investigated with respect to designing controllers for the control method. To decouple the system a pre-compensator is designed and implemented in order to treat the system as two single-input single-output systems. One single-input single-output system where the velocity is controlled by the input signal up and is independent on the input signal ur and another single-input single-out system where the rod pressure is entirely controlled by the input signal ur. For this control strategy two PI-controllers are designed where the design procedure is that the pressure controller is a factor of 10 faster than the velocity controller to eliminate eventually controller interference. The controllers are implemented and firstly tested on the linear system where the performance for the velocity and rod pressure tracking is acceptable. The controllers are then tested on the non-linear model where the velocity tracking is seen to not be acceptable when the velocity reference is negative. The poorly performance for the velocity tracking in the non-linear model is due to changing model parameters when the velocity changes to negative. To overcome the varying model parameters and the poorly velocity tracking for a negative velocity reference, a set of new PI-controllers are designed based on a linear model conducted for negative velocities, which should be activated when the velocity reference is negative. This is implemented with a switch depending on the reference velocity which switches between the controllers designed for positive and negative velocity. Implementing this control structure improved the overall tracking results of the velocity and thereby it can be concluded that a control method for a separate meter-in separate meter-out setup can be designed.

A master thesis from Aalborg University

[Investigation of Separate Meter-In Separate Meter-Out Control Strategies]