Investigation of Separate Meter-In Separate Meter-Out Control Strategies
Authors
Winberg, Rikke Helena ; Aarup, Jonas Tomczyk ; Jakobsen, Andreas
Term
4. term
Education
Publication year
2020
Submitted on
2020-05-29
Pages
176
Abstract
Denne afhandling undersøger styringsstrategier for en hydraulisk cylinder med Separate Meter-In Separate Meter-Out (SMISMO), hvor til- og afstrømning kan styres uafhængigt. I første del identificeres input–output‑koblinger, der opfører sig så uafhængigt som muligt, og de vurderes med en Relative Gain Array (RGA)‑analyse – et standardværktøj til at matche indgange og udgange i multivariabelle systemer. RGA’en viser stærk krydskobling ved visse frekvenser og peger på, at den bedste kobling ændrer sig med stempelpositionen. I anden del modelleres, valideres og analyseres et konkret hydraulisk system. Omkring et valgt worst‑case lineariseringspunkt er systemet dårligt konditioneret, dvs. små ændringer kan give store effekter. Der designes og sammenlignes tre MIMO‑controllere til regulering af hastighed og tryk: polplacering, Linear Quadratic Regulator (LQR) og Linear Quadratic Integral (LQI). LQI‑controlleren giver samlet set de bedste resultater med hurtig forbigående respons, præcis stationær nøjagtighed, robusthed over for parametervariationer og god undertrykkelse af forstyrrelser.
This thesis explores control strategies for a hydraulic cylinder using Separate Meter-In Separate Meter-Out (SMISMO), where inflow and outflow are controlled independently. In the first part, we identify input–output pairings that behave as independently as possible and assess them with a Relative Gain Array (RGA) analysis—a standard tool for matching inputs to outputs in multivariable systems. The RGA shows strong cross-coupling at certain frequencies and indicates that the best pairing changes with piston position. In the second part, we model, validate, and analyze a specific hydraulic system. Around a chosen worst-case operating point (linearization), the system is ill-conditioned, meaning small changes can cause large effects. We design and compare three multi-input multi-output (MIMO) controllers to regulate velocity and pressure: Pole Placement, Linear Quadratic Regulator (LQR), and Linear Quadratic Integral (LQI). The LQI controller provides the best overall performance, with fast transient response, accurate steady state, robustness to parameter variations, and good disturbance rejection.
[This abstract was generated with the help of AI]
Documents