Premium
Optimal feedforward control of hydraulic drive systems with long pipelines
Author(s) -
Pietschnig Christopher,
Marko Lukas,
Kemmetmüller Wolfgang,
Kugi Andreas
Publication year - 2019
Publication title -
pamm
Language(s) - English
Resource type - Journals
ISSN - 1617-7061
DOI - 10.1002/pamm.201900195
Subject(s) - feed forward , control theory (sociology) , hydraulic cylinder , discretization , hydraulic machinery , pipeline transport , actuator , hydraulic motor , optimal control , engineering , control engineering , trajectory , automotive industry , controller (irrigation) , control system , computer science , mechanical engineering , mathematical optimization , control (management) , mathematics , mathematical analysis , electrical engineering , artificial intelligence , agronomy , physics , astronomy , biology , aerospace engineering
Hydraulic drives are used in many technical systems, ranging from automotive and aerospace applications to industrial production systems. They provide high forces and torques at a compact design. The typical setup of hydraulic drive systems comprises a valve which controls a hydraulic cylinder or motor. In many practical applications, the valve is connected to the hydraulic cylinder via hydraulic pipelines, whose length can amount to several meters. Typically, the distributed‐parameter characteristics of these pipelines is not considered in the controller design. For very fast operation scenarios, wave effects can be excited which may cause a significant reduction of the control accuracy or even lead to instability of the closed‐loop system. In this paper, a model‐based control strategy is proposed which allows to systematically account for long pipelines. One main element is the spatial discretization of the partial differential equations of the pipelines by means of the spectral elemental method (SEM), which yields a low‐dimensional, but yet accurate model. Based on this model, an optimal feedforward control strategy is developed, which also enables to take into account actuator constraints. The resulting optimal control problem is efficiently solved on a moving horizon, resulting in a real‐time capable moving horizon trajectory planning strategy. An observer is added to account for uncertain model parameters. The high potential of the proposed approach is demonstrated in simulation by applying the method to the actuation system of a looper in a hot strip rolling mill.