Vapor Compression Cycle Control for Automotive Air Conditioning Systems with a Linear Parameter Varying Approach

TL;DR

This paper develops an LPV control framework for automotive A/C vapor compression cycles, enabling robust output tracking despite parameter variations, using a tensor product transformation and H-infinity control design.

Contribution

It introduces a novel LPV modeling and control approach for automotive A/C systems based on tensor product transformation and H-infinity performance guarantees.

Findings

Successful simulation validation of the control framework.

Enhanced robustness and stability in A/C system control.

Effective handling of parameter variations in the vapor compression cycle.

Abstract

This paper investigates an output tracking problem for the vapor compression cycle in automotive Air Conditioning (A/C) systems using Linear Parameter Varying (LPV) techniques. Stemming from a recently developed first-principle A/C model, Jacobian linearization is first exploited to develop an LPV-based model that is nonlinearly dependent on time-varying system parameters such as evaporator pressure and superheat temperature. To facilitate the control implementation, a Tensor Product (TP) model transformation is applied to transform the LPV-based model to a TP-type convex polytopic model. LPV controllers are then designed to guarantee system stability, robustness and H-infinity performance. Simulations are presented to demonstrate the efficacy of the developed framework.