Experimental study of a novel capacity control algorithm for a multi-evaporator air conditioning system

Abstract

The use of a multi-evaporator air conditioning (MEAC) system is advantageous in terms of installation convenience, high design flexibility, being easy to maintain and commission, better indoor thermal comfort control and higher energy efficiency. While MEAC units worth billions of dollars are sold worldwide, the detailed accounts on compressor capacity control and refrigeration flow distribution amongst evaporators remain unavailable in public domain, mainly due to commercial confidentiality. Limited control algorithms for MEAC systems have been developed based on system simulation, and no experimental-based capacity controller developments and their controllability tests may be identified in open literature. In the study reported in this paper, a novel capacity control algorithm, which imitated On–Off control of a single evaporator air conditioning (A/C) system in each indoor unit of a MEAC system by using variable speed compressor and electronic expansion valves (EEVs), was developed. Controllability tests under various settings for experimentally validating the novel capacity control algorithm were carried out and the control algorithm was further improved based on the experimental results.

Introduction

A multi-evaporator air conditioning (MEAC) system, which is direct expansion (DX) based, consists of an outdoor condensing unit and multiple indoor units. In each indoor unit, there are a DX evaporator, a throttling device such as an electronic expansion valve (EEV) and an indoor air blower. No air duct or chilled water piping is required. The use of MEAC systems is advantageous in terms of installation convenience, high design flexibility, being easy to maintain and commission, better indoor thermal comfort control and higher energy efficiency [1]. While the MEAC technology gradually gains its market place in both Europe and North America, the use of MEAC technology has been common in some of Asian countries. For example, in Japan, MEACs are used in ∼50% of medium sized commercial buildings and in one-third of large commercial buildings [2]. Furthermore, according to a report compiled by a Chinese Market Research Group in Mechanical and Electrical Information, MEAC systems were likely to take 30% market share of air conditioning (A/C) installations in 2009 in Mainland China [3].

Variable speed compressors, instead of constant speed compressors, are standard provisions for MEACs. EEVs are usually used as throttling devices, having a superior operational performance to conventional expansion devices such as thermal expansion valves (TEVs), in terms of stability and capacity control accuracy [4]. Nonetheless, it still remains a challenge to ensure that each indoor unit gets the exact amount of refrigerant flow required for space cooling. Choi and Kim [4] carried out an experimental study on the operating performance of a MEAC system having two indoor units equipped with EEVs under various indoor load conditions. Open-loop system responses obtained during the experiments clearly suggested that the changes in operating parameters of one indoor unit would definitely affect those of the other. However, no system modeling was included and no capacity control algorithms described in the study.

A simulation study on a variable speed compressor MEAC unit having two indoor units was conducted by Park et al. [5]. The steady-state simulation model developed in the study was mainly based on the performance data from the manufacturer of the MEAC unit. With the model developed, a parametric study through varying compressor speed, space cooling load and the opening of EEVs was carried out. However, no experimental validation of the model was reported. Shah et al. [6] reported a study on dynamic modeling and control of MEAC systems. A generic model applicable to commercial applications was introduced, and open loop responses to changes in different system actuators simulated. The simulated open-loop behaviors indicated the effect of cross-coupling among operating variables in different evaporators. However, the model was only validated against a set of previously published experimental data from a two-evaporator heat pump unit [7], thus limiting the effectiveness of the generic model developed. Although a brief discussion on closed loop control strategies was presented, no actual controllability tests of these strategies were carried out and reported.

Lin and Yeh [8] reported a study on designing a feedback controller for a three-evaporator air conditioning (TEAC) system through experimental identification which produced a low-order linear model suitable for controller design. The feedback controller designed was multi-input-multi-output (MIMO) based and possessed a cascade structure for dealing with fast and slow system dynamics. However, the mathematical model obtained through system identification and the controller developed based on the model may not be applicable to other MEAC units having different configurations and/or being operated under other operating conditions, thus not generic.

Chen et al. [9] carried out a simulation study on developing a dynamic model and control algorithms for a TEAC unit equipped with a variable speed compressor and EEVs. A simplified lumped parameter thermodynamic model for the TEAC was developed. A novel control algorithm was proposed and its controllability numerically tested with the TEAC model developed. The suction pressure was taken as a controlled variable to regulate compressor speed while indoor air temperature to regulate the opening of individual EEVs. A self-tuning fuzzy algorithm with a modifying factor was incorporated into the controller. The results of controllability tests showed that the proposed control algorithms could achieve the desired control accuracy for the controlled variables. However, no experimental validation of both the mathematical model and control algorithms was conducted.

A limited number of reported studies on MEAC may be identified, with most existing efforts focusing on introducing basic unit configuration, operation and function [10], [11]. Others were however related to fundamental system modeling or experimental work [4], [5], [12], [13], [14], [15], [16], [17], [18]. On the other hand, although several control algorithms were developed based on the system simulation [6], [8], [9], no experimental-based capacity controller developments and their controllability tests for MEACs may be identified in open literature. Hence, while acknowledging that different capacity control algorithms may co-exist, openly developing an effective and robust capacity control strategy for MEACs to ensure adequate control accuracy for controlled indoor air parameters such as temperature and with adequate experimental controllability tests is urgently desired.

In the study reported in this paper, a novel capacity control algorithm (NCCA), which imitated On–Off control for a single-evaporator air conditioning (SEAC) system in each indoor unit in a MEAC system jointly using a variable speed compressor and EEVs, was developed. Controllability tests under various test conditions for experimentally validating the NCCA were carried out and satisfactory test results obtained. Based on the test results, the control algorithm was further modified and the improved test results of better control accuracy obtained.