| 摘要: |
| 在假设系统输出显、潜冷量的相对值在不同的蒸发器入口空气状态下不发生明显变化的前提下，本文针对实验用变速直膨式空调系统建立了稳态人工神经网络（ANN）模型，预测其在不同压缩机、风机转速组合下的系统输出，利用输出显、潜冷量的相对值可以消除室内空气状态对系统输出的影响。通过稳态实验获得数据训练、检测并验证ANN模型预测变速直膨式系统运行特性的准确性，并通过非训练状态点下的稳态实验验证所提出假设与ANN模型的适用性。ANN模型的训练、检测以及验证实验结果的最大误差均小于5%，平均误差均小于3%，表明该稳态ANN模型可以在训练状态点以及非训练状态点较为准确地预测变速直膨式系统的运行特性。 |
| 关键词: 空气调节系统 稳态ANN模型 稳态实验 直接膨胀 |
| DOI： |
|
|
| 基金项目: |
|
| A Steady State ANN Model for the Operational Characteristics of a Variable Speed (VS) Direct Expansion (DX) Air Conditioning (A/C) System |
|
Li Zhao1, Deng Shiming2
|
| (1.School of Environment and Architecture, University of Shanghai for Science and Technology;2.Department of Building Services Engineering, The Hong Kong Polytechnic University, Hong Kong SAR) |
| Abstract: |
| In this study, an ANN model for an experimental VS DX A/C system was developed to predict its outputs with an assumption that the indoor air states will not influence the relative value of system outputs. Steady state experiments under a fixed indoor air state were carried out in this study to obtain the operational characteristics of the system for training and testing the ANN model and the ANN model developed was validated through steady state experiments under the non-training indoor air state. The experimental results showed that the max errors of prediction in training, testing and validating were below 5% and the average errors were below 3%, suggesting that the ANN model developed was capable of predicting the system outputs with satisfactory accuracy at both training and non-training indoor air states. |
| Key words: air-conditioning system steady state ANN model steady state experiment direct expansion |
