Elsevier

 应用能源


第 211 卷,2018 年 2 月 1 日,第 1150-1170 页
Applied Energy


空气源热泵机组结霜、除霜过程改进综述

https://doi.org/10.1016/j.apenergy.2017.12.022 获取权利和内容

 强调


  • 回顾了 2000 年至 2017 年发表的缓霜和除霜研究。


  • 对两类12项缓冻措施进行了分类分析。


  • 总结了5种除霜方法和6种改进方法。


  • 提出了除霜操作的启动和终止控制策略。


  • 研究工作中存在的差距被识别并分为5个方面。

 抽象的


空气源热泵(ASHP)机组以其高效节能、环保等优点在全球得到广泛应用。 ASHP 机组室外盘管表面的霜沉积和积聚是不可避免的,并且总是产生显着的负面影响。为了准确预测和控制结霜-除霜循环,应清楚地了解霜、融化霜以及空气-霜界面(移动边界条件)内相互关联的热量、质量和动量传输现象。本文重点介绍了2000年至2017年空气源热泵机组缓霜和除霜研究进展,首先介绍了12项缓霜措施和5种除霜方法,然后介绍了逆循环除霜过程中的6种典型系统优化方法。由此提出了启动和结束除霜操作的替代控制策略。在前人分析的基础上,找出了缓霜和除霜研究工作中存在的空白,并根据本文作者的观点提出了建议。这种围绕整个结霜-除霜循环的全面、系统的回顾可以为学者、研究人员、产品开发人员和政策制定者提供分析工具的概述,并为 ASHP 装置的设计和性能优化提供新的思路。

 缩写


ASHP空气源热泵COP性能系数CSDD压缩机停机除霜DEV除霜均匀度EHD电加热除霜FEV结霜均匀度值HGBD热气旁通除霜HWSD热水喷淋除霜PCM相变材料RCD逆循环除霜RH相对湿度TEST蓄热TEV热力膨胀阀

 关键词


空气源热泵缓霜措施除霜方式逆循环除霜控制策略节能

 一、简介


热泵机组是维持室内空间热舒适度的环保且可靠的手段,并且可以以高运行效率用于空间供暖和制冷。在制冷季节,它将热量从室内空间传递到散热器,就像空调一样。在供暖季节,它从热源中提取热量,并将提取的热能输送到加热的室内空间。从全球来看,世界90%以上的人口居住在适合使用热泵机组进行室内环境控制的地区[1]、[2]、[3]。与使用煤炭或电力的传统空间供暖和/或供热方法相比,研究表明使用热泵机组具有大幅减少温室气体,特别是二氧化碳 2 排放的潜力。随着单位能源成本的上升成为世界关注的焦点,人们对使用热泵技术作为节能手段越来越感兴趣。


空间供暖热泵机组有多种热源可供使用,例如空气、地下水和土壤[4]。在这些热源中,空气和水是空间供暖热泵机组最常见的热源[5]。因此,空气对空气热泵机组、空气对水热泵机组、水对空气热泵机组、水对水热泵机组普遍应用于建筑或工业中。其中,空气源热泵(ASHP)机组安装相对容易且成本低廉,因此多年来一直是应用最广泛的热泵机组类型。根据蒸汽压缩制冷原理,空气源热泵机组使用包含压缩机和冷凝器的制冷系统,在一个地方吸收热量并在另一个地方释放热量。


热泵技术起源于卡诺循环,由卡诺于1824年提出,历史悠久,已有近200年的历史,已相当成熟。冬季空气源热泵机组在制热模式下工作时,周围空气温度为-7~5℃,相对湿度(RH)大于65%,其室外盘管表面常会结霜[6]。积霜会增加制热运行期间的传热阻力和气流通道阻力,从而不利地降低系统性能,甚至导致意外停机。然后,对 ASHP 装置进行了广泛的实验和理论研究,以研究其在结霜和/或除霜条件下的运行性能。近年来关于这个话题的研究非常热门,这也是与冬季空气污染以及相应的政府政策(例如中国的煤改电)有关的。尽管一些评论文章围绕家用热泵[7]、[8]、空气对空气热交换器[9]、[10]、[11]、约束霜方法[12]和除霜方法[13]等主题发表。 ],[14],但没有给出空气源热泵机组的缓霜、除霜和控制策略。另外,上述文章主要集中在2010年之前发表的研究。然而,过去几年已经发表了大量的研究文章。因此,为了为学者、研究人员、产品开发商和政策制定者提供分析工具的概述,本文对 2000 年至 2017 年有关 ASHP 机组结霜/除霜研究的现有文献进行了全面系统的分析。


本次审查工作路线图如图1所示。首次报道了空气源热泵机组防霜措施的研究综述,内容包括改变室外盘管入口环境空气参数、优化室外盘管结构等。缓霜措施。其次回顾了以往关于ASHP机组各种除霜方式的相关研究,特别是逆循环除霜过程中系统运行优化的实验和理论研究。包括对除霜期间 ASHP 装置启动和结束除霜操作的控制策略的审查。最后,确定了进一步广泛研究工作以实现 ASHP 装置更好的结霜/除霜性能的问题。


  1. 下载:下载高分辨率图像 (333KB)

  2. 下载:下载全尺寸图像


图1.当前审查工作的路线图。


2 ASHP机组缓霜措施


如前所述,ASHP 装置中室外盘管表面结霜和积聚是一种不良现象。在一个结霜-除霜循环中,结霜持续时间占运行时间的80%以上,因此缓霜措施探索对于空气源热泵机组的优化具有重要意义。为了提高其运行性能,缓霜措施越来越受到人们的关注。以往关于防冻措施开发的研究分为两种类型,系统外型如图2(a)所示,系统内型如图2(b)所示。


  1. 下载:下载高分辨率图像 (524KB)

  2. 下载:下载全尺寸图像


图2. ASHP 机组的缓霜措施。


2.1.改变室外盘管入口处的环境空气参数


不难理解,空气源热泵机组室外盘管表面结霜与空气源热泵机组运行时的环境空气条件,如气温、相对湿度、风量等密切相关。因此,许多学者致力于改变空气源热泵机组室外大肠杆菌入口处的环境空气参数,定量研究这些参数对缓霜的影响。这些系统外类型研究对于在固定气候环境下精确控制 ASHP 机组的运行性能具有重要意义。


2.1.1.降低入口空气湿度


由于水是霜的来源,研究人员首先研究了结霜性能与降低进气湿度的关系[6],证明了进气湿度在这些参数中的主导作用。同时,在 ASHP 装置上进行了实验研究,无论在入口空气之前放置或不放置固体干燥剂,干燥剂的使用都会显着降低结霜率 [15]。还使用固体干燥剂,提出了一种新型无霜 ASHP 装置,并对其缓霜特性进行了实验和数值研究 [16]、[17]、[18]。采用这种方法,不仅降低了空气湿度,而且通过吸收固体干燥剂的热量而提高了空气温度,从而在一定的结霜时间内室外盘管表面结霜较少。因此,替代除湿方法和材料引起广泛关注。例如,液体干燥剂被用于对 ASHP 装置的入口空气进行除湿,因为它具有低气压降、空气净化效果和较低再生温度等诸多优点。表1列出了2000年至2017年精选的10个关于用于空气源热泵机组防冻的固体和液体干燥剂的典型研究。显然,这个话题至今仍然很受欢迎,并且同时考虑了实验方法和数值方法。缓霜特性也成为检验除湿性能的标准。对于 ASHP 机组来说,降低进气湿度是延缓结霜的最基本措施。然而,由于增加干燥剂设施,整个系统的初始成本和占地面积会增加。干燥剂再生消耗较多的热能,从而同时增加运行成本。 该措施的实际应用受到这些经济问题的限制。


表 1. ASHP 装置中用于防霜的固体和液体干燥剂的研究(2000-2017 年)[19]、[20]、[21]、[22]、[23]、[24]。

ItemYearAuthorTypea 使用干燥剂Results
12017 王等人。 [17]N 固体(硅胶)
R22、R407C和R134a的无霜时间分别为29、34和35分钟。在给定环境温度 -10°C 和相对湿度 85% 的情况下,R134a 的平均性能系数 (COP) 分别为 3.3 和 8.6%,高于其他制冷剂
22017 苏和张 [19]N
液体干燥剂(膜基、氯化锂溶液)

新型系统的 COP sen 和 COP tot 比传统除霜系统的 COP reverse 高至少 37.7% 和 64.3%分别分析参数的变化范围。回热器对空气的加湿可以满足环境空气在0℃、70%RH以上时室内热舒适的需要
32017 王等人。 [18]N 固体(硅胶)
在相对湿度为 80% 的情况下,当环境温度从 -10°C 升至 0°C 时,系统平均 COP 增加了 56.2%。 0℃时相对湿度从75%增加到85%时下降6.7%
42015 王等人。 [16]E 固体(硅胶)
系统在环境温度-3℃、0℃、3℃、85%RH条件下,制热模式下可保持蒸发器无霜32、34、36min
52015 张等人。 [20] 电子与网络
液体(氯化锂溶液)

干燥毡的风速(V a )、温度(T b )和含水量(C w )影响平均总传质系数(K total )。 K total 随着 V a 的增加而增加,但随着 T b 的增加而减少,因为 T b 与dC w /dX a
62014 王等人。 [21]E 固体(硅胶)
除湿后空气相对湿度可降至52%,室外热交换器在0℃、相对湿度80%的条件下可保持34分钟无霜
72014 江等人。 [22]E 液体(甘油)
如果溶液喷射子系统连续运行,室外盘管表面永远不会结霜
82012 张等人。 [23]N 固体(未给出)
无霜 ASHP 热水器系统的 COP 比除霜 ASHP 热水器系统可提高 5-30%
92010 张等人。 [24]E
液体(氯化锂溶液)

可延缓结霜,混合动力系统夏季和冬季COP分别提高约20%和100%
102005 王等人。 [15]E
固体(沸石板和活性炭)

解决了结霜问题,提高了热泵机组冬季性能
a


N:数值研究; E:实验研究。


2.1.2.预热进气


预热室外盘管的进气是一种简单但有效的技术,可减少或防止 ASHP 装置结霜。某些加热元件可以放置在进气风道中,以便当室外空气温度降至霜点以下时,加热元件被激活。为了防止结霜,室外盘管上游的入口空气温度必须始终高于结霜点 [25]。 Liu等人首先将热回收技术与这种缓霜方法结合起来。 2007年[26]。排出的室内空气和环境空气在进入室外盘管之前混合,从而延长了结霜时间并降低了结霜速度。实验性 ASHP 装置以及热回收设施。然后,在 ASHP 机组室外盘管上游安装电加热器,对缓霜效果进行实验和定量研究[27]、[28]。结果表明,在室外空气温度低于2℃/1℃(干球温度/湿球温度)时,开启电加热器加热进风,制热能力提高了38.0%,COP提高了57.0%。 - 灯泡温度)。然而,在极寒地区预热进气的缺点是预热能耗较高[29]。对不同防霜措施的比较表明,在室外空气温度长期处于-54至10°C的极低水平的地区,预热入口空气并不经济[30]。因此,为了提高空气源热泵机组的经济性,预热进气的能量应该是废热,例如从排出的室内空气或废水中回收的热量。


2.1.3.增加室外盘管的入口气流速率


增加室外盘管的入口气流速率是另一种系统外型防霜措施。 Da Silva 等人对考虑风扇特性的翅片管热蒸发器上的霜形成进行了实验研究。 2011年[31],证明气流速度降低是蒸发器能力下降的主导因素。进一步建议在结霜条件下风扇-蒸发器应被视为耦合系统。为了预测室外盘管的性能(考虑到霜生长导致的气流减少),Ye 和 Lee 于 2013 年开发并验证了一个数值模型 [32]。假设霜层均匀分布在换热器表面。模拟的传热率和积霜质量与实验数据分别吻合7%和9%。实际中,为了获得更好的缓霜效果,可以将上述空气参数一起改变。不同环境空气参数的变化会影响缓霜率。然而,同年,Moallem 等人进行了一项关于结霜对微通道蒸发器不利影响的实验研究。 2013年[33],表明蒸发器的空气面速度对霜生长速率的影响较小。此外,风扇电力成本的增加和噪音水平的增加也难以避免。更多学者关注其与后续其他防霜措施的结合。


2.2.使用附加设备进行霜冻破坏


另一种系统外类型的防霜措施是使用附加设备(例如超声波振动或空气喷射技术)进行除霜。与上述缓霜措施不同的是,此类措施不使用热量。仅使用机械能来破坏霜的形成和生长过程。


2.2.1.超声波振动技术


Yan等人首先将超声波技术用于缓霜。 2003年[34]和Li等人。 2010年[35]。据报道,由于超声波的作用,平坦表面上的霜形成过程被显着抑制。冷冻水滴形状的比较如图3所示。显然,有超声波的图中的水滴比没有超声波的图中的水滴要小。经定量分析,超声波作用下霜冻覆盖率均小于52%,而无超声波作用时霜冻覆盖率超过65%。冰层上的霜晶和霜枝可以被有效地破碎和去除[36],[37]。 Tan等人利用间歇超声波振动对翅片管蒸发器的除霜性能进行了实验研究。 2015年[38]表明间歇式超声波振动可以有效去除翅片表面积霜。 ASHP机组除霜能耗降低约3.14%~5.46%,制热量提高2.2%~9.03%,COP提高6.51%~15.33%。据悉,室内侧热舒适度明显提高。然而,效果有限,因为超声波振动无法去除鳍上的基本冰层。认为超声波抑霜的机理主要是高频超声波机械振动使霜晶和霜层破碎,霜在重力作用下脱落,而不是超声波空化效应或热效应。这些研究促进了该领域的发展,但超声振动技术因其初始成本高、控制系统复杂而限制了其实际应用。


  1. 下载:下载高分辨率图像 (736KB)

  2. 下载:下载全尺寸图像


图3 冷冻水滴形状比较[35]

 2.2.2.空气喷射技术


喷气技术是另一种有效的缓霜措施,它不需要热能来融化霜[30]。首先在浸入气固流化床中的水平单排冷却管阵列上进行了尝试。对冷却管的传热和除霜特性进行了实验研究,流化床产生的气固颗粒撞击射流有效地去除了管表面的霜层。经验证,在进风温度-7℃、进风相对湿度80%、管表面温度-17℃的条件下,冷却管可无霜运行。 Fei和Mao[39]利用压缩空气实验研究了缓霜效果,表明该措施可以及时除霜。这意味着该措施可以应用于有压缩空气的地方。此外,Sonobe 等人还进行了热交换器的防霜措施,其中通过空气射流加速固体颗粒撞击热交换器表面。 2015年[40]。基于空气喷射技术,这项研究的动机是开发用于高超音速飞机发动机的低温热交换器,并表明该措施比除霜更有效。然而,与超声波振动技术一样,关于喷气技术缓霜的文献报道很少。除了初始成本和运行成本较高、使用不便等缺点外,空气源热泵机组的缓霜效果不成熟或不够好也限制了其实际应用。


2.3.优化室外盘管结构


除了前面从系统外部角度描述的这些缓霜措施外,从系统内部角度来看,通过优化室外盘管的结构,还制定了一些延迟或防止结霜的措施,以减轻负面影响。结霜对 ASHP 装置运行性能的影响。


2.3.1.调整翅片和管的几何形状


Young 和 Watters 等人首先建议使用具有更宽翅片空间的室外盘管来减缓霜的生长。 2002年[41]。随后,Yan等人对此进行了实验研究。 2003 年 [42] 以及 Sommers 和 Jacobi 2005 年 [43]。据报道,在空气侧雷诺数为500至1300时,通过调整翅片结构产生涡流,空气侧热阻降低了35%至42%[43]。这意味着结霜过程被有效延迟。如图 4(a) 所示,2006 年,Yang 等人。文献[44]提出了结霜工况下家用冰箱翅片管换热器设计参数的最佳值,以提高其热性能,延长其运行时间,从而延缓结霜。翅片和管几何形状优化后,平均传热率和运行时间分别提高了6.3%和12.9%。如图 4(b) 所示,2010 年,Lee 等人。 [45]测量并分析了结霜条件下不同翅片节距、管排数和管排列方式的扁平翅片管换热器的空气侧换热特性,表明翅片节距和管排列方式对气流减少的影响较大,从而起到缓霜作用。最近,Park等人对不等百叶节距的百叶窗翅片的结霜行为和热性能进行了研究。 [46],证明当使用不等百叶节距设计时,前侧百叶窗之间的空间的霜冻阻塞被延迟,并且热性能提高了21%。如图4(c)所示,可以明显看出,等百叶间距上的霜积累量比不等百叶间距上的霜积累多。 此外,百叶窗间距从空气入口区域到重定向区域连续减小的设计提供了更均匀的霜生长和改进的热性能。由于上述的缓凝作用,优化翅片和管的几何形状也越来越受到工程师的关注。


  1. 下载:下载高分辨率图像 (628KB)

  2. 下载:下载全尺寸图像


图 4. 翅片和管几何形状优化后的结霜行为 [44]、[45]、[46]


2.3.2.调节翅片类型


翅片类型调整也被用作防霜措施。 2005 年,Yan 等人。 [47]通过实验研究了带有平板翅片、单面百叶窗翅片和重定向百叶窗翅片的磨砂翅片管换热器的运行性能。在其他条件相同的情况下,带改向百叶窗翅片的换热器结霜量最大。董等人。 [48] 通过在住宅 ASHP 装置的室外盘管中使用三种翅片类型,通过实验比较了周期性结霜-除霜性能的影响。具有扁平翅片的室外盘管在 ASHP 装置的定期结霜/除霜循环中表现出最佳的热性能,其次是具有波状翅片和百叶翅片的室外盘管。张和Hrnjak [49]在干、湿和霜条件下实验研究了三种具有百叶窗翅片几何形状的换热器:(1)带有挤压扁管的平行流蛇形翅片,(2)带有挤压扁管的平行流平行翅片,和(3)圆管波片翅片。如图所示,在结霜工况下,圆管波板翅片换热器由于其表面积最大,可以使用最长的时间。带挤压扁管的平行流平行翅片换热器空气侧压降的增加是最低的。尽管某些类型的翅片可用于缓霜,但翅片类型的总数以及缓霜效果是有限的。最后,不可避免的是,越来越多的研究人员将注意力转向翅片表面涂层处理研究以减缓霜冻。


2.3.3.翅片表面涂层处理


有研究报道了室外盘管翅片表面涂层处理对结霜和除霜性能的影响。 2000 年,Okoroafor 和 Newborough [50] 发现,通过交联亲水性聚合物涂层,可以显着减少暴露于温暖潮湿气流的冷表面上的霜生长。与使用未涂层的金属表面相比,霜厚度减少了 10-30%。一年后,Wu 和 Webb [51] 研究了亲水和疏水表面的结霜和除霜过程,结果表明亲水涂层更好,可以减少结霜和滞留水,如图 5(a) 所示。 2011年,蔡等人。 [52]实验研究了普通铜表面、疏水涂层(车蜡涂层)表面和吸湿涂层(甘油涂层)表面的磨砂情况,如图5(b)所示。在霜形成初期,同时使用疏水涂层和吸湿涂层均可抑制霜的生长,且亲水涂层的厚度与霜的缓凝效果成正比。 Jhee 等人也证明了类似的结果。 [53]和刘等人。 [54]。据定量报道,使用表面亲水性聚合物涂料可延缓霜形成达3小时,霜厚度减少至少40%,且涂层表面形成的霜层疏松且易于去除[54] 。


除了常规的尺度研究外,学者们还试图寻找结霜/除霜过程中表面处理的机理。通常,在 2013 年,Chen 等人。 [55]报道了一种分层表面,它可以抑制液滴间冻结波传播并有效去除霜,如图 5(c)所示。结果表明,性能的增强主要是由于分层表面微尺度边缘效应的激活,增加了冰桥的能量势垒,并在除霜过程中产生液体润滑。他们相信,利用表面形态在两个相反的相变过程中实现卓越性能的概念可能为各种应用的新型材料的开发提供新的思路。如表 2 所示,2000-2017 年广泛发表了规则/纳米尺度表面处理防霜机理研究。


  1. 下载:下载高分辨率图像 (1MB)

  2. 下载:下载全尺寸图像


图5.不同表面上的结霜和除霜图像[51],[52],[55]


表2 表面处理防冻机理研究(2000-2017年)[56],[57],[58],[59],[60],[61],[62],[63],[64 ]、[65]、[66]、[67]、[68]、[69]、[70]、[71]、[72]、[73]、[74]。

ItemYearAuthorCountryScaleaResult
12017 左等人。 [56]ChinaN-P
在-10°C 下,所制备的超疏水 ZnO 表面上的霜形成被有效延迟了 140 分钟以上
22017
达席尔瓦等人。 [57]
BrazilR-HEX
蒸发器在-5℃下进行测试,结霜结构更加致密,与蒸发器维持在-10℃时观察到的霜结构有很大不同,后者呈现树枝状和针状冰晶
32017 吴等人。 [58]ChinaR-F
交叉沟槽表面积霜最少,平行沟槽表面融水排水性能最佳
42016 森谷等人。 [59]JapanN-P
与未涂层的表面相比,氟碳基涂层将霜的形成延迟到-6°C
52016 索默斯等人。 [60]USAR-F
亲水表面霜密度比基线表面高20%至-26%。在疏水表面观察到霜密度降低了 37-41%
62015 赵等人。 [61]ChinaN-P
报道了一种基于阳极氧化铝棒状纳米孔可控制备的铝基凝结水微滴自推进功能薄膜,具有自清洁、防霜、防结露功能
72015 金等人。 [62] 韩国R-F
当制冷剂温度为-10℃或-12℃时,超疏水表面缓霜效果显着增强;当制冷剂温度为-8℃时,水接触角大于150°时超疏水效果减弱
82015 梁等人。 [63]ChinaR-F
亲水翅片、裸翅片、疏水翅片和超疏水翅片表面融霜所需时间分别为36s、25s、23s和22s
92015 王等人。 [64]ChinaR-HEX
超疏水换热器结霜厚度和质量分别比裸换热器减少17.1%和28.8%
102014
巴拉蒂达桑等人。 [65]
IndiaR-F
亲水性表面涂层比疏水性有机硅涂层表现出更高的冰粘附强度。超疏水涂层表现出最佳性能
112014 李等人。 [66]ChinaR-HEX
滞留经历了水、水和冰3个阶段,然后主要是冰。第三阶段出现“多年冻土区”,占换热面积的20%
122013 陈等人。 [55] 香港N-P
据报道,分层表面允许抑制液滴间冻结波传播和有效除霜。性能的增强主要是由于分层表面中微尺度边缘效应的激活
132013 米尔科维奇等人。 [67]USAN-P
显着增强了冷凝传热,并有望采用低成本且可扩展的方法来提高大气集水和除湿等应用的效率
142013 金等人。 [68] 韩国R-HEX
在疏水换热器处没有观察到前沿效应,并且由于霜阻滞,疏水单元在重复的结霜和除霜实验循环中表现出最高的总传热率
152012 莫阿勒姆等人。 [69]USAR-HEX
霜在霜的类型、外观和图案上有明显的差异。微通道盘管上的疏水和亲水涂层对结霜条件下的传热能力影响高达 15%
162009
库利尼奇和法尔扎内 [70]
CanadaR-F
在具有低润湿滞后的超疏水表面上,观察到冰粘附强度比裸露抛光铝低约 5.7 倍
172009 黄等人。 [71]ChinaR-HEX
有涂层的亲水翅片在整个测试过程中没有结霜,而无涂层的翅片则完全被致密厚实的霜层覆盖
182009 博雷科等人。 [72]USAN-P
据报道,在没有任何外力的情况下,超疏水表面上自发发生连续滴状冷凝
192002 杰伊等人。 [73] 韩国R-HEX
亲水处理主要影响结霜行为,而疏水处理则影响除霜行为
202000 奥科罗阿福等人。 [74]UKR-HEX
交联亲水聚合物涂层是最大限度减少暴露于潮湿空气的冷表面上霜生长的潜在途径
a


N-P:纳米级板材; R-F:常规比例的翅片; R-HEX:常规规模的热交换器。


虽然通过调整翅片间距和排列方式来优化室外盘管的结构,但改变翅片类型和翅片表面涂层处理会增加室外盘管的设计和制造工作的难度,同时也会提高空气源热泵机组的初始成本,这些措施可以有效延缓霜冻。翅片表面处理越来越受到人们的关注,许多学者致力于探索翅片表面处理的机理,这是可以理解的。


2.4.系统调整优化


空气源热泵机组也有一些系统外式缓霜措施,是通过调整和优化系统结构而采取的。在这些措施中,缓霜消耗的所有能量都来自于内部制冷剂传递到管子和翅片的热量。


2.4.1.蒸气注射技术


蒸汽喷射技术自 1979 年以来一直用于室内空调,但最近其在热泵机组中的应用受到了更多关注,因为该技术可以在寒冷气候下延缓结霜 [75]。 2002 年,Zhnder 等人。 [76]在入口空气温度为-7°C的情况下测试了空气-水蒸气喷射热泵机组,报告称与没有使用的测试相比,热输出分别增加了28%,COP提高了15%。注射。同样在 -7 °C 的环境温度下,Nguyen 等人。 [77] 使用 R-407C 评估了闪蒸罐蒸气喷射循环和过冷器蒸气喷射循环的热性能,报告其加热 COP 分别比单级循环高 24% 和 10%。与此同时,在-20至-15°C的室外温度下,Shao等人。 [78]的结论是,蒸汽喷射热泵机组可以提供足够的加热能力。在-25°C的环境温度下,Ma和Zhao[79]通过实验研究了带有闪蒸罐和涡旋压缩机的ASHP装置的运行性能,表明ASHP装置比带有过冷器的系统更高效在-25至-7°C。这意味着该技术对于 ASHP 装置的防霜非常有用。


2.4.2.两阶段技术


与蒸气注射技术一样,两阶段技术也引起了许多研究人员的关注。王等人。实验研究了双级热泵供暖系统,该系统将 ASHP 机组和水源热泵机组耦合起来 [80]。结果显示,测试期间平均能效比最高为3.2,最低为2.5,室内平均温度超过19.5℃,最低为18℃。与传统的空气源热泵机组相比,耦合系统的运行特性得到了很大的改善,表明该系统在寒冷地区具有巨大的应用潜力。李等人。 [81]提出并实验测试了一种新型无霜空气源热泵系统,表明该新型系统在冬季比传统空气源热泵机组运行效率更高,并且无需定期除霜。 Heo 等人报道,在-15°C 的环境温度下。 2010年[82]发现两级注汽循环的COP和加热能力分别提高了10%和25%。 Wang等人发现,在-17.8°C的环境温度下。 [83] 两级热泵系统的 COP 最大提高了 23%。 2008年,Bertsch和Groll[84]测试了专门设计的R410A两级ASHP机组,在-30℃的环境温度下观察到加热COP为2.1。这意味着耦合系统中的这项技术可以在寒冷气候下延缓霜冻。显然,该技术使系统比使用蒸汽喷射技术更加复杂。

2.4.3. Adding outside heating source

It is easy to understand that adding outside heat source could improve the system operating performance at frosting condition. It was qualitatively demonstrated by Masaji that the performance of an ASHP unit with a kerosene fired burner placed either in its indoor unit or under its outdoor coil could be improved at low ambient temperatures [30]. Using the same method, Mei et al. [85] reported that the heating capacity of an ASHP unit could be increased, and the frost accumulation on its outdoor coil retarded by heating up the liquid refrigerant in its accumulator. By heating liquid refrigerant, the frequency of defrosting cycles was shown to be reduced by a factor of 5 in Knoxville, Tennessee, USA, and indoor supply air temperature raised by 2–3 °C because of the increased compressor suction pressure. Different from preheating inlet air of outdoor coil, the measure of adding outside heating source takes outside thermal energy by refrigerant. However, they both consumed a lot of energy in heating surround cold air. Therefore, to improve the economy of the ASHP units, the heating source should be waste heat, such as heat recovered from exhausted indoor air or waste water. Finally, this type of frost retarding measure is also limited in application, due to its disadvantages of high running cost and inconvenience.

2.4.4. Adjusting refrigerant distribution

For an outdoor coil used in an ASHP unit, on its refrigerant side, multiple parallel circuits are commonly used for minimized refrigerant pressure drop and enhanced heat transfer efficiency. As shown in Fig. 6(a), the frost accumulations are always uneven on the surface of a multi-circuit heat exchanger [86], [87], [88]. To clearly describe this phenomenon, it was defined as uneven frosting [89]. Further, to quantitatively evaluate the phenomenon of uneven frosting, frosting evenness value (FEV) was defined as the ratio of the minimum frost accumulation among three circuits to the maximum one [[89], [90]. Recently, in order to investigate the relationship of FEV and frost retarding effect, an experimental study on even frosting performance of an ASHP unit with a three-circuit outdoor coil was conducted [90]. System frosting performance when frost accumulated at different FEVs was compared, with airside of surface conditions of outdoor coil shown in Fig. 6(b). As indicated, when the FEV was increased from 75.7% to 90.5%, the COP could be increased from 4.10 to 4.26 at a 3600 s frosting process, and increased from 3.18 to 4.00 at its last 600 s. That means supplying the same thermal energy by the system, less frost accumulated on the surface of outdoor coil when the FEV is higher, and refrigerant distribution adjustment could retard frost for an ASHP unit having a multi-circuit outdoor coil.

  1. Download : Download high-res image (2MB)
  2. Download : Download full-size image

Fig. 6. Uneven frosting and different FEVs [86], [87], [88], [90]

All the energy consumed for frost retarding comes from refrigerant at the inside of system, although the energy is inputted into the system at different positions or components. It is essentially different from the outside of system type measures, such as increasing inlet air temperature or decreasing inlet air RH. In these frost retarding measures, inlet air works as the energy medium. As listed in Table 3, evaluation results of 12 frost retarding measures are given. Nearly all the measures would increase the initial cost and/or the running cost, except two measures of adjusting fin and tube geometry, and fin type. System adjustment and optimization would increase the system complexity, and decrease system stability. Additional thermal energy is in need for the measures of preheating inlet air and adding outside heat source. Four measures need enlarge the floor space for adding equipment, such as desiccant bed for reducing inlet air humidity, and air jet or ultrasonic vibrations facilities to destroy frost formation or growth. Among all the measures, reducing inlet air humidity and preheating inlet air have the best frost retarding effect. Considering the comprehensive values of listed measures, preheating inlet air with waste heat and coating treatment on fin surface with new materials are highly recommended to be further explored.

Table 3. Evaluation results of 12 frost retarding measures.

Frost retarding measureInitial costaRunning costaSystem complexityaFloor spaceaAdditional thermal energybSystem stabilityaDefrosting effectcComprehensive value c
Outside of system1Reducing inlet air humidity×332
2Preheating inlet air333
3Increasing inlet airflow rate×222
4Ultrasonic vibration technique×222
5Air jet technique×122

Inside of system6Adjusting fin and tube geometry×311
7Adjusting fin types×311
8Coating treatment on fin surface×323
9Vapor-injection technique×111
10Two-stage technique×122
11Adding outside heat source221
12Adjusting refrigerant distribution×212
a

↑: Increased; →: Unchanged.

b

√: In need; ×: Not in need.

c

3:The best; 1: The worst.

3. Defrosting methods for ASHP units

As discussed earlier, the presence of frost on tube surface of the outdoor coil in an ASHP unit would deteriorate its operating performance, energy efficiency, reliability and life span. While the use of frost retarding measures can delay frost formation or growth, these measures can be expensive or consume additional energy, and frost that is present after delaying would have to be removed. Therefore, periodic defrosting becomes necessary for guaranteeing the satisfactory operation of ASHP units. To distinguish frosting retarding and defrosting, some of their differences are summarized in Table 4.

Table 4. Differences of frost retarding and defrosting.

ItemDifferent aspectsFrost retardingDefrosting
1Operation initiationDuring frost formingAfter frost accumulated
2Operation terminationNeverPeriodic
3During operationHeating mode continuedHeating mode stopped
4Operation effectsNot all frost removedAll the frost removed
5Evaluation indexCOPDefrosting efficiency; Defrosting duration
6Duration in a frosting-defrosting cycleMore than 80% timeLess than 20% time

In this Section, defrosting methods reviewed are all based on the assumption, normal frosting, that no frost retarding measures are implemented during heating/frosting. Generally speaking, there are 5 types of defrosting methods, including: (1) compressor shutdown defrosting, (2) electric heating defrosting, (3) hot water spraying defrosting, (4) hot gas bypass defrosting, and (5) reverse cycle defrosting. In order to carefully distinguish the 5 defrosting methods, diagrams of system at frosting and different defrosting operations are illustrated in Fig. 7.

  1. Download : Download high-res image (747KB)
  2. Download : Download full-size image

Fig. 7. Operation conditions of 5 defrosting methods.

3.1. Compressor shutdown defrosting (CSDD)

For compressor shutdown defrosting method, ambient air is used as the heat source of defrosting. Therefore, it is normally applied to where ambient air temperature is not lower than 1 °C. When defrosting is needed, as shown in Fig. 7(b), the compressor is shutdown but outdoor coil air fan continues to move the ambient air at >1 °C to pass through the outdoor coil to melt the frost. To test this defrosting method, Ameen et al. experimentally investigated the defrosting performance of an ASHP unit using warm air under controlled conditions in an air-conditioned wind tunnel [14]. The effectiveness of compressor shutdown defrosting method was also demonstrated by Shang [91] in 2009, by experimentally investigating the effect of pre-start fans on defrosting performance of an ASHP unit. At first, advantages of low initial cost, no reconstruction work, and easy to control make this method widely applied. Then, since the energy comes from ambient air, this defrosting method consumes a long time. Defrosting duration would be prolonged, as well as the indoor thermal comfort degraded. Finally, researchers pay more attention to other defrosting methods.

3.2. Electric heating defrosting (EHD)

Electric heating defrosting usually involves electrically heating up the surface of an outdoor coil to melt off frost, as illustrated in Fig. 7(c). Kim et al. [92], Bansal et al. [93] and Ozkan et al. [94] conducted comparative studies using different types of defrosting heaters, but no quantitative defrosting results or frosting conditions were provided. In 2013, Melo et al. [95] carried out a series of experiments through a purposely-built testing apparatus. In three types of heaters, the highest efficiency of approximately 48% was obtained with a glass tube heater. A calrod heater seemed to be mostly appropriate not only because of its efficiency, which was compatible with that of the other heaters, but also due to its low cost and easy installation. Based on air bypass circulation and electric heater, Yin et al. [96] comparatively and experimentally studied a new cold storage method with different defrosting heaters and air circulation modes. When using the new method, defrosting duration was shortened by 62.1%, and defrosting energy consumption reduced by 61.0%. The defrosting efficiency was increased to 77.6%, which was 2.93 times of that by using a conventional electric heating defrosting method. However, additional electrical energy required to melt frost is high quality energy. Meanwhile, an ASHP unit is out of operation during defrosting, which results in the interrupt of indoor air heating. Therefore, this method is limited in civilian application domain, such as household air conditioning and residual ASHP units.

3.3. Hot water spraying defrosting (HWSD)

Hot water spraying defrosting method can be applied to where hot water for defrosting is available. As shown in Fig. 7(d), during defrosting, the indoor and outdoor fans are turned off, with the hot water spraying the outdoor coil. And thus, the frost could be melted and is flowing away with the water. However, only limited reported studies can be identified including a patent from Tanker, and Abdel-Wahed experimentally investigated applying hot water spraying defrosting method to a horizontal flat plate surface [14]. Obviously, hot water source is one important limitation for application, especially low cost and continuous hot water. In addition, at the termination of hot water spraying defrosting, there would be some water retained on the surface of fin due to surface tension [97]. The retained water would degrade system operating performance when it changes to heating mode. Finally, this defrosting method is not widely investigated or applied.

3.4. Hot gas bypass defrosting (HGBD)

Hot gas bypass defrosting is mainly applied to industrial ASHP units. As shown in Fig. 7(e), the superheated refrigerant vapor discharged from compressor is directed into an evaporator, or outdoor coil, bypassing a condenser and an expansion device. Most likely latent heat of condensation of refrigerant vapor is used as heat source, however, sensible heat of highly superheated refrigerant vapor may also be used [98]. On the basis of hot gas bypass defrosting, in 2009, Fu et al. [99] divided an outdoor coil into two parts, a front part and a rear part, which were used as an evaporator and a condenser respectively during defrosting. It was indicated that energy was used more efficiently, and thus defrosting duration was shorter and defrosting loss less, than those of using reverse cycle defrosting. A novel dual hot gas bypass defrosting method was also developed to remove frost from the outdoor coil of an ASHP unit [100], showing that the proposed method could overcome the main disadvantages for reverse cycle defrosting and hot gas bypass defrosting. However, defrosting duration is always very long, due to the fact that energy use for defrosting comes from the power input to the compressor. In addition, it is easy for compressor to suck in liquid during a hot gas bypass defrosting process due to insufficient energy supply, which impacts badly on the safety of compressor [30]. Finally, the hot gas bypass defrosting method is mostly used in industry units.

3.5. Reverse cycle defrosting (RCD)

When an ASHP unit is operated at reverse cycle defrosting mode, its outdoor coil acts as a condenser and its indoor coil as an evaporator [81], [82]. Operation illustration of reverse cycle defrosting for an ASHP unit is shown in Fig. 7(f), and the normal operation cycle during heating for an ASHP unit is reversed by using a four-way valve. During defrosting, hot gas is pumped into an outdoor coil to melt off the frost. When the frost is melted and drained away from the coil, the ASHP unit returns to heating operation. Besides requiring a four-way valve, reverse cycle defrosting does not need more complicated or space demanding components. That means the system is simple and easily installed [83], [84]. The energy used for reverse cycle defrosting mainly comes from four sources: (1) thermal energy of indoor air, (2) metal energy storage of indoor coil, (3) electricity input to indoor air fan, and (4) electricity input to compressor. Meanwhile, the energy is consumed in five aspects, (1) heating outdoor coil metal, (2) melting frost, (3) heating the melted frost, (4) vaporizing retained water, and (5) heating ambient air. The sufficient energy sources makes the duration of a reverse cycle defrosting operation much shorter than that of hot gas bypass defrosting [101]. In fact, reverse cycle defrosting has been the most widely used standard defrosting method for ASHP units for many years.

To clearly distinguish the 5 defrosting methods, their operation differences and evaluation results are summarized in Table 5, Table 6, respectively. To decrease the defrosting duration, only reverse cycle defrosting needs turning on the indoor air fan, and compress shutdown defrosting would turn on the outdoor air fan. For hot gas bypass defrosting and reverse cycle defrosting, compressor should be turned on to supply enough defrosting energy. Although the system stability for compress shutdown defrosting and electric heating defrosting is highly evaluated, the former results in bad defrosting effect and the later costs much high quality energy. Consequently, the two defrosting methods have the lowest comprehensive evaluation value, and are not widely applied. Hot water spraying defrosting is limited by its inconvenience, discontinuity, and high cost hot water source. Hot gas bypass defrosting has good system stability and defrosting effect, however, more electric energy is needed than that consumed in reverse cycle defrosting. Finally, the comprehensive evaluation value of reverse cycle defrosting is the highest.

Table 5. Operation differences of 5 defrosting methods.

ItemMethodsIndoor fanOutdoor fanCompressorThermal source
1CSDDOffOnOffAmbient air
2EHDOffOffOffElectricity
3HWSDOffOffOffHot water
4HGBDOffOffOnElectricity
5RCDOnOffOnElectricity

Table 6. Evaluation results of five defrosting methods.

ItemMethodsSystem complexityaSystem stabilitybDefrosting effectbComprehensive valueb
1CSDD311
2EHD331
3HWSD222
4HGBD222
5RCD133
a

↑: Increased; →: Unchanged;

b

3: The best; 1: The worst.

4. Improvement methods for reverse cycle defrosting

Currently, the most widely used standard defrosting method is reverse cycle defrosting. Reverse cycle defrosting system is simple and easily controlled, but a reverse cycle defrosting operation is a complex process involving spatial and time variations of the temperatures of refrigerant, metal and air, as well as many other indeterminate factors resulted from transient cycling which may last for only a few minutes [30]. Also, an energy balance on the airside of an outdoor coil is complex due to the fact that the energy extracted from hot refrigerant gas is utilized in 5 different ways. As previously discussed, defrosting for an ASHP unit consumes energy and causes undesirable fluctuations of indoor air temperature and other operational problems, such as low-pressure cut-off or wet compression. Therefore, with both experimental and numerical approaches, extensive research work has been carried out to improve the operating performance of ASHP units during reverse cycle defrosting.

4.1. Experimental studies on reverse cycle defrosting

4.1.1. ME1: Original component optimization

Series of investigations about original component optimization is easy to be considered, due to system operation characteristic of reverse cycle defrosting. First, thermal expansion valve (TEV) was used to experimentally investigated the transient reverse cycle defrosting performance of a nominal 3-ton residential ASHP unit. It was found that the accumulator of the ASHP unit and the TEV impacted significantly on the dynamic responses of system. With either a scroll or a reciprocating compressor, the cycle performances during reverse cycle defrosting for an ASHP unit was further experimentally compared [14]. Then, the effects of an accumulator in suction line on frosting/defrosting performance was investigated, showing that the removal of the accumulator produced a 10% reduction in defrosting duration but a 25% reduction in the integrated cyclic COP [30]. Also, using a refrigerant charge compensator instead of an accumulator, it was found an increase in refrigerant flow rate and higher suction and discharge pressures of compressor in an ASHP unit during its defrosting [102]. Defrosting effect was improved by the addition of compensator with an increased circulation. However, effect of original component optimization is limited, and thus researchers tried to add thermal energy storage (TES) system in an ASHP unit.

4.1.2. ME2: PCM-TES based RCD

During reverse cycle defrosting, the indoor air fan in an ASHP unit is normally switched off to avoid blowing cold air directly to a heated indoor space, affecting thermal comfort of occupants [103]. Because the defrosting operation always occurs at night, when the ambient air temperature is at lower temperature than at day, the sleep thermal comfort is always degraded [104], [105], [106]. Hence, the energy available from the indoor coil is basically that stored in coil metal but there is an insignificant amount of energy available from indoor air because of a negligibly small airside convective heat coefficient resulted from de-energized indoor air fan during defrosting. Consequently, low-pressure cut-off or wet compression may take place, which may cause the ASHP unit to shutdown and possibly damage the compressor. To avoid these aforementioned problems, the technologies of TES and phase change materials (PCMs) found their wide applications due to their advantage of high density energy storage [107], [108]. First, DX40 was used as a thermal storage material [109] in a heat source tank for defrosting, demonstrated a higher defrosting efficiency reached after the PCM-TES used. Then, inorganic type PCM, CaCl2⋅6H2O, was used in an ASHP unit [[87], [110]. As indicated, this PCM-TES based defrosting method could help achieve improved indoor thermal comfort, with a shorter defrosting period and a higher indoor supply air temperature during reverse cycle defrosting. The same conclusions were given in similar studies [111], [112]. As summarized in Table 7, PCM-TES based reverse cycle defrosting were widely explored in 20002017.

Table 7. Studies of PCM-TES based reverse cycle defrosting (2000–2017) [113], [114], [115].

ItemYearAuthorPCM typeTES typeHeat transfer fluidConclusions
12017Qu et al. [112]WaterShell and tubeRefrigerantDefrosting duration was shortened by 71.4–80.5%, and defrosting energy consumption reduced by 65.1–85.2%
22015Dong et al. [111]65 mol% Capric Acid with 35 mol% Lauric AcidFinned tubeRefrigerantDefrosting duration was reduced by 60%, compressor suction pressure and temperature increased by 0.28 MPa and 13.6 °C, COP increased about 1.8
32015Wang et al. [16]CaCl2·6H2OFinned tubeRefrigerantThe regeneration time was 4 min longer at −3 °C than 3 °C, but the regeneration efficiency was reduced by 15%, due to more heat stored in PCM-TES during heating mode
42014Zhang et al. [113]65 mol% Capric Acid with 35 mol% Lauric AcidFinned tubeRefrigerantCompared to the traditional method, defrosting time was shortened by 65%, the total energy consumption was less than 27.9%
52014Daikin [30]Polyethylene glycolFinned tubePolyethylene glycolThe heat capacity was improved by about 10% and COP by 50%
62012Qu et al. [87]CaCl2·6H2OShell and tubeRefrigerantDefrosting duration was reduced by 36.4%
72012Wu et al. [114]75 wt% paraffin and 25 wt% expanded graphiteTube and finRefrigerantWhen the outlet temperature drops to the minimum temperature for usage (40 °C), the efficiency of water tank with PCM is 94.23%, while the efficiency of water tank without PCM is 97.85%, meaning a water tank with PCM has a lower energetic efficiency than that without PCM
82011Hu et al. [110]CaCl2·6H2OShell and tubeRefrigerantDefrosting duration was reduced by ∼3 min or 38%, and compressor suction pressure increased by about 200 kPa, the mean indoor coil surface temperature increased about 25 °C.
92011Dong et al. [115]CaCl2·6H2O with SrCl2·6H2OShell and tubeRefrigerantDefrosting duration was reduced by 60% and energy consumption during defrosting reduced by 48.1%
102006Chen et al. [109]DX40Plate tableAirIndoor air temperature was kept steady, and the indoor thermal comfort improved

4.1.3. ME3: Improving defrosting evenness value

During reverse cycle defrosting, while most melted frost drains off from the surface of outdoor coil, some may however retain on the surface of the finned coil. The retained water should be removed to prevent it from becoming ice when the ASHP unit returns to heating/frosting mode [66]. Therefore, a complete reverse cycle defrosting process covers both melting frost and drying coil surface [116], [117]. As previously mentioned, multiple parallel circuits outdoor coils are commonly used in an ASHP unit. On its airside, there is usually no segmentation corresponding to the number of refrigerant circuits [118]. In order to highlight this point, multi-circuit outdoor coil used in the experimental studies are specially summarized and listed in Table 8.

Table 8. Multi-circuit outdoor coil used in the experimental studies (2000–2017) [119], [120], [121], [122], [123], [124], [125], [126], [127], [128], [129], [130], [131].

ItemCircuit numberAuthor
12Ding et al. [119], [120]; Hu et al. [110]; Dong et al. [111], [115];
23Song et al. [89], [116], [117], [121], [122], [123];
34Payne and O'Neal [30]; Liu et al. [26]; Qu et al. [87], [124]; Wang et al. [125]; Kim et al. [126], [127];
45Tan et al. [38];
56Stoeker et al. [90]
67Wang et al. [102]
710Yao et al. [6]
812Choi et al. [100]; Huang et al. [128], [129];
9UnknownCho et al. [130]; Huang et al. [131];

However, very limited reported studies on defrosting characteristics over the surface of multi-circuit outdoor coils may be identified in the open literature. For example, O’Neal et al. and Qu et al. both investigated the transient defrosting performances of ASHP units, each with a vertically installed four-circuit outdoor coil [87]. Stoeker et al. investigated defrosting phenomena using a six-circuit outdoor coil [90]. Wang et al. [102] tested defrosting performance of an ASHP unit with at a seven-circuit outdoor coil. In the aforementioned experimental studies, phenomena of uneven defrosting are all found, which was defined by Song et al. [121]. To quantitatively evaluate the uneven defrosting, defrosting evenness value (DEV) was further defined as the ratio of the minimum duration for tube or fin surface temperature reaching the pre-set defrosting termination temperature to the maximum value in the several circuits [101], [122]. A lower DEV always means a longer defrosting duration and lower defrosting efficiency. Then, to improve the DEV, the negative effects of downwards flowing of the melted frost along a multi-circuit outdoor coil surface were numerically demonstrated by Qu et al. [132] and experimentally investigated by Song et al. [116], [117]. Although experimental results suggested that the use of the water collecting trays helped shorten the defrosting duration by 9.2% and reduce the defrosting energy use by 10.4%, the uneven defrosting phenomenon still existed. In addition, for an existing multi-circuit outdoor coil in an ASHP unit, it is hard to modify its structure. Therefore, traditional vertical multi-circuit outdoor coil was tried to be installed horizontally to reduce the flow path of melted frost, and change the flow directions of hot refrigerant and cold melted frost from opposite to orthogonal. As expected, a better defrosting performance was reached with an increase of 9.8% in defrosting efficiency [122]. However, changing the installation style of its outdoor would take bigger floor space.

4.1.4. ME4: Improving frosting evenness value

Another reason of uneven defrosting attracted attention of Song et al., which is that frost accumulations on each circuit’s surface of a multi-circuit outdoor coil in an ASHP unit were different at the start of defrosting operation [89]. As demonstrated, less frost accumulated on the surface of outdoor coil when the FEV is higher, and thus frost retarding is expected by adjusting the refrigeration distribution in a multi-circuit outdoor coil. However, when we focus on the defrosting period, it was demonstrated that, defrosting efficiency could be improved by increasing the FEV at the start of a defrosting operation [89], [101]. As reported, the negative effects on defrosting performance for uneven frosting as a defrosting process start were confirmed, and an increase of 6.8% in defrosting efficiency was reached after the FEV increased from 82.6% to 96.6%. [89] Furthermore, when melted frost locally drained away by installing water collecting trays between circuits [101], and the FEV increased from 79.4% to 96.6%, the defrosting duration would be shortened by 23 s, from 198 s to 175 s, or about 11.2% less. At the same time, the total energy supply during defrosting could be decreased by about 3.7%, from 781.8 kJ to 678.8 kJ, and defrosting efficiency increased by about 5.7%, from 45.0% to 50.7%. Airside surface conditions of the outdoor coil at the start of defrosting operation at different FEVs without and with water collecting trays installed are shown in Fig. 8(a) and (b), respectively. Therefore, with the method of FEV adjustment at frosting period, the reverse cycle defrosting performance could also be improved by starting at a higher FEV.

  1. Download : Download high-res image (832KB)
  2. Download : Download full-size image

Fig. 8. Airside surface conditions of the outdoor coil [89], [101]

4.1.5. ME5:Airflow and refrigerant distribution adjustment

Mal-distribution of refrigeration or airflow might result in uneven defrosting phenomenon, and thus degrade the defrosting performance. In 2000, Aganda et al. found that airflow mal-distribution reduced the performance of an evaporator circuit [133], [134]. With refrigerant flow controlled by one TEV, the worst performing circuit affected the performance of the whole outdoor coil, by as much as 35%. Kim et al. [126], [127] experimentally and numerically investigated a hybrid-individual degree of superheat control method for refrigerant flow balancing in a multi-circuit evaporator: upstream versus downstream flow control. Results showed that the upstream refrigerant flow control consistently outperformed the downstream refrigerant flow control, and recovered most of the loss in cooling capacity and COP due to non-uniform airflow distribution. Based on these conclusions, they utilized the model to further evaluate the effects of uneven air and refrigerant flow distributions and the benefits of upstream hybrid control during defrosting for an ASHP unit [127]. That means adjusting airflow and/or refrigerant distribution could improve defrosting performance.

However, it is easy to be confused that the improvement method of airflow and refrigerant distribution adjustment is the same to another frost retarding method of FEV adjustment, due to both adjusting the refrigerant distribution. In fact, the same FEV might result from totally different control strategies, by adjusting air and/or refrigeration distribution dynamically, or structure of a multi-circuit outdoor coli. The FEV works as an index to evaluate these control strategies. For example, three control strategies to alleviate uneven defrosting were numerically and comparatively investigated, including adjusting the refrigeration distribution and the compressor speed [135]. In addition, this defrosting improvement method by adjusting refrigeration distribution focuses on the defrosting process, while frost retarding method on heating/frosting process. Therefore, frost retarding and reverse cycle defrosting improvement can be both considered and reached by adjusting the refrigerant distribution in a multi-circuit outdoor coil. This control strategy might be considered in the designing of an intelligent ASHP unit.

4.1.6. ME6:Sensible heat defrosting method

To avoid adverse shock and ‘oil rush’, which were commonly seen in conventional reverse cycle defrosting operations, sensible heat defrosting method was proposed and numerically investigated by Liang et al. [98], by using a self-organizing fuzzy control system in an ASHP unit. In this new defrosting method, the four-way valve did not act when the ASHP unit switched its mode from heat supply to defrosting. Therefore, it is similar to the hot gas bypass defrosting method, as shown in Fig. 7(e). Refrigerant was discharged from the compressor when it was sucked and compressed by the compressor, and passed through the hot gas solenoid valve and the discharge conduit. After throttled by the expansion valve, the refrigerant with high temperature and low pressure flowed into the outdoor coil and experienced the heat exchange process with the outer frost layer. After that, the refrigerant passed through the accumulator and was sent back to the compressor for recompression. In this way, the cycle continued. However, in order to ensure the normal operation of the system, it should be guaranteed that there would not be any condensation of the refrigerant. This also causes the energy hard to be effectively used [136]. Moreover, the defrosting duration would be prolonged due to all energy supply from the electricity input to the compressor, and thus the indoor thermal comfort adversely affected [103]. This method might be suitable for industry applications, but limited in residential ASHP units.

Evaluation results of 6 improvement methods are listed in Table 9, in which the initial cost, running cost, system complexity, defrosting effect and comprehensive value are considered and compared. As seen, the ME2, PCM-TES based reverse cycle defrosting, is the most recommended method to enhance the defrosting for an ASHP unit. Then, the ME4, improving frosting evenness value, is evaluated the lowest, due to its limited optimizing effects.

Table 9. Evaluation results of 6 defrosting enhancement methods.

MethodsInitial costaRunning costaSystem complexitybDefrosting effectbComprehensive valueb
ME122
ME233
ME322
ME411
ME522
ME622
a

↑:; →: Unchanged.

b

3: The best; 1: The worst.

4.2. Theoretical studies on reverse cycle defrosting

As a non-linear, moving-boundary, variable-density, multi-dimensional, phase-continuously-changing, transient, dynamic heat- and mass-transfer process, the defrosting process is a common and complicated physical phenomenon. During a defrosting operation, the frost on the surface of an outdoor coil will not necessarily be melted uniformly throughout the surface. The frost over some parts of the coil surface remains attached to the coil surfaces until it is completely melted and sublimated while the frost at other locations may be partially melted, and then detaches from the coil surface, falling down to the coil surface at a lower level or to a drainage tray. Two types of models were classified in this paper, one type around components and system, and the other one focus on multi-circuit outdoor coil.

4.2.1. Defrosting model of components and system

From the open literature available, modeling a defrosting process has attracted lots of research attentions. Early modeling work focused mainly on outdoor coils of simple geometry, such as finite slabs, horizontal flat plate, or flat plate cooler. Then, on a cylindrical coil cooler, a semi-empirical model for electric defrosting was presented, and an analytical model was developed to predict the evaporation, sublimation and melting rates during defrosting. A moving boundary technique was used and the defrosting process was divided into two stages, pre-melting and melting [14]. Later, Alebrahim and Sherif [137] reported an electric defrosting model for a finned-tube outdoor coil using the enthalpy method to predict defrosting duration and frost surface temperature profiles. Thereafter, a number of studies on modeling a defrosting process in ASHP units were carried out. Noticeably, a reverse cycle defrosting model for an outdoor coil was developed, where the process of frost melting on the surface of an outdoor coil was subdivided into four stages: pre-heating, melting, vaporizing and dry heating [14], as illustrated in Fig. 9. A number of heat and mass transfer parameters required for simulating defrosting performance, e.g., the maximum mass of surface water, free-convection air film conductance, air/water film conductance and surface water vaporization coefficient, were however experimentally determined. On the basis of the aforementioned model, a validated defrosting model for an ASHP unit using capillary tube was developed by Liu et al. [138] in 2003. Distributed modeling was used for both evaporator and condenser because of their importance during reverse cycle defrosting.

  1. Download : Download high-res image (332KB)
  2. Download : Download full-size image

Fig. 9. Schematic diagram illustrating defrosting of an outdoor coil cell.

4.2.2. Defrosting model of multi-circuit outdoor coil

Although the aforementioned defrosting models [14], [137], [138] were developed and used in studying defrosting performance, none of them in fact considered the negative effects of downwards flowing of the melted frost due to gravity along the surface of an outdoor coil on defrosting performance [117], [116], by either assuming a stable water layer or no water retention on coil surface. In 2012, Qu et al. [132] reported on a modeling analysis where a semi-empirical model for the defrosting on the airside of a four-circuit outdoor coil in an ASHP unit was developed. Different from aforementioned defrosting modeling studies, the negative effects of melted frost on defrosting performance were considered and quantitatively studied in this model. It was further predicted that if the melted frost could be drained away locally, the defrosting efficiency for the ASHP unit could be increased by up to 18.3%. Similar energy consumption ratio was also reported by Dong et al. [139] in a study on the energy consumption analysis on vaporizing the melted frost and heating ambient air during reverse cycle defrosting in an ASHP unit.

In consideration of the energy used to heat the metal in an outdoor coil accounted for as much as 16.5% of the total defrosting energy use during defrosting [132], Song et al. developed two new models for with and without water collecting trays installed between circuits [140]. Two empirical models, corresponding to the two previous experimental studies [116], [117], were therefore validated with those experimental data. Basing on the two models developed, a modeling study on alleviating uneven defrosting for a vertical three-circuit outdoor coil in an ASHP unit during reverse cycle defrosting was further carried out [135]. It was demonstrated that defrosting energy use could be decreased to 94.6% by fully closing the modulating valve on the top circuit when its defrosting terminated. Meanwhile, a reduction of 7 s in defrosting duration was reported. To clearly understand the two modeling studies reported by Qu and Song, their differences are summarized and listed in Table 10.

Table 10. Differences between two defrosting models.

ItemDifferent aspectsQu’s defrosting model [132]Song’s defrosting model [135], [140]
1Multi-circuit outdoor coil4 circuits3 circuits
2Metal energy storageWithout consideredConsidered
3Experimental conditionsUneven frosting startEven frosting start
4Assumptions given69
5Stages divided3 stages4 stages
6Model validation parameters14
7Defrosting efficiency increased after melted frost drained away40.8% (18.3% increased)Without prediction (Given in experimental studies)
8Model extension studyWithout extensionControl strategies

The main equations used in two modes were listed in Table 11. Based on Qu’s model, a more accurate model was developed by Song, with more assumptions given, more stages divided, and more experimental conditions considered. In Qu’s model, it was predicted that defrosting efficiency was increased by 18.3% after the melted frost drained away from each circuit. In Song’s model, defrosting efficiency was not predicted, but it was reported in their experimental study that, after the water collecting trays installed between circuits for the three-circuit outdoor coil, defrosting efficiency could be increased about 13.2%, from 43.5% to 56.7% [117]. Finally, as the potential uses and limitations of the modeling work described [135], the defrosting model study for an ASHP unit with multi-circuit outdoor coil should be further developed.

Table 11. Main energy equations used in two defrosting modeling studies.

Qu’s defrosting model [128]
1First stage: frost melting without water flowqj=Lsfmf,j+cpd(Mw,jTw,j)dt
2Second stage: frost melting with water flowqj+cpmw,j-1Tw,j-1=Lsfmf,j+cpMw,maxdTw,jdt+cpmw,jTw,j+hc,w(Tw,j-Ta)Af-a
3Third stage: water layer vaporizingqj=cpd(Mw,jTw,j)dt+hc,w(Tw,j-Ta)Aw-a+hc,d(Tr,j-Ta)Ad-a+mv,jLv

Song’s defrosting model [131], [133]
1First stage: preheatingqj=Lsfmf,j+cpd(Mw,jTw,j)dt+qMe
2Second stage: frost melting without water flowing away from a circuitqj=Lsfmf,j+cpd(Mw,jTw,j)dt+hc,w(Tw,j-Ta)Af-a+qMe
3Third stage: frost melting with water flowing away from a circuitq1=Lsfmf,1+cpMw,maxdTw,1dt+cpmw,1Tw,1+hc,w(Tw,1-Ta)Af-a+qMeqj+cpmw,j-1Tw,j-1=Lsfmf,j+cpMw,maxdTw,jdt+cpmw,jTw,j+hc,w(Tw,j-Ta)Af-a+qMe
4Fourth stage: water layer vaporizingqj=cpd(Mw,jTw,j)dt+hc,w(Tw,j-Ta)Aw-a+hc,d(Tr,j-Ta)Ad-a+mv,jLv+qMe

As the most popular defrosting methods for ASHP units, reverse cycle defrosting attracted more and more attentions. Series of experimental investigations on it were undertaken, consisting of optimizing original component, PCM-TES based reverse cycle defrosting, improving defrosting evenness value, adjusting frosting evenness value, adjusting air and refrigerant distribution, and sensible heat defrosting method. Among all the experimental studies on reverse cycle defrosting optimization, PCM-TES based reverse cycle defrosting was the most strongly recommended, with its advantages of easy and inexpensive installation, and good defrosting effect. However, both of system frosting and defrosting performance should be tested, when the system is optimized with any defrosting enhancement methods. If the refrigerant distribution adjustment solenoid valves installed on each circuit and water collecting trays placed between circuits, the FEVs could be increased, and thus the increases of the COP and defrosting efficiency for an ASHP unit are both expected. Therefore, coupled using the previously mentioned system optimization methods during reverse cycle defrosting would be better.

On the other hand, numerical studies on reverse cycle defrosting were conducted, with defrosting models of system and outdoor coil developed. For the defrosting model of system, component optimization and their effects on defrosting efficiency attracted most of attentions. Differently, in defrosting model of an outdoor coil, the defrosting procession was always introduced more detailed. For example, the defrosting process was divided into only two stages, pre-melting and melting [14], in the defrosting model of system, while it was divided into four stages in Song’s defrosting model. In addition, based on this defrosting model of multi-circuit outdoor coil, three control strategies were further numerically investigated [135].

5. Control strategies

Frosting seriously affect the operating performance of an ASHP unit by reducing its COP and output heating capacity. Hence, periodic defrosting is necessary. Defrosting initiation and termination control strategies impact system reliability and energy efficiency [141]. Studies on selecting suitable control parameters were therefore carried out.

5.1. Initiation of a defrosting operation

There are two types of control strategies to start a defrosting operation: time-based start defrosting and demand start defrosting. For the former, defrosting start is controlled by a pre-set timer. Due to the advantages of simplicity and low-cost, many of the earlier ASHP units employed time-based start defrosting method. Usually, for every 60–90 mins of frosting time, a defrosting operation would be initiated [14], [87], [89], [139]. However, performances of these earlier ASHP units could well suffer from some unnecessary defrosting operations, resulting in a degraded operational efficiency. Time-based start defrosting results in two typical mal-defrosting problems, one is unnecessary defrosting cycles when no or few frost accumulated on surface of outdoor coil, and the other is no defrosting in progress when necessary. As reported by Wang et al., the mal-defrost phenomenon was found with more than 60% frosted area of the outdoor coil after the system running 5 days [125]. During this frosting period, the system COP was significantly degraded, only 2.3 under an environment temperature of 7.9 °C. The mal-defrost phenomenon decreased the system COP by up to 40.4% and the heating capacity by up to 43.4%. To avoid the mal-defrosting problem, and thus improve the energy efficiency of system, temperature and/or pressure parameters were considered in time-based start defrosting method. When the parameter(s) reached pre-set value(s), the timer starts. These time-based start defrosting methods become widely used in applications.

Another control strategy was applied to initiate a defrosting cycle was firstly proposed by Eckman, and named as demand start defrosting [30]. As the name suggests, it could start a defrosting operation only when needed. A demand start defrosting control strategy defrosts the display cabinet when sufficient frost is formed to adversely affect the operating performance. It would lead to: (1) better temperature control, (2) increased product quality and life, (3) reduced product losses, and (4) significant energy savings. When applying this control strategy, an ASHP unit would start defrosting only when an adequate frost buildup was detected. Thus, it is important to accurately detect the presence and growth of frost. A number of frost detecting techniques have been developed over the years, including: (1) measuring the thermal conductivity of ice, (2) calculating air pressure differential across an evaporator, (3) calculating the degree of refrigerant superheat [142], (4) sensing the temperature difference between the air and evaporator surface, and (5) sensing outdoor air fan power [14], [30], etc.

Based on aforementioned direct and indirect frost accumulation sensing technologies, recent defrosting start control strategy includes: (1) measuring the ice thickness using holographic interferometry technique [143], (2) measuring the frost surface temperature by infrared thermometer [144], (3) sensing refrigerant flow instability [145], (4) sensing frost using photo-coupler, photo-optical systems or fiber-optic sensors [[14], [146], (5) modeling the amount of frost on coil surface by applying neural networks [30], and (6) calculating effective mass-flow fraction by fin surface temperature [147]. Particularly, principle of photoelectric technology for frost detection using photo-coupler was systematically investigated by Wang et al. with experimental, numerical and theoretical approaches [[125], [148], [149], [150]. Based on a frosting map for the ASHP unit, a novel Temperature-Humidity-Time defrosting control strategy was further proposed by them [151]. There are three typical regions existed in this map, frosting region, condensing region, and non-frosting region. The working condition of an ASHP unit will be clearly presented in this frosting map, and thus the corresponding control strategy could be accurately made. This frosting map is fundamental and meaningful for the intelligent control designing work of ASHP unit.

5.2. Termination of a defrosting operation

Mal-defrosting was found and reported by many scholars. It was defined as defrosting operations are carried out a long time after a ‘critical’ level of frost has been reached, or when they are not necessary by Wang et al. in 2011 [152]. Clearly, this definition focuses on the start of defrosting, but neglects the conditions that a defrosting operation is terminated at earlier or later of the ‘critical termination time’. With respects to defrosting termination, related research is much less seen. It should be noted that for defrosting on an ASHP unit, a complete defrosting process covers both melting frost and drying coil surface. Otherwise, once the ASHP unit returning to heating operation, retained melted water on outdoor coil surface would become ice. This may change the structure of a frost layer, increase the density and enhance thermal conductivity of the frost layer [153]. During defrosting, not only a great deal of energy for melting frost and vaporizing melted frost off outdoor coil surface is consumed, but also the occupants thermal comfort may be adversely affected [154]. Therefore, shortening a defrosting period is always one of the defrosting control purposes for ASHP units. For example, Chinese Standard GB/T 7725-2004 specifies that the defrosting duration for an ASHP unit should not exceed 20% of its total working hours.

In practical applications, a reverse cycle defrosting operation can be terminated based on tube or fin surface temperature of an outdoor coil, refrigerant pressure difference across an outdoor coil, or defrosting operation time [14]. Although the control strategy of time-based start defrosting is widely used, a defrosting operation always does not terminated based on time. Currently, the mostly used method for terminating a defrosting operation is based on the tube surface temperature of an outdoor coil. A temperature sensor is usually placed on the tube surface at the exit of the lowest liquid-line circuit of a vertically installed multi-circuit outdoor coil [122]. A defrosting operation will be terminated once a pre-set temperature is reached. When the multi-circuit outdoor coil is horizontally installed, temperatures of all circuits will be considered and the longest circuit defrosting duration is used as the defrosting duration of a reverse cycle defrosting cycle [[122], [155]. It is obviously that when the pre-set temperature is higher or lower, the defrosting duration would be prolonged or more residual water left, respectively. Both of them result in potential energy waste for an ASHP unit, or even adversely degrade the indoor thermal comfort [103]. However, there is no standard defrosting termination temperature (DTT) or even a fixed range given in application, due to the diversity of equipment and operating climates. Different DTT settings were used and reported in 20002017, from 10 °C [129] to 50 °C [156], as summarized in Table 12. The temperature range covering 40 °C is too big. In open literature, nearly no scholar present any relative studies, or even point out this fundamental problem. It is worth mentioning that, to save energy for an ASHP unit, an experimental methodology was firstly presented and validated by Song et al. [157], [158]. As seen in Fig. 10, the defrosting efficiency reaches its peak at 175 s, at about 60.6%. That means the corresponding DTT at 175 s is the best value. Finally, a suitable DTT at the range of 20–25 °C, or around 22 °C, was suggested for this experimental unit. This method could be widely used to fix the termination temperature, and thus saving energy for ASHP units.

Table 12. DTT settings for ASHP units (2000–2017).

ItemDTT (°C)Circuit numberCapacity (kW)YearAuthor
1101255 (Cooling)2009Huang et al. [125]
21228.82 (Cooling)2004Ding et al. [116]
315455–350 (Heating)2013Wang et al. [83]
41846.8 (Heating)2010Qu et al. [100]
520//2005Cho et al. [126]
6201216 (Cooling)2011Choi et al. [97]
7221250 (Cooling)2004Huang et al. [127]
82436.8 (Heating)2014Song et al. [113]
92436.8 (Heating)2015Song et al. [112], [117]
102436.8 (Heating)2016Song et al. [86], [87]
112436.8 (Heating)2017Song et al. [152]
12241255 (Cooling)2007Huang et al. [124]
132446.8 (Heating)2012Qu et al. [84]
1425237.5 (Heating)2015Dong et al. [107]
152624.8 (Heating)2012Dong et al. [132]
163040.88 (Compressor)2003Liu et al. [24]
173322.8 (Cooling)2012Dong et al. [153]
183522.5 (Heating)2011Dong et al. [111]
195022.5 (Heating)2011Hu WJ [149]
  1. Download : Download high-res image (728KB)
  2. Download : Download full-size image

Fig. 10. Defrosting efficiency calculated and the most suitable DTT [157], [159], [160]

Different control strategies to start and end a defrosting were carried out to improve the ASHP system operating performance at an entire frosting-defrosting cycle. Currently, developed time-based control strategies are widely used to start a defrosting cycle, due to their advantages of strategy simplicity and reliability, although the accuracy is still questionable. For an ASHP unit with a multi-circuit outdoor coil, as summarized, terminating a reverse cycle defrosting operation based on the tube surface temperature at the exit of the lowest circuit is the most widely used method. Also, it is still questionable in practical applications due to the DTT setting seems random, at a wide temperature range. More defrosting start and termination control strategies should be explored.

6. Outlook for future research

Extensive related studies on the performance of ASHP units under frosting or defrosting conditions have been undertaken, with experimental, theoretical and numerical approaches. Various measures to address the problem associated with the frosting/defrosting operation were considered. However, there are still 5 important areas where further in-depth research work is required.

  • (1)

    System and component optimization. When optimizing the original components in an ASHP unit, more attention should be payed to the entire system operating performances. Experimental or numerical studies on the frosting/defrosting performance for an ASHP unit at different frost accumulations, circuit numbers, total heat exchanger areas, etc., still have not been reported. When an ASHP unit horizontally installed multi-circuit outdoor coil was applied at some special places, such as on the roof of room or vehicles, its system performance could be further explored.

  • (2)

    New method and material. New defrosting method and control strategy, taking its economic, energy and environmental performances into consideration, should be explored. New hydrophobic materials with good durability, high thermal conductivity, and self-cleaning function are meaningful and expected in industry application.

  • (3)

    Model development. New defrosting models of an ASHP unit with PCM-TES, with the indoor thermal comfort or sleep thermal comfort considered, should be developed. New frosting and defrosting models for a flat plate or fin at micro/nano scales should be developed, for example, using lattice Boltzmann method.

  • (4)

    Control strategy optimization. A defrosting control strategy of refrigerant distribution adjustment should be explored, which leading the melted frost to play positive effects on defrosting performance in an ASHP unit with a vertically installed multi-circuit outdoor coil. More indoor and outdoor environmental parameters could be considered to avoid mal-defrosting problem. Operation control strategy coupled with big data of climate can be considered.

  • (5)

    Mechanism study. For some mechanical based frost retarding methods, such as the ultrasonic vibration and air jet techniques, their mechanisms of frosting process on the fin surface should be explored. In order to improve the FEVs for a multi-circuit outdoor coil, and thus improving the system frosting/defrosting performance, mechanism of multi-phase refrigerant distribution into each circuit during uneven heat transfer between refrigerant and outside frost, melted frost and ambient air, should also be explored.

7. Conclusions

An update on recent developments in frosting/defrosting investigations for ASHP units are reviewed and reported, covering strictly selected articles published in 2000–2017. The current status of research in respect to this topic is summarized above, to identifying the finished work, unsolved problems, the potential and barriers to practical application as well. Findings described in the literatures demonstrate that, considering the comprehensive values of 12 listed frost retarding measures, preheating inlet air with waste heat and fin surface treatment with new materials are highly recommended. Among the classified 5 typical defrosting methods, reverse cycle defrosting is highly evaluated and 6 enhancement methods for it were thereby listed. A defrosting operation is always started based on time, at 60–90 mins, and one or two auxiliary parameters. It is always terminated when the tube surface temperature at the exit of the lowest circuit in a multi-circuit outdoor coil reaching a pre-set value at the temperature range of 10–50 °C. Fellow up studies relating to various uncertainties surrounding frosting/defrosting for ASHP units, should be addressed, regarding the work in the 5 areas outlined above.

Acknowledgements

The authors acknowledge the financial supports from the National Natural Science Foundation of China (No. 51606044 & No. 51006073), Natural Science Foundation of Guangdong Province (No. 2017A030313300), and The University of Tokyo for the work reported in this paper. The first author is grateful for the financial support of the Japan Society for the Promotion of Science (JSPS). The third author was partially supported by the JSPS Grants-in-Aid for Scientific Research (C) no 16K06109 and 25420150.

References