Welcome to visit Advances in New and Renewable Energy!

Review of Heating Value Estimating Models for Municipal Solid Waste

  • Dan WANG 1, 2 ,
  • Xu-qing LI 1, 2 ,
  • Wan-qin YANG , ††, 1, 2
Expand
  • 1. School of Life Science, Taizhou University, Taizhou 318000, Zhejiang, China
  • 2. Institute of Soil Ecology and Remediation, Taizhou University, Taizhou 318000, Zhejiang, China

Received date: 2021-09-20

  Revised date: 2021-10-14

  Online published: 2022-02-28

Copyright

版权所有 © 《新能源进展》编辑部

Abstract

Heating value (HV) of municipal solid waste (MSW) is an important factor in deciding whether MSW can be treated by incinerate, and also an important factor in determining the design and operation of incineration plants. Empirical models are usually used to estimate the HV of MSW. In this work, the research on HV estimating models of MSW at home and abroad were summarized, and the representation of HV, accuracy of models, methods to build HV models, data sizes and sources were systematically reviewed and compared. The results showed as follows: the representations of HV were inconsistent; the sample size of model was small; there was lack of generalized model which can be applied internationally, and the application of artificial neural networks needs further exploration. Building generalized HV estimating model with good performance which can be applied worldwide might be hot point of future research.

Cite this article

Dan WANG , Xu-qing LI , Wan-qin YANG . Review of Heating Value Estimating Models for Municipal Solid Waste[J]. Advances in New and Renewable Energy, 2022 , 10(1) : 69 -79 . DOI: 10.3969/j.issn.2095-560X.2022.01.010

0 引言

全球生活垃圾年产量已超过20亿t,预测在2050年将超34亿t[1]。2020年,我国生活垃圾清运量已达2.35亿t(图1)(数据来源于国家统计局2004-2021年中国统计年鉴,http://www.stats.gov.cn/)。不当的垃圾管理会造成严重的环境污染[2]。同时,垃圾中蕴含丰富的物质资源和能源[3,4]。垃圾焚烧处理具有减量化、安全化、快速稳定化、用水少、周期短、易操作、可处理的垃圾种类丰富、占地面积小等优点[5,6,7,8,9,10],是可持续生活垃圾管理中不可缺少的部分[9,11-12],已成为我国垃圾处理的主流方式(图1)。目前,我国垃圾焚烧发电装机规模、发电量均居世界第一[13],焚烧厂日处理能力达719 550 t,发电功率达12 386 988.9 kW,主要分布在东部地区(图2[14]
Fig. 1 Quantity of collected and safely treated municipal solid waste, and the number of waste incineration plants in China during 2003-2020 (from China Statistical Yearbook 2004-2021)

图1 2003-2020年我国生活垃圾清运量、无害化处理量及焚烧厂数量(数据来源于国家统计局2004-2021年的中国统计年鉴)

Fig. 2 Spatial distribution of MSW incineration plants in China (from Automatic Monitoring Data Disclosure Platform for Municipal Solid Waste Incineration Power Plants, https://ljgk.envsc.cn/)

图2 我国生活垃圾焚烧厂分布图(数据来源于生活垃圾焚烧发电厂自动监测数据公开平台,https://ljgk.envsc.cn/)

目前国内垃圾焚烧发电工艺有机械炉排炉、流化床、回转窑和热解4种类型[15]。机械炉排炉应用最广(图2[14],可以处理热值范围宽的生活垃圾,对预处理的要求不高,运行及维护简便,国产化程度高,投资成本适中[15]。流化床应用较少,一般用于中小型城市的焚烧厂,该工艺成本较低且可使垃圾燃烧充分,但对入炉垃圾要求极高,需要加入煤炭助燃,且容易造成空气污染[16]。回转窑焚烧炉燃烧适应性好,运行稳定,但占地面积大,热效率低,一般用于小规模特种垃圾如医疗废物、工业废物的处置[17]。热解工艺技术先进、可靠,但对垃圾含水率有一定要求,对干燥段控制要求较高,不能满足大型垃圾焚烧厂的要求,目前还未应用于生活垃圾焚烧,但在西部人口稀疏、位置偏远地区具有一定应用优势[18]
垃圾能否采取焚烧处理以及焚烧厂的设计和运行取决于垃圾的热值[19,20]。热值影响处理技术的选择、辅助燃料的添加及用量、焚烧厂的运行维护、运营管理及经济效益[21,22]。热值可通过量热仪测定或模型计算。模型计算方便快捷、经济实惠[23],可用于计算历史垃圾热值[24]。模型研究可以为精确计算垃圾热值及其可持续管理提供科学依据。

1 垃圾热值的表示

热值通常表示为高位热值(higher heating value, HHV)或低位热值(lower heating value, LHV)。高位热值表示单位质量垃圾完全燃烧后,所有产物冷却到标准状态(298 K,1 atm),水以液态形式存在时所释放的热量[25,26]。低位热值指完全燃烧后产物在150℃,水以气态形式存在且其中热量未被利用时释放的热量[26,27]。根据《生活垃圾采样和分析方法》(CJ-T313-2009),高位热值、低位热值可通过式(1)~ 式(3)进行转化。
${{Q}_{\text{(h)}}}=\frac{1}{m}\sum\limits_{j=1}^{m}{{{{{Q}'}}_{j(\text{h})}}\times \frac{100-{{C}_{(\text{W})}}}{100}}$ (1)
${H}'=\sum\limits_{i=1}^{n}{\left( {{{{H}'}}_{i}}\times \frac{{{{{C}'}}_{j}}}{100} \right)}$ (2)
${{Q}_{(\text{l})}}={{Q}_{(\text{h})}}-24.4\times \left( {{C}_{(\text{W})}}+9{H}'\times \frac{100-{{C}_{(\text{W})}}}{100} \right)$ (3)
式中:${{{Q}'}_{j\text{(h})}}$为干基高位热值,kJ/kg;${{Q}_{(\text{h})}}$为湿基高位热值,kJ/kg;${{Q}_{(\text{l})}}$为湿基低位热值,kJ/kg;${H}'$为干基氢元素含量,%;${{C}_{(\text{W})}}$为样品含水率,%;${{{C}'}_{i}}$为某成分干基含量,%;j为重复测定序数;m为重复测定次数;i为各成分序数;n为成分数量;24.4为水的凝缩热常数,kJ/kg。
热值的表示形式、基准和单位在不同研究中不同。有些甚至未说明是高位热值还是低位热值[28]。热值的表示基准包括湿基[29]、干基[5]、风干基[30]等,部分研究并未指出采用的基准[4,5]。热值的常用单位有kJ/kg[31,32]、kcal/kg[33,34]、Btu/lb[35]和MJ/kg[10,36]。这些差异主要是研究目的和使用的采样和分析标准不同导致。国际上常用的生活垃圾采样和分析标准是美国材料实验协会(American Society of Testing Materials, ASTM)(https://www.astm.org/)的系列标准,如ASTM D5468-02。我国采用的是由中华人民共和国住房和城乡建设部发布的《生活垃圾采样和分析方法》(CJ-T313-2009)和《生活垃圾化学特性通用检测方法》(CJ-T96-2013)。上述这些差异增加了不同研究之间和不同地区之间比较的难度,可能对后续的研究和应用产生一定的影响。

2 生活垃圾热值计算模型种类

2.1 元素含量分析模型

基于元素含量分析的生活垃圾热值计算模型从杜龙公式演变而来[37],使用元素含量作为参数[5](见表1)。碳、氢、氧三种元素含量对热值具有明显影响,因而包含在绝大多数模型中。其他元素如硫、氮、氯也被包含在一些模型中。极少数模型被简化到只包含了一种(例如C)或两种(例如C、H)元素。由于生活垃圾的地域差异性较大,这类简化模型的使用范围可能会受到限制。
Table 1 Summary of heating value predictive models for MSW based on ultimate analysis

表1 基于元素含量分析的生活垃圾热值计算模型

模型建立者 公式 单位 样本量 数据来源 模型评价指标
R2 MAPE / %
KHAN等[35] ${{Q}_{l}}=145.4\text{C}+620\left( \text{H}-{}^{\text{O}}/{}_{8} \right)+41\text{S}$ Btu/lb 86 文献 - -
DULONG公式 ${{Q}_{l}}=81\text{C}+342.5\left( \text{H}-{}^{\text{O}}/{}_{8} \right)+22.5\text{S}-6\left( 9\text{H}+{{C}_{(\text{W})}} \right)$ kCal/kg - - - -
LIU等[19] ${{Q}_{l}}=19.96\text{C}+44.3\text{O}-671.82\text{S}-19.92{{C}_{(\text{W})}}+1558.8\text{O}$ kCal/kg 40 实测 0.926 -
COOPER等[38] ${{{Q}'}_{(\text{l})}}\text{=}17050\text{C}+32030\left( \text{H}-{}^{\text{O}}/{}_{8}-{}^{\text{Cl}}/{}_{35.5} \right)\text{+}4591\text{S}-791$ Btu/lb 40 文献 0.948 -
CHANNIWALA等[39] ${{Q}_{h}}=0.3491\text{C}+1.1783\text{H}+0.1005\text{S}-0.1034\text{O}-0.0151\text{N}-0.0211\text{A}$ MJ/kg 275 文献 - 2.58
MERAZ等[25] ${{Q}_{h}}=\left( 1-\frac{{{C}_{(\text{W})}}}{100} \right)\left( -0.3708\text{C}-1.1124\text{H}+0.1391\text{O}-0.3178\text{N}-0.1391\text{S} \right)$ MJ/kg 101 文献 - -
KATHIRAVALE等[29] ${{{Q}'}_{(\text{h})}}=416.638\text{C}-570.017\text{H}+259.031\text{O}+598.955\text{N}-5829.078$ kJ/kg 30 实测 0.625 -
THIPKHUNTHO等[30] ${{{Q}'}_{(\text{h})}}=492.5\text{C}-926.0\text{H}+117.6\text{O}+19.3\text{S}$ kJ/kg 219 实测 0.891 10.8
${{{Q}'}_{(\text{h})}}\text{=}406.4\text{C}-210.5\text{H}+160.4\text{O}+154.8\text{S}-151.2\text{N}-23.8\text{A}$ kJ/kg 219 实测 0.905 9.3
${{{Q}'}_{(\text{h})}}=134.3\text{C}-1502.1\text{H}-2.7\left[ {{\text{O}}^{2}}/(1-\text{A}/100) \right]+29132.8(1-\text{A}/100)$ kJ/kg 219 实测 0.893 10.4
孙晓杰等[40] ${{{Q}'}_{(\text{h})}}=337\text{C}+2837\left( \text{H}-0.125\text{O} \right)+93\text{S}+23\text{N}$ kJ/kg 15 实测 - -
${{{Q}'}_{(\text{h})}}=\left[ 337\text{C}+2837\left( \text{H}-0.125\text{O} \right)+93\text{S}+23\text{N} \right]\left( 1-{{C}_{(\text{W})}} \right)-2420\left( {{C}_{(\text{W})}}+9\text{H} \right)$ kJ/kg 15 实测 - -
AKKAYA等[41] ${{Q}_{h}}=\left( 1-\frac{{{C}_{(\text{W})}}}{100} \right)\left( 0.327\text{C}+1.241\text{H}-0.089\text{O}-0.26\text{N}+0.74\text{S} \right)$ MJ/kg 100 文献 0.983 -
SHI等[5] ${{{Q}'}_{(\text{h})}}=0.361\text{C}+1.05\text{H}-0.16\text{N}+1.24\text{S}-0.0658\text{O}-1.46$ MJ/kg 193 实测+文献 0.938 -
${{{Q}'}_{(\text{h})}}=0.35\text{C}-1.01\text{H}-0.826\text{O}$ MJ/kg 193 实测+文献 0.936 6.73
EBOH等[42] ${{{Q}'}_{(\text{h})}}=0.364\text{C}+0.863\text{H}-0.075\text{O}+0.028\text{N}-1.633\text{S}+0.062\text{Cl}$ MJ/kg 86 实测 0.95 5.738
HAN等[43] ${{Q}_{h}}=35\text{C}+120\text{H}-16\text{O}$ MJ/kg - 实测 0.93 6.67
KHURIATI等[34] ${{Q}_{h}}=5751.94+52.67\text{C}+75.9\text{H}-4.14\text{N}-1044.03\text{S}-97.68\text{O}$ kCal/kg 29 实测 0.99 -
${{Q}_{h}}=114.63\text{C}+310.55\text{H}-2762.68$ kCal/kg 29 实测 0.98 0.85
${{Q}_{h}}=143.33\text{C}-1737.55$ kCal/kg 30 实测 0.94 1.35
IBIKUNLE等[44] ${{Q}_{h}}=1.3849+85.0807\text{C}-28.9675\text{H}-666.125\text{N}+11.6296\text{S}-97.68\text{O}$ MJ/kg 62 实测 0.837 -
BOUMANCHAR等[45] ${{Q}_{h}}=0.484\text{C}-4.1307$ MJ/kg 187 文献 - -
${{Q}_{h}}=3.1451\text{H}-0.8268$ MJ/kg 187 文献 - -
${{Q}_{h}}=0.3805\text{C}+0.77\text{H}-4.0219$ MJ/kg 187 文献 - -
$\begin{align} & {{Q}_{h}}=2.775+\text{H}+0.004027\text{C}+0.004027{{\text{C}}^{2}}+\frac{0.05706}{\text{H}-12.97}+\frac{0.02323}{\text{H}-6.661}+ \\ & \frac{0.009398}{\text{H}-5.961}+\frac{12.97-\text{H}}{{{\text{H}}^{3}}-5.922\text{C}} \\ \end{align}$ MJ/kg 187 文献 - -
AMEN等[46] ${{Q}_{h}}=4.392+2.514\times {{10}^{5}}{{\text{C}}^{4}}-3.281\times {{10}^{-30}}{{\text{C}}^{22}}$ MJ/kg 36 实测 0.824 -
$\begin{align} & {{Q}_{h}}=11.8\text{HS}+0.2367\text{NO}+12.91{{\text{N}}^{2}}-0.3428- \\ & 0.2871\text{N}{{C}_{(\text{W})}}-0.7196\text{S}{{C}_{(\text{W})}}-66.04\text{NS} \\ \end{align}$ MJ/kg 36 实测 0.644 -
$\begin{align} & {{Q}_{h}}=5.734+41.96\text{S}+0.9586\text{CN}+0.4347\text{O}{{\text{N}}^{2}}- \\ & 22.42\text{N}-52.33\text{NS}-0.01236\text{CN}{{C}_{(\text{W})}} \\ \end{align}$ MJ/kg 36 实测 0.919 -
MATEUS等[37] ${{{Q}'}_{(\text{h})}}=0.254811\text{C}+1.64176\text{H}$ MJ/kg 458 实测 0.992 3.47
$\begin{align} & {{{{Q}'}}_{(\text{h})}}=0.008854{{C}_{(\text{W})}}+0.492754\text{C}+ \\ & 0.614578\text{H}-0.057788\text{O}-5.047684 \\ \end{align}$ MJ/kg 458 实测 0.996 1.55
$\begin{align} & {{{{Q}'}}_{(\text{l})}}=0.008859{{C}_{(\text{W})}}+0.492715\text{C}+ \\ & 0.408739\text{H}-0.057778\text{O}-5.047003 \\ \end{align}$ MJ/kg 458 实测 0.991 1.65
${{{Q}'}_{(\text{l})}}=0.517644\text{C}+0.514339\text{H}-8.215895$ MJ/kg 458 实测 0.989 1.81

备注:${{Q}_{l}}$为低位热值;${{{Q}'}_{(\text{l})}}$为干基低位热值;${{Q}_{\operatorname{h}}}$为高位热值;$Q_{(\operatorname{h})}^{'}$为干基高位热值;C为碳含量,%;H为氢含量,%;O为氧含量,%;S为硫含量,%;Cl为氯含量,%;A为灰分,%;R2为决定系数;MAPE为平均绝对百分比误差。

元素含量分析模型精度通常较高,但模型精度评价指标及方法各不相同,对比较难。理论上,水分对热值产生负影响,但其系数在一些模型中为正[37];硫元素氧化过程为放热反应,但其系数在一些模型中为负[25,34,42,46]。元素分析通常耗时耗力(4 ~ 5 d),对实验设备和操作人员要求高[47]。另外,用于元素分析的样品质量是毫克级,但生活垃圾性质复杂,样品的代表性存在争议[29]

2.2 工业特性分析模型

工业特性分析模型以生活垃圾含水率、挥发分和固定碳为参数[13,48],见表2。基于R2和MAPE的评价结果,这类模型精度较元素含量分析模型低。进行工业特性分析的样本质量为克级,但代表性依然存在问题[29,31]。工业特性分析同样耗时(4 ~ 5 d)耗力,对操作人员技术要求较高。因此,这类模型较少。
Table 2 Summary of heating value predictive models for MSW based on proximate analysis

表2 基于工业特性分析的生活垃圾热值计算模型

模型建立者 公式 单位 数据量 数据来源 模型评价指标
R2 MAPE / %
LIU等[19] ${{Q}_{l}}=44.75\text{V}-5.85{{C}_{(\text{W})}}+21.1$ kCal/kg - - - -
KATHIRAVALE等[29] ${{{Q}'}_{(\text{h})}}=356.248\text{V}-6998.497$ kJ/kg 30 实测 0.682 -
${{{Q}'}_{(\text{h})}}=356.047\text{V}-118.035\text{FC}-5600.613$ kJ/kg 30 实测 0.691 -
THIPKHUNTHOD等[30] ${{{Q}'}_{(\text{h})}}=255.75\text{V}+283.88\text{FC}-2386.38$ kJ/kg 219 实测 0.899 9.1
${{{Q}'}_{(\text{ah})}}=278.07\left( \text{V}+\text{FC} \right)-50.44{{C}_{(\text{W})}}-2875.52$ kJ/kg 219 实测 0.901 8.9
${{{Q}'}_{(\text{h})}}=219.98+327.44\text{FC}-68.39{{C}_{(\text{W})}}$ kJ/kg 219 实测 0.881 11.4
IBIKUNLE等[44] ${{Q}_{h}}=0.151721\text{V}+0.116768\text{FC}-0.34728{{C}_{(\text{W})}}-7.19477$ MJ/kg 62 实测 0.705 -
AMEN等[46] ${{Q}_{h}}=0.744767+0.240652\text{V}+\frac{5.214473}{{{\text{V}}^{2}}}$ MJ/kg 36 实测 0.818 4 -
${{Q}_{h}}=0.184563\text{V}+3.570487$ MJ/kg 36 实测 0.818 4 -

备注:${{{Q}'}_{(\text{ah})}}$为风干基高位热值;FC为固定碳含量,%;V为挥发分,%。

2.3 物理组成分析模型

物理组成分析模型以生活垃圾可燃物理成分(纸、塑料、木竹、食物等)及水分为模型参数[31],见表3,少数模型也将不可燃成分作为参数[49]。纸、塑料、食物三类在生活垃圾中占比大,是绝大多数模型的参数,部分模型简化为只包含这三种参数[20,35]。也有少数模型使用生活垃圾拟组分,如纤维素、木质素、聚乙烯等作为模型参数[50]
生活垃圾物理组成分析方法简单,操作便捷,对操作人员要求较低[19,31,36],因此,这类模型数量较多[51],但精度参差不齐。一般而言,塑料热值高,模型中系数应当大于其他成分,但在部分模型中其系数小于纸类[20,34,51],甚至为负数[10,32],这不符合其化学性质;可能是由于物理组成分析模型只将生活垃圾的大类作为参数,但同一类垃圾热值幅度较大,例如,塑料为17.8 MJ/kg ~ 47.5 MJ/kg,纸类为10.4 MJ/kg ~ 27.3 MJ/kg [5]。生活垃圾的物理成分和热值易受到环境(地理位置、气候、季节等)[58,59,60]、社会经济(工业发展、经济发展、生活水平等)[61,62,63]及垃圾管理方式等因素的影响[64,65,66,67]。因此,物理组成分析模型具有较强的地域性、季节性和时效性。
Table 3 Summary of heating value predictive models for MSW based on physical composition analysis

表3 基于物理成分分析的生活垃圾热值计算模型

模型建立者 公式 单位 数据量 数据 来源 模型评价指标
R2 MAPE / %
KHAN等[35] ${{Q}_{l}}=23\left( \text{Fo}+3.6\text{Pa} \right)+160\text{Pl}$ Btu/lb 86 文献 0.971 -
LIU等[19] ${{{Q}'}_{(\text{l})}}=28.16\text{Pl}+7.90\text{Pa}+4.87\text{Ga}-37.28{{C}_{(\text{W})}}+2229.91$ kCal/kg 34 实测 0.967 -
ABU-QUDAIS等[51] ${{Q}_{(\operatorname{l})}}=267.0\left( {\text{Pl}}/{\text{Pa}}\; \right)+2285.7$ kCal/kg 15 实测 0.940 -
TIAN等[52] $\begin{align} & {{Q}_{(\operatorname{l})}}=\left[ 458\text{Pl}+141.1\left( \text{Te}+\text{Fo}+\text{Pa}+\text{Yr} \right)+8.2\text{A} \right]\times \left( 100-{{C}_{(\text{W})}} \right)/100- \\ & 25\left( {{C}_{(\text{W})}}+9\text{H} \right) \\ \end{align}$ kJ/kg - 实测 - -
董长青等[53] ${{Q}_{l}}=237.79\text{Pl}+95.44\text{Pa}+53.37\text{Te}+18.77\text{Wo}+4.33\text{Fo}+1393.37$ kJ/kg 108 实测 - -
KATHIRAVALE等[29] ${{Q}_{(\operatorname{l})}}=112.157\text{Ga}+183.386\text{Pa}+288.737\text{Pl}+5064.701$ kJ/kg 30 实测 0.779 -
${{Q}_{(\operatorname{l})}}=112.815\text{Ga}+184.366\operatorname{Pa}+298.343\text{Pl}-1.920{{C}_{(\text{W})}}+5130.380$ kJ/kg 30 实测 0.779 -
CHANG等[23] ${{{Q}'}_{(\text{l})}}=\left( 35.19\text{Pa}+71.17\text{Pl}+36.24\text{Te}+48.06\text{Wo}+42.21\text{Fo}+44\text{Mi} \right)\frac{\left( 100-{{C}_{(\text{W})}} \right)}{{{C}_{(\text{W})}}}-6{{C}_{(\text{W})}}$ kCal/kg 180 实测 0.983 5.56
${{{Q}'}_{(\text{l})}}=\left( 39.04\text{Pa}+101.47\text{Pl}+38.47\text{Fo} \right)\frac{100-{{C}_{(\text{W})}}}{{{C}_{(\text{W})}}}-6{{C}_{(\text{W})}}$ kCal/kg 180 实测 0.974 10.7
LIN等[33] ${{{Q}'}_{(\text{l})}}=\left( \begin{align} & 47.3\text{Pa}+58.6\text{Pl}+38.6\text{Te}+32.4\text{Wo}+ \\ & 45.2\text{Fo}+62.3\text{Ru}+50.1\text{Mi} \\ \end{align} \right)\frac{100-{{C}_{(\text{W})}}}{{{C}_{(\text{W})}}}-6{{C}_{(\text{W})}}$ kCal/kg 497 实测 0.987 11.6
${{Q}_{(\operatorname{l})}}=22.1\text{Pa}+28.1\text{Pl}+24.6\text{Te}+12.7\text{Wo}+6.0\text{Fo}+57.4\text{Ru}+17.2\text{Mi}$ kCal/kg 497 实测 0.954 17.7
KHURIATI等[34] ${{Q}_{l}}=2997-4.6\text{Pa}+7\text{Pl}+11\text{Ru}-27\text{Te}+20\text{Wo}-28\text{Yr}-26\text{Fo}-6\text{Mi}$ kCal/kg 24 实测 0.491 -
${{Q}_{l}}=141+23\text{Pa}+8\text{Pl}+40\text{Ru}+49\text{Wo}+2.5\text{Fo}+22\text{Mi}$ kCal/kg 24 实测 0.491 -
OZVEREN [54] ${{Q}_{(\operatorname{l})}}=20\text{Fo}+83\text{Pl}+187\text{Pa}+105\text{Wo}+170\text{Te}$ kJ/kg 89 实测 0.906 15
NWANKWO等[32] ${{Q}_{h}}=17712.04\text{W}{{\text{o}}^{-0.0094}}\text{F}{{\text{o}}^{-0.0063}}\text{L}{{\text{e}}^{0.041}}\text{M}{{\text{i}}^{-0.019}}\text{P}{{\text{a}}^{-0.044}}\text{P}{{\text{l}}^{0.084}}\text{T}{{\text{e}}^{0.025}}$ kJ/kg 10 实测 0.999 -
$\begin{align} & {{Q}_{h}}=22402-25.677\text{Fo}+122.132\text{Le}-56.697\text{Mi}-104.471\text{Pa}+ \\ & 49.728\text{Pl}+4.442\text{Te}-64.129{{C}_{(\text{W})}} \\ \end{align}$ kJ/kg 10 实测 0.994 -
苏肇基等[55] ${{Q}_{l}}=2494.019-22.833{{C}_{(\text{W})}}-5.223\text{Ga}-0.926\text{Pa}+2.129\text{Pl}$ kCal/kg 48 - - 15.78
DRUDI等[56] $\begin{align} & {{Q}_{(\operatorname{l})}}=\left( 13.69\text{Or}+20.94\text{Sa}+37.99\text{Pl}+10.48\text{Pa}+19.27\text{Te} \right)\left( 1-{{C}_{(\text{W})}} \right)- \\ & \left( 2.442-{{C}_{(\text{W})}} \right) \\ \end{align}$ MJ/kg 60 实测 0.993 6.48
IBIKUNLE等[44] ${{Q}_{h}}=0.171002+0.010962\text{Ga}+0.008254\text{Ce}+0.010242\text{Po}$ MJ/kg 62 实测 0.977 -
DRUDI等[36] $\begin{align} & {{Q}_{(\operatorname{l})}}=\left( 16.55\text{Or}+20.42\text{Sa}+36.17\text{Pl}+9.06\text{Pa}+22.81\text{Te} \right)\left( 1-{{C}_{(\text{W})}} \right)- \\
& \left[ 2.442\left( 9\text{H}+{{C}_{(\text{W})}}-9\text{H}\times {{C}_{(\text{W})}} \right) \right] \\ \end{align}$
MJ/kg 36 实测 0.996 5.09
李剑颖[57] $\begin{align} & {{Q}_{(\operatorname{l})}}=5529.832-59.618\text{Fo}+87.144\text{Mi}+78.874\text{Pl}- \\ & 118.693{{C}_{(\text{W})}}+541.542\text{Ot}+38.957\text{Br} \\ \end{align}$ kJ/kg 108 实测 0.972 -
吕永[22] ${{Q}_{(\operatorname{l})}}=\frac{100-{{C}_{(\text{W})}}}{100}\left( 145.6\text{Fo}+160.8\text{Pa}+269.9\text{Pl}+195.5\text{Te} \right)-10.3{{C}_{(\text{W})}}-1210$ kJ/kg 100 实测 - -
WANG等[20] ${{Q}_{l}}=-68.06\text{Fo}+91.77\text{Pa}+52.65\text{Pl}+30.73\text{Te}+34.91\text{Wo}+7342.79$ kJ/kg 151 文献 - 22.18
${{Q}_{l}}=-74.42\text{Fo}+83.20\text{Pa}+67.90\text{Pl}+7669.08$ kJ/kg 151 文献 - 21.94

备注:${{Q}_{(\operatorname{l})}}$为湿基低位热值;Pl为塑料,%;L为落叶,%;Le为皮革,%;Pa为纸,%;Wo为木竹,%;Yr为庭院垃圾,%;Gr为草,%;Ga为普通垃圾,%;Fo为厨余,%;Te为纺织物,%;Mi为杂项,%;Ru为橡胶,%;Or为有机垃圾,%;Sa为卫生垃圾,%;Ce为纤维素,%;Po为聚乙烯,%;Ot为其他垃圾,%;Br为砖块,%。

2.4 其他模型

除上述三类模型,研究者也使用经济、社会和环境因子建立模型。孙巍等[68]以人均收入、国内生产总值、第二产业生产总值、第三产业生产总值和年降雨量为参数,建立了生活垃圾热值人工神经网络模型;杨涛[69]利用燃气利用率、年降雨量、城镇居民人均年生活支出和国内生产总值为参数建立了成都市生活垃圾热值人工神经网络模型,这些模型精度都较高。这类模型的研究可以帮助进一步探索生活垃圾热值和社会、经济及环境之间的关系,并且可以减少采样和分析时人力、物力的投入。但此类模型较少,建模样本小,易产生过度拟合,且少有数学模型,应当进行进一步的研究。

3 模型建立方法

3.1 多元线性回归分析

多元线性回归分析是最常用的建模方法,是一种确定因变量${{Y}_{i\left( 1\le i\le n \right)}}$与一个或多个自变量${{X}_{j\left( 1\le j\le n \right)}}$间关系的统计方法[70],变量间的控制方程为:
${{Y}_{i\left( 1\le i\le n \right)}}=\sum\limits_{j=1}^{m}{{{a}_{j}}{{X}_{j}}}+{{b}_{i\left( 1\le i\le n \right)}}$ (4)
其中:m为自变量个数;n为因变量个数;${{a}_{j\left( 1\le j\le m \right)}}$为回归系数;b为常数系数。多元线性回归分析具有模型参数可解释、易使用、所有参数都可进行统计检测以及给予预测置信区间等优点[71]。但该方法只包含统计显著、有限数量的参数[72]。然而,生活垃圾成分复杂、时空差异大,该方法在建立高精度普适模型上有一定缺陷。

3.2 人工神经网络

人工神经网络具有较强的容错性、学习性、自适应性、快速信息处理能力和非线性映射能力[49]。其结构如图3所示。目前,人工神经网络模型基本采用三层结构(表4)。研究表明,模型精度随变量和样本量增加而提高。因此,人工神经网络理论上可将生活垃圾所有成分作为参数进行建模。
Fig. 3 Structure of artificial neural network

图3 人工神经网络结构示意图

Table 4 Summary of heating value predictive models for MSW built by using artificial neural network

表4 生活垃圾热值人工神经网络模型

模型建立者 参数 热值 单位 人工神经网络结构* 样本量 数据来源 模型精度R2
DONG等[73] 塑料、纸、纺织物、草、厨余 LHV kJ/kg 9∶(3,5,7,9)∶1 105 实测 -
SHU等[47] C、H、N、O、S、Cl LHV kCal/kg 6∶15∶1 220 实测 0.93
水分、塑料、纸、厨余、有机垃圾、纺织物、皮革、可燃物、不可燃物 LHV kCal/kg 10∶15∶1 220 实测 0.87
水分、挥发分、灰分 LHV kCal/kg 3∶15∶1 220 实测 0.83
OGWUELEKA等[74] 橡塑、纸、纺织物、厨余、玻璃 LHV kJ/kg 5∶10∶1 60 实测 0.981
AKKAYA等[41] C、H、O、N、S、灰分、水分 HHV MJ/kg 7∶5∶1 100 文献 0.991
KHURIATI等[75] 塑料、纸、橡胶、纺织物、木竹、庭院垃圾、厨余、可燃物 LHV kCal/kg 8∶11∶1 24 实测 0.988 1
8∶23∶1 0.992 7
8∶35∶1 0.992 9
OZVEREN [54] 塑料、纸、纺织物、木竹、厨余、水分 LHV kJ/kg 6∶16∶4∶1 89 实测 0.974 7
丁兰等[76] 低密度聚乙烯、高密度聚乙烯、聚丙烯、聚苯乙烯、涤纶树脂、聚氯乙烯、聚碳酸酯、其他塑料、纸、橡胶、纺织物、木竹、皮革、水分、干基氢含量 LHV kJ/kg - 78 实测 0.941 7
WANG等[20] 厨余、纸、塑料、纺织物、木竹 LHV kJ/kg 5∶3 ~ 35∶1 151 文献 -
厨余、纸、塑料 LHV kJ/kg 3∶3 ~ 35∶1 151 文献 -
BIRGEN等[77] 温度、降水、风力、一周中第几天、一年中第几周 LHV MJ/kg A/N 1 024 实测 -

注:*人工神经网络的结构为输入层变量数∶隐藏层中的节点个数∶输出层变量个数(例如6∶15∶1)。

人工神经网络模型较数学模型精度高。但人工神经网络相当于一个黑匣子,计算过程和结果不能用燃烧的化学机理进行验证[20],不能计算出标准化系数和变量的系数[72]。并且,网络结构和参数设置尚没有标准的方法来确定,参数的设置和调试受研究者经验和时间限制,建模者需具有一定编程背景[78]。另外,受数据量和复杂度限制[78],人工神经网络常用的交叉验证并未应用于生活垃圾热值模型验证。

4 讨论

4.1 非线性模型的必要性

越来越多研究表明,生活垃圾热值与理化成分的关系并不一定是线性[32,45]。随着社会、经济的发展,生活垃圾成分愈加复杂,简单的线性模型可能无法精确计算热值,研究生活垃圾理化成分与热值的非线性关系是必要的。

4.2 数据对模型精度的影响

数据收集一直是生活垃圾热值模型研究中的难点。一方面由于技术和资金缺乏,另一方面是由于数据访问受限。生活垃圾样品采集费时费力,样本量通常较小,容易造成模型过度拟合,尤其是人工神经网络模型。因此,部分研究收集文献中精度较低的数据建模[20],但可能影响模型精度。从表1 ~ 表3可以看出,模型精度,尤其是普适模型精度还有待提高。

4.3 缺乏普适模型

由于生活垃圾的复杂性和地域性,目前还没有适用于全球生活垃圾热值计算的普适模型。KHAN等[35]收集35个国家86个城市的数据建立了物理组成分析模型(表3),但研究中的热值由杜龙公式算得,且模型建于30年前,已经不适用于当下生活垃圾热值的计算。WANG等[20]收集了11个国家44个城市1990-2015年间的数据建立了物理组成分析模型(表3表4),但精度较低,且大部分数据来源于亚洲发展中国家。因此,普适模型精度有待提高。提高精度的可能方案有:(1)对相同来源的生活垃圾建模,研究表明生活垃圾热值和其来源有很高的相关性[32];(2)对相似发展水平(如人均GDP等)或经济模式(如工业型城市、服务型城市等)的城市建模,研究表明工业增加值较高的城市生活垃圾热值相对较高[52];(3)针对相似自然环境的城市建立模型。

4.4 模型的应用与推广

由于生活垃圾成分和热值地域性强,基于本地样本建立的模型用于其他城市时精度通常较低,而普适模型少且精度较低,应用推广较难。另外,基于生活垃圾理化成分、自然环境和社会经济因素建立的模型具有滞后性,不能实现对热值的实时监测,因此,根据焚烧炉运行参数建模[79,80]和利用深度学习进行图像识别[81]来实时预测生活垃圾热值也在研究当中。

5 结论与展望

生活垃圾热值计算模型的研究中存在着热值报告基准、单位等不统一,模型普适性较低、推广难,数据精度低、样本量小等问题。未来热值研究应加强各地区相关研究领域的合作,提高生活垃圾热值计算模型的精度,加强普适模型的研究。
[1]
KAZA S, YAO L C, BHADA-TATA P, et al.What a waste 2.0: a global snapshot of solid waste management to 2050[M]. Washington, DC: World Bank, 2018.

[2]
KHANDELWAL H, DHAR H, THALLA A K, et al.Application of life cycle assessment in municipal solid waste management: a worldwide critical review[J]. Journal of cleaner production, 2019, 209: 630-654. DOI: 10.1016/j.jclepro.2018.10.233.

[3]
BOGNER J, AHMED M A, DIAZ C, et al. Waste management[M]//METZ B, DAVIDSON O R, BOSCH P R, et al. Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, USA: Cambridge University Press, 2017.

[4]
RYU C.Potential of municipal solid waste for renewable energy production and reduction of greenhouse gas emissions in South Korea[J]. Journal of the air & waste management association, 2010, 60(2): 176-183. DOI: 10.3155/1047-3289.60.2.176.

[5]
SHI H H, MAHINPEY N, AQSHA A, et al.Characterization, thermochemical conversion studies, and heating value modeling of municipal solid waste[J]. Waste management, 2016, 48: 34-47. DOI: 10.1016/j.wasman.2015.09.036.

[6]
DRUDI R, ANTONIO G C, TONELI J T C L, et al.Municipal waste heating value modelling using computational and mathematical techniques[C]//Proceedings of the 25th European Biomass Conference and Exhibition. Stockholm, Sweden, 2017: 241-245.

[7]
RYU C, SHARIFI V N, SWITHENBANK J.Thermal waste treatment for sustainable energy[J]. Proceedings of the institution of civil engineers-engineering sustainability, 2007, 160(3): 133-140. DOI: 10.1680/ENSU.2007.160.3.133.

[8]
CHENG H F, HU Y N.Municipal solid waste (MSW) as a renewable source of energy: current and future practices in China[J]. Bioresource technology, 2010, 101(11): 3816-3824. DOI: 10.1016/j.biortech.2010.01.040.

[9]
BRUNNER P H, RECHBERGER H.Waste to energy-key element for sustainable waste management[J]. Waste management, 2015, 37: 3-12. DOI: 10.1016/j.wasman.2014.02.003.

[10]
OUMAROU M B, SHODIYA S, NGALA G, et al.Statistical modelling of the energy content of municipal solid wastes in Northern Nigeria[J]. Arid zone journal of engineering, technology and environment, 2016, 8(12): 103-109.

[11]
CORVELLEC H, CAMPOS M J Z, ZAPATA P. Infrastructures, lock-in, and sustainable urban development: the case of waste incineration in the Göteborg Metropolitan Area[J]. Journal of cleaner production, 2013, 50: 32-39. DOI: 10.1016/j.jclepro.2012.12.009.

[12]
ZHAO X G, JIANG G W, LI A, et al.Technology, cost, a performance of waste-to-energy incineration industry in China[J]. Renewable and sustainable energy reviews, 2016, 55: 115-130. DOI: 10.1016/j.rser.2015.10.137.

[13]
北极星电力网新闻中心. 国内外垃圾焚烧发电现状及发展趋势(3)[N/OL]. 北极星电力网, (2018-10-15) [2021-09-10]. https://news.bjx.com.cn/html/20181015/933827-3.shtml

[14]
生活垃圾焚烧发电厂自动监测数据公开平台[DB/OL]. [2021-09-10].https://ljgk.envsc.cn/

[15]
范妮. 国内生活垃圾焚烧发电项目研究进展[J/OL]. 湖北大学学报(自然科学版). 2021, 43(6), 690-697. DOI: 10.3969/j.issn.1000-2375.2021.06.014.

[16]
房德职, 李克勋. 国内外生活垃圾焚烧发电技术进展[J]. 发电技术, 2019, 40(4): 367-376. DOI: 10.12096/j.2096-4528.pgt.18234.

[17]
刘洋, 胡彦霞, 张晔, 等. 气化式回转窑焚烧炉设计理念与实践[J]. 工业锅炉, 2021(1): 20-25. DOI: 10.16558/j.cnki.issn1004-8774.2021.01.003.

[18]
张益阳. 热解气化炉技术在国内垃圾焚烧发电工程中的应用[J]. 中国金属通报, 2019(10): 262-263. DOI: 10.3969/j.issn.1672-1667.2019.10.161.

[19]
LIU J I, PAODE R D, HOLSEN T M.Modeling the energy content of municipal solid waste using multiple regression analysis[J]. Journal of the air & waste management association, 1996, 46(7): 650-656. DOI: 10.1080/10473289.1996.10467499.

[20]
WANG D, TANG Y T, HE J, et al.Generalized models to predict the lower heating value (LHV) of municipal solid waste (MSW)[J]. Energy, 2021, 216: 119279. DOI: 10.1016/j.energy.2020.119279.

[21]
PUTNA O, KROPÁC J, FRÝBA L, et al. Prediction of heating value of waste and its importance for conceptual development of new waste-to-energy plants[J]. Chemical engineering transactions, 2014, 39: 1273-1278. DOI: 10.3303/CET1439213.

[22]
吕永. 华南城市生活垃圾热值估算模型分析[J]. 中国资源综合利用, 2020, 38(1): 82-85. DOI: 10.3969/j.issn.1008-9500.2020.01.024.

[23]
CHANG Y F, LIN C J, CHYAN J M, et al.Multiple regression models for the lower heating value of municipal solid waste in Taiwan[J]. Journal of environmental management, 2007, 85(4): 891-899. DOI: 10.1016/j.jenvman.2006.10.025.

[24]
WANG D, HE J, TANG Y T, et al.Life cycle assessment of municipal solid waste management in Nottingham, England: past and future perspectives[J]. Journal of cleaner production, 2020, 251: 119636. DOI: 10.1016/j.jclepro.2019.119636.

[25]
MERAZ L, DOMÍNGUEZ A, KORNHAUSER I, et al. A thermochemical concept-based equation to estimate waste combustion enthalpy from elemental composition[J]. Fuel, 2003, 82(12): 1499-1507. DOI: 10.1016/S0016-2361(03)00075-9.

[26]
PATEL S U, KUMAR B J, BADHE Y P, et al.Estimation of gross calorific value of coals using artificial neural networks[J]. Fuel, 2007, 86(3): 334-344. DOI: 10.1016/j.fuel.2006.07.036.

[27]
BILGEN S, KELEŞ S, KAYGUSUZ K.Calculation of higher and lower heating values and chemical exergy values of liquid products obtained from pyrolysis of hazelnut cupulae[J]. Energy, 2012, 41(1): 380-385. DOI: 10.1016/j.energy.2012.03.001.

[28]
李晓东, 陆胜勇, 徐旭, 等. 中国部分城市生活垃圾热值的分析[J]. 中国环境科学, 2001, 21(2): 156-160. DOI: 10.3321/j.issn:1000-6923.2001.02.016.

[29]
KATHIRAVALE S, YUNUS SSOPIAN M N M, et al. Modeling the heating value of municipal solid waste[J]. Fuel, 2003, 82(9): 1119-1125. DOI: 10.1016/S0016-2361(03)00009-7.

[30]
THIPKHUNTHOD P, MEEYOO V, RANGSUNVIGIT P, et al.Predicting the heating value of sewage sludges in Thailand from proximate and ultimate analyses[J]. Fuel, 2005, 84(7/8): 849-857. DOI: 10.1016/j.fuel.2005.01.003.

[31]
LIN X B, WANG F, CHI Y, et al.A simple method for predicting the lower heating value of municipal solid waste in China based on wet physical composition[J]. Waste management, 2015, 36: 24-32. DOI: 10.1016/j.wasman.2014.11.020.

[32]
NWANKWO C A, AMAH V E.Estimating energy content of municipal solid waste by multiple regression analysis[J]. International journal of science and research, 2016, 5(6): 687-691.

[33]
LIN C J, CHYAN J M, CHEN I M, et al.Swift model for a lower heating value prediction based on wet-based physical components of municipal solid waste[J]. Waste management, 2013, 33(2): 268-276. DOI: 10.1016/j.wasman.2012.11.003.

[34]
KHURIATI A, BUDI W S, NUR M, et al.Modeling of heating value of municipal solid waste based on ultimate analysis using multiple stepwise regresion linear in Semarang[J]. ARPN journal of engineering and applied sciences, 2017, 12(9): 2870-2876.

[35]
KHAN M Z A, ABU-GHARARAH Z H. New approach for estimating energy content of municipal solid waste[J]. Journal of environmental engineering, 1991, 117(3): 376-380. DOI: 10.1061/(ASCE)0733-9372(1991)117:3(376).

[36]
DRUDI K C R, DRUDI R, MARTINS G, et al. Statistical model for heating value of municipal solid waste in Brazil based on gravimetric composition[J]. Waste management, 2019, 87: 782-790. DOI: 10.1016/j.wasman.2019.03.012.

[37]
MATEUS M M, BORDADO J M, DOS SANTOS R G. Simplified multiple linear regression models for the estimation of heating values of refuse derived fuels[J]. Fuel, 2021, 294: 120541. DOI: 10.1016/j.fuel.2021.120541.

[38]
COOPER C D, KIM B, MACDONALD J.Estimating the lower heating values of hazardous and solid wastes[J]. Journal of the air & waste management association, 1999, 49(4): 471-476. DOI: 10.1080/10473289.1999.10463816.

[39]
CHANNIWALA S A, PARIKH P P.A unified correlation for estimating HHV of solid, liquid and gaseous fuels[J]. Fuel, 2002, 81(8): 1051-1063. DOI: 10.1016/S0016-2361(01)00131-4.

[40]
孙晓杰, 徐迪民, 李雄, 等. 上海城市生活垃圾的组成及热值分析[J]. 同济大学学报(自然科学版), 2008, 36(3): 356-360, 378. DOI: 10.3321/j.issn:0253-374X.2008.03.015.

[41]
AKKAYA E, DEMIR A.Energy content estimation of municipal solid waste by multiple regression analysis[C]// Proceedings of the 5th International Advanced Technologies Symposium (IATS'09). Karabuk, Turkey, 2009.

[42]
EBOH F C, AHLSTRÖM P, RICHARDS T. Estimating the specific chemical exergy of municipal solid waste[J]. Energy science & engineering, 2016, 4(3): 217-231. DOI: 10.1002/ESE3.121.

[43]
HAN J, YAO X, ZHAN Y Q, et al.A method for estimating higher heating value of biomass-plastic fuel[J]. Journal of the energy institute, 2017, 90(2): 331-335. DOI: 10.1016/j.joei.2016.01.001.

[44]
IBIKUNLE R A, TITILADUNAYO I F, AKINNULI B O, et al.Modelling the energy content of municipal solid waste and determination of its physicochemical correlation, using multiple regression analysis[J]. International journal of mechanical engineering and technology, 2018, 9(11): 220-232.

[45]
BOUMANCHAR I, CHHITI Y, ALAOUI F E M, et al.Municipal solid waste higher heating value prediction from ultimate analysis using multiple regression and genetic programming techniques[J]. Waste management & research, 2019, 37(6): 578-589. DOI: 10.1177/0734242X18816797.

[46]
AMEN R, HAMEED J, ALBASHAR G, et al.Modelling the higher heating value of municipal solid waste for assessment of waste-to-energy potential: a sustainable case study[J]. Journal of cleaner production, 2021, 287: 125575. DOI: 10.1016/j.jclepro.2020.125575.

[47]
SHU H Y, LU H C, FAN H J, et al.Prediction for energy content of Taiwan municipal solid waste using multilayer perceptron neural networks[J]. Journal of the air & waste management association, 2006, 56(6): 852-858. DOI: 10.1080/10473289.2006.10464497.

[48]
ÖZYUĞURAN A, YAMAN S. Prediction of calorific value of biomass from proximate analysis[J]. Energy procedia, 2017, 107: 130-136. DOI: 10.1016/j.egypro.2016.12.149.

[49]
张瑛华, 张友富, 王洪. 基于神经网络的生活垃圾低位热值计算模型的研究与应用[J]. 电力建设, 2010, 31(9): 94-97. DOI: 10.3969/j.issn.1000-7229.2010.09.024.

[50]
LI Q H, LONG Y Q, ZHOU H, et al.Prediction of higher heating values of combustible solid wastes by pseudo- components and thermal mass coefficients[J]. Thermochimica acta, 2017, 658: 93-100. DOI: 10.1016/j.tca.2017.10.013.

[51]
ABU-QUDAIS M, ABU-QDAIS A. Energy content of municipal solid waste in Jordan and its potential utilization[J]. Energy conversion and management, 2000, 41(9): 983-991. DOI: 10.1016/S0196-8904(99)00155-7.

[52]
TIAN W D, WEI X L, WU D Y, et al.Analysis of ingredient and heating value of municipal solid waste[J]. Journal of environmental sciences, 2001, 13(1): 87-91. DOI: 10.3321/j.issn:1001-0742.2001.01.014.

[53]
董长青, 金保升. 神经网络法用于预测城市生活垃圾热值[J]. 热能动力工程, 2002, 17(3): 275-278. DOI: 10.3969/j.issn.1001-2060.2002.03.016.

[54]
OZVEREN U.An artificial intelligence approach to predict a lower heating value of municipal solid waste[J]. Energy sources, part a: recovery, utilization, and environmental effects, 2016, 38(19): 2906-2913. DOI: 10.1080/15567036.2015.1107864.

[55]
苏肇基, 梁桂花, 祁光霞. 某城市生活垃圾低位热值预测模型比较与改进[J]. 广东化工, 2016, 43(15): 187-188, 186. DOI: 10.3969/j.issn.1007-1865.2016.15.088.

[56]
DRUDI K C R, DRUDI R, MARTINS G, et al. Prediction of lower heating value of wastes of Santo Andre using multivariate regression[C]//Proceedings of the 25th European Biomass Conference and Exhibition. Stockholm, Sweden, 2017. DOI: 10.5071/25thEUBCE2017-1DV.1.81.

[57]
李剑颖. 基于多元线性回归的生活垃圾热值影响因素分析[J]. 环境卫生工程, 2019, 27(4): 35-40. DOI: 10.3969/j.issn.1005-8206.2019.04.009.

[58]
王延涛, 曹阳. 我国城市生活垃圾焚烧发电厂垃圾热值分析[J]. 环境卫生工程, 2019, 27(5): 41-44. DOI: 10.3969/j.issn.1005-8206.2019.05.009.

[59]
朱真真, 王进, 王沛丽, 等. 中国城市入炉生活垃圾热值分析[J]. 新能源进展, 2021, 9(2): 110-114. DOI: 10.3969/j.issn.2095-560X.2021.02.004.

[60]
樊荣, 甘志强, 肖国光, 等. 华东某市生活垃圾组成及热值分析[J]. 广东化工, 2018, 45(11): 40-41. DOI: 10.3969/j.issn.1007-1865.2018.11.018.

[61]
王桂琴, 张红玉, 王典, 等. 北京市城区生活垃圾组成及特性分析[J]. 环境工程, 2018, 36(4): 132-136. DOI: 10.13205/j.hjgc.201804027.

[62]
SOKKA L, ANTIKAINEN R, KAUPPI P E.Municipal solid waste production and composition in Finland— Changes in the period 1960-2002 and prospects until 2020[J]. Resources, conservation and recycling, 2007, 50(4): 475-488. DOI: 10.1016/j.resconrec.2007.01.011.

[63]
DEN BOER E, JĘDRCZAK A, KOWALSKI Z, et al. A review of municipal solid waste composition and quantities in Poland[J]. Waste management, 2010, 30(3): 369-377. DOI: 10.1016/j.wasman.2009.09.018.

[64]
ZHOU X, ZHOU P, ZHAO X Q, et al.Applicability of municipal solid waste incineration (MSWI) system integrated with pre-drying or torrefaction for flue gas waste heat recovery[J]. Energy, 2021, 224: 120157. DOI: 10.1016/j.energy.2021.120157.

[65]
YANG Y, LIEW R K, TAMOTHRAN A M, et al.Gasification of refuse-derived fuel from municipal solid waste for energy production: a review[J]. Environmental chemistry letters, 2021, 19(3): 2127-2140. DOI: 10.1007/s10311-020-01177-5.

[66]
KAUR A, BHARTI R, SHARMA R. Municipal solid waste as a source of energy[J]. Materials today, 2021, in press. DOI: 10.1016/j.matpr.2021.04.286.

[67]
DAS S, LEE S H, KUMAR P, et al.Solid waste management: scope and the challenge of sustainability[J]. Journal of cleaner production, 2019, 228: 658-678. DOI: 10.1016/j.jclepro.2019.04.323.

[68]
孙巍, 陶怀志, 陈晓春. 人工神经元网络在市政垃圾热值预测中的应用[J]. 中国环保产业, 2005(8): 34-36. DOI: 10.3969/j.issn.1006-5377.2005.08.017.

[69]
杨涛. 基于经济统计数据的生活垃圾热值计算模型[J]. 环境工程技术学报, 2014, 4(2): 158-163. DOI: 10.3969/j.issn.1674-991X.2014.02.026.

[70]
BOUMANCHAR I, CHHITI Y, ALAOUI F E M, et al.Effect of materials mixture on the higher heating value: case of biomass, biochar and municipal solid waste[J]. Waste management, 2017, 61: 78-86. DOI: 10.1016/j.wasman.2016.11.012.

[71]
WANG Y M, ELHAG T M S. A comparison of neural network, evidential reasoning and multiple regression analysis in modelling bridge risks[J]. Expert systems with applications, 2007, 32(2): 336-348. DOI: 10.1016/j.eswa.2005.11.029.

[72]
EFTEKHAR B, MOHAMMAD K, ARDEBILI H E, et al.Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data[J]. BMC medical informatics and decision making, 2005, 5: 3. DOI: 10.1186/1472-6947-5-3.

[73]
DONG C Q, JIN B S, LI D J.Predicting the heating value of MSW with a feed forward neural network[J]. Waste management, 2003, 23(2): 103-106. DOI: 10.1016/S0956-053X(02)00162-9.

[74]
OGWUELEKA T C, OGWUELEKA F N.Modelling energy content of municipal solid waste using artificial neural network[J]. Iranian journal of environmental health science and engineering, 2010, 7(3): 259-266.

[75]
KHURIATI A, SETIABUDI W, NUR M, et al.Heating value prediction for combustible fraction of municipal solid waste in Semarang using backpropagation neural network[J]. AIP conference proceedings, 2015, 1699(1): 030028. DOI: 10.1063/1.4938313.

[76]
丁兰, 张文阳, 张良均, 等. 基于人工神经网络的居民生活垃圾可燃成分热值预测[J]. 环境工程学报, 2016, 10(2): 899-905.

[77]
BIRGEN C, MAGNANELLI E, CARLSSON P, et al.Machine learning based modelling for lower heating value prediction of municipal solid waste[J]. Fuel, 2021, 283: 118906. DOI: 10.1016/j.fuel.2020.118906.

[78]
张驰, 郭媛, 黎明. 人工神经网络模型发展及应用综述[J]. 计算机工程与应用, 2021, 57(11): 57-69. DOI: 10.3778/j.issn.1002-8331.2102-0256.

[79]
谢承利, 陆继东, 沈凯, 等. 基于焚烧运行参数的垃圾热值软测量模型[J]. 燃烧科学与技术, 2007, 13(1): 81-85. DOI: 10.3321/j.issn:1006-8740.2007.01.017.

[80]
马晓茜, 谢泽琼. 基于BP神经网络的垃圾热值预测模型[J]. 科技导报, 2012, 30(23): 46-50. DOI: 10.3981/j.issn.1000-7857.2012.23.006.

[81]
谢昊源, 黄群星, 林晓青, 等. 基于图像深度学习的垃圾热值预测研究[J]. 化工学报, 2021, 72(5): 2773-2782. DOI: 10.11949/0438-1157.20201481.

Outlines

/