基于BP神经网络和激光诱导击穿光谱的燃煤热值快速测量方法研究

作为煤质评价的重要指标之一,热值的快速、准确测量对电厂燃煤锅炉的优化燃烧和经济运行至关重要。采用激光诱导击穿光谱(LIBS)技术结合BP神经网络定量分析模型和聚类分析,以35个煤粉样品作为研究对象进行热值的定量分析。基体效应对LIBS光谱数据的显著影响,针对基于某类煤粉样品所建立的定标曲线不能直接用于不同煤种的定量分析,采用K-means聚类方法根据热值、灰分、挥发分把样品分为三类对训练集和预测集样品进行优化选择。通过谱线强度和热值变量相关性分析,同时考虑特征谱线的物理意义,最终提取12条元素谱线的峰值强度作为输入参数,建立BP神经网络模型对燃煤热值进行预测。定标结果表明,建立的神经网络模型具有良好的定量分析能力,定标曲线拟合度R2为0.996,热值预测值的相对误差低于3.42%,多次重复测量的相对标准偏差在4.23%以内。对聚类分析中3类样品具有不同的预测能力,采用峰值强度作为输入参数时,能够在一定程度上减弱试验参数波动和基体效应造成的影响。定量分析结果的重复性和准确性可以通过对不同类别的煤种分别建立BP神经网络模型来进一步改善。LIBS技术结合BP神经网络可以对煤粉热值进行定量分析,在现场在线/快速检测领域具有很好的应用价值和潜力。     

Influence of combustion temperature and coal types on alumina crystal phase formation of high-alumina coal ash

The alumina content (more than 40%) of high-alumina coal ash is comparative to the middle content bauxite ores in China. So far, in order to meet the high demand of alumina and the rise of circular economy industrial chain, extracting alumina from coal ash has become a way to comprehensively utilize high-alumina coal ash. However, this process has high requirements on the crystal phase and stability of alumina. Different from most studies, this paper focuses on how to produce coal ash more beneficial to the later refining of aluminum. Therefore, the effects of combustion temperature and coal types by classifying high-alumina coal into dull coal and bright coal on alumina crystal phase formation were studied. Through proximate analysis, ultimate analysis, calorific value analysis, X-ray fluorescence spectroscopy, X-ray diffraction (XRD) and scanning electron microscope (SEM) and other methods, it is found that γ-Al₂O₃ in high-alumina coal ash translated into more stable θ-Al₂O₃ and finally α-Al₂O₃ when combustion temperature is higher than 1000°C. Thus compared with pulverized coal boilers, circulating fluidized bed (CFB) boilers with lower combustion temperature can produce higher quality coal ash. Moreover, at the same combustion temperature, alumina crystal phase in dull coal ash is relatively less stable than that in bright coal ash, which is more suitable to the later refining and electrolysis of aluminum.     

激光诱导击穿光谱技术快速探测煤灰中的重金属锌

煤灰的成分指的是煤中矿物质的完全燃烧,产生各种金属和非金属氧化物和盐,这是使用煤时的重要参数。煤被广泛用于生产和人民生活,作为重要的能源物质。大量来自燃煤燃烧的煤尘（煤灰）被释放到大气中并与大气中的各种物质相互作用而形成雾霾。煤灰中的金属氧化物和空气中的小液滴之间发生一系列物理化学反应,这导致了雾霾的形成。在实验中,采用激光诱导击穿光谱（LIBS）分析煤灰中的元素。实验样品由某钢铁公司提供,分为七个样品,并标上序号。样品分别加入蒸馏水和0.1‰,0.2‰,0.2％,0.4％,0.8％,1％硫酸锌溶液,分别用1～7号标记。为了获得更好的LIBS信号,样品被研磨为粉末状,并使用蒸馏水使硫酸锌与煤灰充分混合。通过使用压片机将煤灰压制成10 mm直径和10 mm厚的煤灰块。为获得准确的元素结果,X射线荧光光谱也被用作参考,并且原始样品不含锌元素。由于光谱分析和波长漂移现象的不确定性,因此实验中,分别选择了铁,钙,钛和铝四种高纯单质。在相同的实验条件下,将四条测量的元素谱线与NIST原子光谱数据库中相应的谱图比较。实验中的所有光谱根据波长差或偏移进行校正。此时,纯单质的元素谱线可以与样品的光谱对齐。当元素谱中的特征线与样品中的谱线对齐时,样品就可以被识别和确认。由于铝元素与目标元素具有相似的化学和物理性质,铝元素是煤灰和地壳中的主要元素之一,具有中等的光谱强度。因此将铝元素作为内标元素,运用内标校准方法来确定样品中锌的浓度。模拟含锌大气气溶胶是通过向煤灰中添加含锌元素来实现的。还有一些其他的金属元素,包括铁,钙,锰,钛和铝也被用来加入煤灰中,用以模拟大气气溶胶。两种方法的相对差异分别为1.78％,3.39％,5.17％,0.20％。造成差异的原因可能是由于光谱仪缺乏分辨率或背景噪声的影响,这是可能导致测量误差的原因之一。由于实验室条件的限制,无法确定基底是否会影响实验结果,这将在未来的实验中得到进一步的证实。实验拟合曲线测得煤灰中锌的线性相关系数为0.995 72,这表明可以通过粗略估算锌的激光强度来估计煤灰中的锌含量的实现。实验结果证明LIBS技术可用于煤灰中金属元素的快速检测,为基于锌含量的大气环境检测提供了一种新方法。在建立元素的校准曲线后,LIBS技术将来可以用来进行更快速,更准确的定量分析。     

Properties of biomass vs. coal fly ashes deposited in electrostatic precipitator

This paper presents comparative experimental studies of the morphology and elemental composition of fly ash particles from coal- and biomass-fired boilers, deposited in each stage of 3-stage electrostatic precipitators (ESPs). It was shown that fly ash morphology, its physical properties, and the percentage of elements in the fly ash taken from each stage of ESP depend on the kind of fuel. The biomass fly ash contains many irregular large particles, which are pieces of unburned wood. Bulk density of biomass fly ash is on average lower than that of coal fly ash, and drastically decreases in the second and third stages of ESP. The resistivity, measured at electric field of 4 kV/cm, of fly ash from biomass-fired boilers is much lower than that from coal, and can be below 10² Ω m, whereas from coal, except the first stage, varies in the range from 10⁷ to 10¹⁰ Ω m. The low resistivity of coal fly ash in the first stage of ESP results from high carbon content, and of biomass is probably an effect of additional high percentage of potassium, calcium and sodium sulfates. The percentage of Si, Al, Na, Fe, and Ti in fly ash from coal-fired boilers is much higher than from biomass, and in the opposite, the percentage of Mg, K, Ca, Mn, Mo, S, Cl, and P in biomass ash exceeds that in coal fly ash. Potential detrimental effects of biomass combustion products (salts, acids, tar) leaving the boiler on the construction elements of the electrostatic precipitator, including electrodes and HV insulators have been discussed in this paper. It was concluded that the long-term effects of biomass co-firing on the electrostatic precipitator performance, including the collection efficiency, have not been sufficiently studied in the literature and these issues require further detailed investigations.     

准东煤中碱金属含量的LIBS激光测量

电站锅炉燃用准东煤后存在的严重结渣、积灰等问题跟煤中高含量的碱金属(Na,K)有关。采用化学方法对准东煤进行处理以制取一定碱金属含量范围的煤样本,采用激光诱导击穿光谱(LIBS)技术对准东煤的Na,K含量进行定标与预测。结果表明:LIBS技术对Na,K元素的测量具有较高的灵敏度、较低的检测极限和较小的预测误差均方根(RMSEP)。尽管元素含量较低时,LIBS测量数据的相对标准偏差(RSD)较大,但当元素含量达到一定值后,其RSD趋于稳定,通过大量的重复测量可以保证测量的正确性。由于LIBS技术本身有快速同位测量的优点,因此LIBS可以作为在线测量煤中碱金属含量的一种手段。     

煤粉发热量近红外光谱检测的预处理方法研究

煤炭的发热量是评价煤炭品质的重要指标。首先对比分析了平滑处理、微分处理、多元散射校正(MSC)以及标准归一化(SNV)等光谱与处理方法在改善煤粉近红外漫反射光谱信噪比的效果,然后利用偏最小二乘法(PLS)和主成分分析方法(PCR)分别对采用每种预处理方法处理后的光谱建立煤粉发热量模型,发现采用5点平滑处理、多元散射校正和标准归一化处理可使模型的性能有较明显的改观,5点平滑效果最好,相关系数、校正标准差和预测标准差分别为：0.989 9,0.0004 9和0.0005 2,采用25点平滑处理产生了过平滑现象,导致模型的性能变坏,采用微分预处理后的光谱建立的模型没有明显变化,对模型的性能影响不大。     

激光诱导击穿光谱精确测定燃煤工业分析指标的研究

燃煤工业指标的在线精确分析对于指导燃煤工业优化生产、降低燃煤煤耗至关重要。利用激光诱导击穿光谱(LIBS)分析燃煤煤质时,因受我国复杂多样煤种所导致的“基体效应”,测量精度有待提高。实验中对激光诱导燃煤等离子体光谱至燃煤工业分析指标转化过程中的光谱预处理和定标建模方法进行了优化选择。实验结果表明,利用单/多峰Lorentzian光谱拟合计算谱线强度相比于传统计算方法,谱线强度RSD均值可由12.1%降至9.7%;对于核函数参数寻优,相比于网格参数(Grid)和遗传算法(GA),粒子群算法(PSO)的平均绝对误差(MAE)最小;采用PSO参数寻优式支持向量机(SVM)回归建模的预测均方根误差(RMSEP)小于偏最小二乘回归分析法(PLS);采用单/多峰Lorentzian光谱拟合方法和PSO参数寻优式SVM回归建模,对燃煤工业分析指标预测的平均绝对误差(AAE)为：灰分为16%～30%时AAE为1.37%,灰分大于30%时AAE为1.77%,发热量为9～24 MJ·kg-1时AAE为0.65 MJ·kg-1,挥发分低于20%时AAE为1.09%,挥发分大于20%时AAE为1.02%。     

An investigation of the effect of ultrasonic waves on the efficiency of silicon extraction from coal fly ash

Burning of coal accounts for an enormous proportion of the current energy supply, especially in developing countries. Burning of coal produces large amounts of coal fly ash, which causes serious environmental problems unless it is managed properly. Using chemical analysis, we found that coal fly ash could be a promising source of Si, Al, Ca and some rare earth elements, especially with the assistance of some measures such as ultrasound. In this study, we extracted silicon from coal fly ash using an alkaline dissolution strategy and investigated the effects of temperature and ultrasonic power on the efficiency of silicon extraction. During a 70 min reaction, the efficiency of silicon extraction increased markedly, from 9.41% to 34.96%, as the reaction temperature increased from 70 °C to 110 °C. With ultrasound assistance, ultrasonic waves enhanced the extraction of silicon at both 80 °C and 110 °C at 720 W ultrasound, increasing the efficiency of silicon extraction from 6.01% to 15.36% and from 34.96% to 54.42%, respectively. However, at 900 W ultrasonic power, extraction was slightly inhibited at both temperatures, causing a little decrease in efficiency.     

不同来源煤灰激光诱导击穿光谱测量特性对比研究

飞灰含碳量的定量分析需要不同含碳量梯度的飞灰样品作为定标之用。通常做法是用煤粉按照快速灰化法的要求灼烧得到不同含碳量的灰样,用于定标分析。但是这与实际锅炉飞灰的成分存在一定差异,需要对光谱特性的差异进行研究来指导实际定标工作。因此,对比了快速灰化法制备的不同含碳量的煤灰样品与锅炉飞灰在特征谱线强度、等离子温度等等离子光谱特征方面的差异。实验证明快速灰化法制备的煤灰样品的Fe,Mg和Al谱线强度强于锅炉飞灰样品,飞灰等离子温度低于所制备的煤灰样品的等离子体温度,这可能是不同处理过程的物理化学特性差异造成的。用主成分分析法考察了导致光谱差异的主要原因,认为Fe,Mg,Al和Si等元素是导致二者在光谱特性差异的主要因素,这可能因为实验室内按照快速灰化法进行制样的飞灰相应矿物质组成不同所导致的。在用灼烧后的煤灰定标未燃碳时,应注意由于成灰过程不同所造成的Fe,Mg,Al和Si等元素含量和形态不同所带来的影响。     

Insight of arsenic transformation behavior during high-arsenic coal combustion

The release of arsenic vapors (As³⁺) during high-arsenic coal combustion not only raises serious environmental concerns, but also causes catalyst deactivation in selective catalytic reduction (SCR) systems. To illuminate the mechanisms involved in the transformation of arsenic vapors towards less troublesome arsenates (As⁵⁺) during coal combustion, the accessory minerals in the high-arsenic coal were identified and the association relationship of these compounds with arsenic in fly ash was estimated. The results showed that Si/Al were the main inorganic elements in high-arsenic coal while the content of Ca was quite low. Ca was mostly transformed into sulfates during coal combustion and the effect of Ca on the arsenic transformation was limited. Al/Fe played a more significant role in arsenic speciation transformation and arsenic in the fly ash was predominantly bound with Al/Fe-oxides as arsenates. It was further confirmed that Al in kaolin/metakaolin showed good capacity on arsenic capture. In addition, few arsenic vapors were captured through the physical adsorption mechanism and the large fraction of As³⁺ in some fine particles was mostly attributed to the chemical reactions between arsenic vapors and Al-compounds. Meanwhile, a certain amount of arsenic vapors were converted into As₂O₅(s) under the influence of SCR catalyst and then carried by flue gas to participate in fly ash. Besides, part of arsenic distributed in the fly ash was through the stabilization of ash matrix in high temperature conditions. The transformation of arsenic from vapors towards particulate arsenic favored arsenic emission control by particulate matter control devices.     

光谱对煤粉发热量检测系统的研究

在现有近红外检测技术的基础上,设计了一种基于近红外的煤粉发热量检测系统。选取100个煤粉样品建立煤粉发热量的近红外模型,利用验证集的50个样本对模型的精度和稳定性进行了验证分析,结果表明:煤粉发热量的预测值与真实值的相关系数达到0.995 8,相对偏差小于2%。系统具有良好的预测精度和稳定性,能够满足对煤粉发热量的快速检测的需求,而且该系统体积小、结构简单、操作方便,具有很好的可移植性。     

Ash formation and deposition during combustion of pulverized torrefied wood and coal in a 100 kW downflow combustor

Torrefied wood originating from beetle-killed trees is an abundant biomass fuel that can be co-fired with coal for power generation. In this work, pulverized torrefied wood, a bituminous coal (Sufco coal) and their blended fuel with a mixing ratio of 50/50 wt.%, are burned in a 100-kW rated laboratory combustor under similar conditions. Ash aerosols in the flue gas and ash deposits on a temperature-controlled surface are sampled during combustion of the three fuels. Results show that ash formation and deposition for wood combustion are notably different from those for coal combustion, revealing different mechanisms. Compared to the coal, the low-ash torrefied wood produces low concentrations of fly ash in the flue gas but significantly increased yields (per input ash) of ash that has been vaporized. All the mineral elements including the semi- or non-volatile metals in the wood are found to be more readily partitioned into the PM₁₀ ash than those in the coal. The inside layer deposits sticking to the surface and the loosely bound outside deposits exposed to the gas both show a linear growth in weight during torrefied wood test. Unlike coal combustion, in which the concentration of (vaporized) ash PM₁ controls the inside deposition rate, wood combustion shows that the formation of porous bulky deposits by the condensed residual ash dominates the inside deposition process. Co-firing removes these differences between the wood and coal, making the blended fuel to have more similar fly ash characteristics and ash deposition behavior to those of the bituminous coal. In addition, results also show some beneficial effects of co-firing coal with torrefied wood, including reduction of the total deposition rate and the minimization of corrosive alkali species produced by wood.     

Mechanism on the contribution of coal/char fragmentation to fly ash formation during pulverized coal combustion

In this paper, the correlations between coal/char fragmentation and fly ash formation during pulverized coal combustion are investigated. We observed an explosion-like fragmentation of Zhundong coal in the early devolatilization stage by means of high-speed photography in the Hencken flat-flame burner. While high ash-fusion (HAF) bituminous and coal-derived char samples only undergo gentle perimeter fragmentation in the char burning stage. Simultaneously, combustion experiments of two kinds of coals were conducted in a 25?kW down-fired combustor. The particle size distributions (PSDs) of both fine particulates (PM_1-10) and bulk fly ash (PM₁₀₊) were measured by Electrical Low Pressure Impactor (ELPI) and Malvern Mastersizer 2000, respectively. The results show that the mass PSD of residual fly ash (PM₁₊) from Zhundong coal exhibits a bi-modal shape with two peaks located at 14?µm and 102?µm, whereas that from HAF coal only possesses a single peak at 74?µm. A hybrid model accounting for multiple-route ash formation processes is developed to predict the PSD of fly ash during coal combustion. By incorporating coal/char fragmentation sub-models, the simulation can quantitatively reproduce the measured PM₁₊ PSDs for different kinds of coals. The sensitivity analysis further reveals that the bi-modal mass distribution of PM₁₊ intrinsically results from the coal fragmentation during devolatilization.     

Application of LIBS and TMA for the determination of combustion predictive indices of coals and coal blends

Presented work brings results of Laser-Induced Breakdown Spectroscopy (LIBS) and Thermo-Mechanical Analysis (TMA) of coals and coal blends used in coal fired power plants all over Spain. Several coal specimens, its blends and corresponding laboratory ash were analyzed by mentioned techniques and results were compared to standard laboratory methods. The indices of slagging, which predict the tendency of coal ash deposition on the boiler walls, were determined by means of standard chemical analysis, LIBS and TMA. The optimal coal suitable to be blended with the problematic national lignite coal was suggested in order to diminish the slagging problems.Used techniques were evaluated based on the precision, acquisition time, extension and quality of information they could provide. Finally, the applicability of LIBS and TMA to the successful calculation of slagging indices is discussed and their substitution of time-consuming and instrumentally difficult standard methods is considered.     

An in-situ method for time-resolved sodium release behaviour during coal combustion and its application in industrial coal-fired boilers

To mitigate the slagging, fouling and high-temperature corrosion problems caused by alkali metals during coal combustion process, measurement of time-resolved alkali metals release is very important. The paper proposed an in-situ approach for measuring sodium (Na) release in coal combustion by Flame Emission Spectroscopy (FES). Through the analysis of spontaneous emission spectra and a calibration procedure, the concentration of gas phase Na, temperature and thermal radiation can be obtained. Firstly, experimental measurement of Zhundong coal particles burning in a flat flame burner was done. Two kinds of Zhundong coal with similar proximate and ultimate analyses, but different ash composition were used. The Na-release history measured by FES was compared with that by LIBS. Results showed that the Na-release at the devolatilization, char, and ash stages can be distinguished by FES. The higher Si/Al content in ash can suppress the Na-release at the ash stage. Moreover, FES method was extended to the measurement of Na-release in four industrial boiler furnaces of two Zhundong coal-fired power plants. Results showed the Na-release measured by FES can reflect the change of fuel and load, and both temperature and thermal radiation play key roles in Na-release in coal combustion.     

In situ measurements of the spectral emittance of coal ash deposits

The spectral emittance of deposits left by bituminous and sub-bituminous coals under oxidizing conditions have been measured in situ. Pulverized coal is injected into a down-fired entrained-flow reactor. Ash accumulates on a probe in the reactor effluent and radiation emitted by the ash layer is recorded using a Fourier transform infrared (FTIR) spectrometer. Values for the spectral emissive power emitted by the ash and the surface temperature of the ash are extracted from these data. These results are then used to calculate the spectral emittance of the deposit. The spectral emittances of ash deposits formed by burning Illinois #6 (bituminous) coal and Powder River Basin (sub-bituminous) coal were measured between 3000 and 500 wavenumbers. The spectral emittance of the deposit left by the bituminous coal has a constant value of approximately 0.46 between 3000 and 2400 wavenumbers. Between 2200 and 1200 wavenumbers, the spectral emittance of the deposit increases from approximately 0.47 to approximately 0.61. Between 1200 and 500 wavenumbers, the spectral emittance is relatively constant at 0.61. The spectral emittance of the deposit left by the sub-bituminous coal is also relatively constant between 3000 and 2400 wavenumbers at a value of 0.29. Between 2200 and 500 wavenumbers, the spectral emittance of deposits from the sub-bituminous coal increases from approximately 0.29 to 0.55. Differences between these spectral emittance measurements and those measured ex situ illustrate the importance of making in situ measurements. Band emittances were calculated using the measured spectral emittances, and band emittances of the deposits are reported as functions of temperature.     

直接酸提取电感耦合等离子体发射光谱法测定煤中主要成灰 …

本文提出了一种全新的以酸直接提取煤中的矿物元素、ＩＣＰ－ＡＥＳ测定其含量的方法。考察了最佳酸配比、酸用量、最佳提取时间、仪器最佳操作参数以及元素间的干扰等诸多因素。与其灰分含量相比,煤中矿物质的提取率在９５％以上,各元素加标回收率为９６％～１０４％,九次测定相对标准偏差小于４％。方法准确、快速、适合于不同煤样中矿物元素的分析测定。     

The fluorescence interference in Raman spectrum of raw coals and its application for evaluating coal property and combustion characteristics

Fluorescence interference in Raman spectrum is a big barrier for rapid and precise analysis of coal structures by Raman spectroscopy. Dealing with fluorescence interference suitably is one of the key tasks before efficient application of Raman spectroscopy in coal assessment. In this study, Raman spectra and coal combustion characteristics of 32 kinds of Chinese coals were respectively obtained in a micro-Raman spectrometer and Thermal Gravimetric Analyzer. The degree of fluorescence interference in Raman spectrum was firstly defined and quantified as the drift coefficient α using a simple method without curve-fitting the spectrum. The correlations between the degree of fluorescence interference and coal property, coal combustion characteristics were set up and multivariable analysis was done. The results indicate that the degree of fluorescence interference is well related to the coal structures, and it is synthetically determined by volatile, moisture and ash content in coal. With the increase of volatile, moisture content in coal, the fluorescence interference increases continuously, and it can be reduced but not eliminated by drying the moisture in coals. Significant mathematical relations between the drift coefficient α and volatile, moisture content, coal combustion characteristic temperatures have been found. Coal with more evident fluorescence interference in Raman spectrum usually has lower degree of coalification, more polar functional groups, and burns at a lower temperature. The drift coefficient α can act as an efficient probe for coal property and coal combustion characteristics. This study provided a new and simple approach for evaluating coal property and coal combustion characteristics by fluorescence interference in Raman spectrum.     

Partitioning and transformation behavior of selenium during coal combustion

The partitioning of selenium in coal-fired flue gas and desulfurization wastewater is of great threat to the ecological environment and human health. However, the unclear understanding of interactions between selenium vapors and fly ash hinders the emission control of selenium from coal-fired power plants. To further illuminate the mechanism of selenium partitioning and transformation, this study carefully estimated selenium distribution characteristics in the coal combustion byproducts from several industrial power plants. The effective temperature range as well as the key ash components for selenium retention by fly ash was clarified by multiple-scale experiments and theoretical perspectives. The results showed that gaseous selenium tended to be captured by fly ash at a medium-to-low temperature range (i.e. below 650 °C). The limited residence time resulted in the incomplete capture of gaseous selenium by fly ash. Mullite, quartz, iron oxides, and anhydrite in fly ash were found to be the main trappers for gaseous selenium. Among these components, iron oxides showed excellent selenium adsorption performance at a wide temperature range of 150-700 °C, which was realized by the strong chemical adsorption. By contrast, as the dominant phases in fly ash for the physical adsorption of gaseous selenium, mullite and quartz mainly captured gaseous selenium below 300 °C. On the other hand, sulfur dioxides had priority over gaseous selenium to react with calcium-containing ash components by forming anhydrite in the high-temperature region. The formed anhydrite had a limited selenium adsorption capacity, which was confirmed to capture gaseous selenium through a combination of physical adsorption and weak chemical adsorption. For the in-depth control of selenium emitted into desulfurization system and atmosphere environment, these findings provided a comprehensive insight into the behavior of selenium partitioning and transformation into fly ash during coal combustion.     

炉内喷钙脱硫对锅炉热效率的影响

根据炉内喷钙脱硫的化学反应过程,经过化学分析和数学推导,得出了锅炉采用炉内喷钙脱硫后化学反应热、飞灰量和烟气量的增加对锅炉热效率影响的计算公式。采用三种试样的试验结果,定量分析了煤的含硫量不同时,随着钙硫摩尔比的变化喷钙脱硫引起的热损失的变化。结果表明,化学反应热损失大于飞灰和烟气量增加造成的热损失,当煤的含硫量为3.46％时,可使锅炉热效率降低1％左右。