玉米、小麦秸秆原料、热解过程及固体产物特性NIRS快速分析研究

发布时间：2017-12-27 02:24

  本文关键词：玉米、小麦秸秆原料、热解过程及固体产物特性NIRS快速分析研究　出处：《中国农业大学》2016年博士论文　论文类型：学位论文

  更多相关文章： 玉米秸秆 小麦秸秆 热解 基础特性 快速分析

【摘要】：综合高效利用生物质是解决能源危机和环境污染两大问题的有效方法之一,其中,生物质热解是一项基本和有效的利用技术。热解利用的基础特性主要包括原料组分和热值、热解过程参数和热解产物的燃料特性。这些基础特性的检测可为评价生物质适用性,了解反应过程和机理,预测反应速率及难易程度,指导实际生物质热解设备和工艺的工程设计,有效控制生物质热解提供数据和技术支持。传统的检测方法费时费力,近红外光谱技术(NIRS)作为一种快速、高效的检测方法,对生物质热解利用中原料、过程和产物基础特性的快速定量检测具有很大潜力。本研究选取主要农作物秸秆玉米、小麦秸秆,基于NIRS对不同粒度原料组分和热值进行实验室和在线定量预测研究,对热解特性、热解活化能和低温热解固体产物的燃料特性进行快速定量预测研究。研究结果表明,NIRS可以快速定量预测原料的组分和热值、热解特性、热解活化能和低温热解固体产物的燃料特性。论文取得的主要创新性成果有：1、NIRS可以快速定量预测玉米、小麦秸秆粗粉原料的纤维素、半纤维素、木质素、可溶性糖、水分、灰分、挥发分、固定碳、C、H、N、O、K、Mg和热值,而对于S含量的定量预测,模型需进一步研究。粗粉光谱模型与细粉光谱模型相比,细粉模型精度高于或与粗粉模型精度相当,粗粉和细粉最优模型的光谱预处理方式不同,表明样品状态不同,需要不同预处理方法。2、在线光谱采集参数对不同近红外光谱仪器的光谱重复性影响不同,因此,对不同光谱仪需采用不同采集参数,以保证光谱质量。采用优化后光谱仪和采集参数,建立的在线模型可以快速定量预测样品的纤维素、半纤维素、木质素、可溶性糖、水分、灰分、挥发分、固定碳、C、H、 O、N、K、Mg、热值,而对于S的预测效果较差。与静态粗粉模型相比,在线模型精度与静态粗粉模型相当或略好。3、不同热解速率下的热解过程特性存在显著区别,小麦秸秆与玉米秸秆的热解过程特性存在显著区别。NIRS可以快速定量预测热解外推起始温度、外推结束温度、总失重率、失重率、峰值速率、最大峰温度,模型的RSD都小于10%。玉米秸秆的活化能显著高于小麦秸秆。活化能随转化率的增加先增大后减小再增大。峰值转化率点对应的活化能随升温速率增加无显著变化。NIRS可以快速定量预测其热解平均活化能及转化率为0.3-0.6阶段的活化能,模型RSD都小于10%。4、不同终温对低温热解固体产物的燃料特性影响非常显著。相对于终温来说,不同升温速率,不同氮气吹扫速率对低温热解固体产物的燃料特性的影响较小。玉米秸秆低温热解固体产物和小麦秸秆低温热解固体产物的燃料特性有显著区别。利用NIRS可以快速定量预测玉米秸秆和小麦秸秆低温热解固体产物的能量产率、质量产率、热值、挥发分、固定碳、灰分、C、H、O、 N、燃料比率,其模型交互验证RSD分别为4.66%、5.12%、3.29%、7.01%、7.61%、7.44%、2.10%、8.18%、5.31%、7.02%、11.89%,对水分和S含量的预测精度较低,需进一步研究。
[Abstract]:Comprehensive and efficient utilization of biomass is one of the effective ways to solve the two major problems of energy crisis and environmental pollution. Biomass pyrolysis is a basic and effective utilization technology. The basic characteristics of the pyrolysis use mainly include the component of the raw material and the calorific value, the parameters of the pyrolysis process and the fuel characteristics of the pyrolysis products. The detection of these basic characteristics can provide data and technical support for evaluating biomass applicability, understanding reaction process and mechanism, predicting reaction rate and difficulty degree, guiding practical engineering design of biomass pyrolysis equipment and process, and effectively controlling biomass pyrolysis. The traditional detection method is time-consuming and laborious. Near infrared spectroscopy (NIRS) as a fast and efficient detection method has great potential for rapid quantitative detection of the basic characteristics of raw materials, processes and products in biomass pyrolysis. This study selected the main crop straw corn, wheat straw, NIRS different granularity of raw material composition and calorific value of laboratory and on-line quantitative based on the characteristics of fuel and low temperature pyrolysis solid product activation on the pyrolytic characteristics and the research of rapid quantitative prediction. The results show that NIRS can quickly predict the composition and calorific value of raw materials, pyrolysis characteristics, pyrolysis activation energy and the fuel properties of low temperature pyrolytic solid products. The main innovative achievements of this paper are: 1, NIRS corn and wheat straw coarse powder material prediction fast quantitative cellulose and hemicellulose, lignin, soluble sugar, moisture, ash, volatile matter and fixed carbon, C, H, N, O, K, Mg and calorific value, and for quantitative S the amount of prediction, further research is needed to model. Compared with the fine powder spectral model, the accuracy of the fine powder model is higher than that of the coarse powder model. The spectral preprocessing way of the optimal model of coarse powder and fine powder is different, indicating that different state of the sample requires different pretreatment methods. 2, online spectral acquisition parameters have different effects on spectral repeatability of different near infrared spectrometer. Therefore, different acquisition parameters are needed for different spectrometers, so as to ensure spectral quality. Using the optimized spectrometer and acquisition parameters online, the model can be samples of cellulose, hemicellulose and lignin, soluble sugar, moisture, ash, volatile matter and fixed carbon, C, H, O, N, K, Mg, and for the rapid and quantitative prediction of calorific value, the prediction effect of S is poor. Compared with the static coarse powder model, the accuracy of the on-line model is equivalent to or slightly better than that of the static coarse powder model. 3. There is a significant difference in the characteristics of the pyrolysis process at different pyrolysis rates, and there is a significant difference between the characteristics of the pyrolysis process of the wheat straw and the corn straw. NIRS can quickly and quantitatively predict the initial temperature, the temperature of the extrapolation, the total weightlessness rate, the weightlessness rate, the peak rate and the maximum peak temperature of the pyrolysis extrapolation. The RSD of the model is less than 10%. The activation energy of corn straw was significantly higher than that of wheat straw. The activation energy increases first and then decreases with the increase of conversion rate. The activation energy corresponding to the peak conversion point has no significant change with the increase of temperature. NIRS can quickly predict the average activation energy of its pyrolysis and the activation energy of 0.3-0.6 phase, and the model RSD is less than 10%. 4. The effect of different final temperature on the fuel properties of low temperature pyrolysis solid products is very significant. Compared with the final temperature, the different heating rates and different nitrogen blowing rates have little effect on the fuel properties of the low temperature pyrolysis solid products. The fuel characteristics of the low temperature pyrolysis solid products of corn straw and the low temperature pyrolysis of wheat straw are significantly different. NIRS can be used for rapid quantitative prediction of corn straw and wheat straw pyrolysis solid product yield, energy yield, quality of calorific value, volatile, fixed carbon, ash, C, H, O, N, fuel ratio, the model of cross validation of RSD were 4.66%, 5.12%, 3.29%, 7.01%, 7.61%, 7.44%, 2.10%. 8.18%, 5.31%, 7.02%, 11.89%, the prediction accuracy of water and the content of S is low, need further study.
【学位授予单位】：中国农业大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：S216.2
，

本文编号：1339869