基于LED的便携式近红外整粒小麦成分分析仪的研制
摘要
小麦是最主要的粮食作物之一，快速无损测定小麦营养成分，在粮食的收购及生产加工中具有十分重要的意义。近红外仪器检测方法在谷物检测领域得到了广泛的认可。本课题旨在利用LED体积小、功耗低以及滤光片带宽相对较窄的优点，研制LED与滤光片结合型的便携式近红外光谱仪器，为小麦成分的现场检测提供了一个经济、快捷的便携式设备。
本仪器采用长光程透射法。以880、950nm的近红外LED为光源，14个窄带干涉滤光片为单色器，以硅光电池为检测器，经AD转换得到数字信号，送单片机处理。经多次重复装样以减少样品颗粒度大带来的影响。通过RS232串口，将建模样品的光谱上传到PC机。在PC机上采用逐步回归建模，得到营养成分浓度与光谱的关系，将模型的系数传同仪器，即可预测未知样品营养成分的浓度。
本文详细介绍了仪器的硬件设计、软件设计、抗干扰和低功耗措施，以及对小麦蛋白质的建模和预测应用实例。对实验结果分析，校准集中预测值与化学值相关系数R=0.84，变异系数为5.45％，检验集中预测值与化学值的相关系数R=0.93，变异系数为5.54％，表明本仪器可以满足实际应用的需要。
与其它近红外仪器相比，本课题研制的仪器具有体积小、结构简单、低功耗、抗震动等特点。该测量仪可在粮库甚至田间现场对整粒小麦成分进行无损检测。
Wheat is one of the most important grain plants. Quickly and non-damage to measure the nutrition component of wheat is very important in the field of wheat's purchase and production. The measures based on NIR apparatus have been widely certificated in the field of grain measurement. Making use of advantages of LED's small dimension, low power waste and the relatively narrow bandwidth of interference filter, we develop a portable apparatus which made up of NIR LED and interference filter, provide a economical and quickly non-damage measurement of wheat components' concentration on fieldwork.
The apparatus adopt method of long light path transmission. The apparatus uses 880,950nm NIR LED as light source, interference filter as homochromy device, Si photronic as detector. Digital signal comes from AD transformation, processed by microprocessor system. To reduce the influence which taken by huge grain, we put the sample for many times. The spectrum used for modal set is transmitted to PC by RS232.To get the relation between the spectrum and nutrition component's concentration, we set up the modal used method of stepwise regression in PC, transmit the coefficient of modal to apparatus, then we can predict the nutrition component's concentration of the wheat sample.
This paper introduced the design of hardware, software, anti-disturb, low power waste and the model establish & predict the concentration of wheat's protein detailed. The following is the result: for calibration set, correlation coefficient R = 0.84, coefficient of variation is 5.45%; for prediction set, correlation coefficient R = 0.93, coefficient of variation is 5.54%. The analysis of experiment's result shows that this apparatus can fulfill the application of practice using.
Compared with other instrument, the apparatus has many advantages, such as compact space, simple structure, low power waste, and anti-shake. The apparatus can realize non-damage measurement of wheat components' concentration on fieldwork.
本仪器采用长光程透射法。以880、950nm的近红外LED为光源，14个窄带干涉滤光片为单色器，以硅光电池为检测器，经AD转换得到数字信号，送单片机处理。经多次重复装样以减少样品颗粒度大带来的影响。通过RS232串口，将建模样品的光谱上传到PC机。在PC机上采用逐步回归建模，得到营养成分浓度与光谱的关系，将模型的系数传同仪器，即可预测未知样品营养成分的浓度。
本文详细介绍了仪器的硬件设计、软件设计、抗干扰和低功耗措施，以及对小麦蛋白质的建模和预测应用实例。对实验结果分析，校准集中预测值与化学值相关系数R=0.84，变异系数为5.45％，检验集中预测值与化学值的相关系数R=0.93，变异系数为5.54％，表明本仪器可以满足实际应用的需要。
与其它近红外仪器相比，本课题研制的仪器具有体积小、结构简单、低功耗、抗震动等特点。该测量仪可在粮库甚至田间现场对整粒小麦成分进行无损检测。
Wheat is one of the most important grain plants. Quickly and non-damage to measure the nutrition component of wheat is very important in the field of wheat's purchase and production. The measures based on NIR apparatus have been widely certificated in the field of grain measurement. Making use of advantages of LED's small dimension, low power waste and the relatively narrow bandwidth of interference filter, we develop a portable apparatus which made up of NIR LED and interference filter, provide a economical and quickly non-damage measurement of wheat components' concentration on fieldwork.
The apparatus adopt method of long light path transmission. The apparatus uses 880,950nm NIR LED as light source, interference filter as homochromy device, Si photronic as detector. Digital signal comes from AD transformation, processed by microprocessor system. To reduce the influence which taken by huge grain, we put the sample for many times. The spectrum used for modal set is transmitted to PC by RS232.To get the relation between the spectrum and nutrition component's concentration, we set up the modal used method of stepwise regression in PC, transmit the coefficient of modal to apparatus, then we can predict the nutrition component's concentration of the wheat sample.
This paper introduced the design of hardware, software, anti-disturb, low power waste and the model establish & predict the concentration of wheat's protein detailed. The following is the result: for calibration set, correlation coefficient R = 0.84, coefficient of variation is 5.45%; for prediction set, correlation coefficient R = 0.93, coefficient of variation is 5.54%. The analysis of experiment's result shows that this apparatus can fulfill the application of practice using.
Compared with other instrument, the apparatus has many advantages, such as compact space, simple structure, low power waste, and anti-shake. The apparatus can realize non-damage measurement of wheat components' concentration on fieldwork.
引文
[1] Norris K H, Barnes R F, Moore J E, et al. Prediction forage quality by NIRS. J. Animal Sci, 1976, 43(4): 899～897
[2] Law D P. Determination moisture content in wheat by NIR diffuse reflectance spectrophotometry. Cereal chem, 1977, 54(4): 874～881
[3] Williams P C, Norris K H, Sobering D C. Determination of protein and moisture in wheat and barley by NIR transmission. J. Agric Food Chem, 1985, 33: 239～244
[4] Hartwig R A, Hurburgh C R. Near Infrared Refletance Measurement of Moisture. Protein and Oil content of Ground Crambe Seed, JAOCS, 1990, 67(7):435～437
[5] 彭玉魁，李菊英，祁振秀．近红外光谱分析技术在小麦营养成分鉴定上的应用．麦类作物，1997，17(2)
[6] 吴建国，石春海，张海珍，等．应用近红外光谱法整粒测定小样品油菜籽含油量的研究．作物学报，2002，28(3)：421～425
[7] 金翀，耿燕茹，江明能，等．使用IM9100型近红外整粒谷物分析仪分析籼米直链淀粉含量．中国稻米2002，1：32
[8] 何东健，前川孝昭，森岛博．水果内部品质在线近红外分光检测装置及试验．农业工程学报，2001，17(1)：146～148
[9] 张广军，Suranjan Panigrahi．近红外透射式谷物蛋白质含量在线监测系统．光电工程，2001，28(2)：19～22
[10] 王文真，付翠真．利用近红外反射光谱快速测定大豆籽粒蛋白质、脂肪和部分氨基酸含量．作物品种资源．1994，1：31～32
[11] 石井良市，在线使用的近红外成分分析计，现代计量测试，1994，3：59～60
[12] 徐广通，袁洪福，陆婉珍．近红外光谱仪器概况与进展．现代科学仪器，1997，3：9～11
[13] Jouko Malinen, Markku Kansakoski, Raimo Rikola, et al. LED-based NIR spectrometer module for hand-held and process analyser applications. Sensor and Actuators B: Chemical Volume, 1998, 51: 220～226
[14] http://www.zeltex.com
[15] 严衍禄，高文淑，张录达，等．付里叶变换近红外漫反射光谱分析测定谷物粗蛋白质的规范化方法．北京农业大学学报，1990，16(增)：57～60
[16] Bochereau L, Bourgine P, Palagos B. A method for prediction by combining data analysis and neural networks: application to prediction of apple quality using near infra-red spectra. Journal of Agricultural Engineering Research. 1992, 51: 207～216
[17] Jasper W J, Kovacs E T. Using neural networks and NIR spectrophotometry to identify fibers. Textile Research Journal. 1994, 64(8): 444～448
[18] 陆婉珍，袁洪福，徐广通，等．现代近红外光谱分析技术．北京：中国石化出版社，2000
[19] 裴鑫德．多元统计分析及其应用．北京：北京农业大学出版社，1991，473～478
[20] 张伦，冯新强，吴常津。传感器与信号调理器件应用技术．北京：科学出版社，2002，123～126
[21] 扬振江，孙占彪，王曙梅，等．智能仪器与数据采集系统中的新器件及应用．西安：西安电子科技大学出版社，2001
[22] 北京光电器件厂．光电器件产品手册．1987
[23] 方佩敏，张国华．最新集成电路应用指南．北京：电子工业出版社．1996，343～344
[24] 扈啸，周旭升．单片机数据通信技术从入门到精通．西安：西安电子科技大学出版社，2002，233～236
[25] 廖焕霖．Protel99电路板设计者必读．北京：冶金工业出版社，2000
[26] 曾峰，侯亚宁，曾凡雨．印刷电路板(PCB)设计与制作．北京：电子工业出版社，2002，62～69
[27] 纪宗南．单片机外围器件实用手册输入通道器件分册．北京：北京航空航天大学出版社，1998，400～403
[28] 徐爱钧，彭秀华．单片机高级语言C51应用程序设计．北京：电子工业出版社，1998
[29] 赖麒文．8051单片机C语言彻底应用．北京：科学出版社，2002
[30] 林德杰，林均淳，许锦标，等．电气测试技术．北京：机械工业出版社，1999
[31] 林月芳．粮库温湿度测量与防火防盗监测系统：[硕士学位论文]．北京：中国农业大学，2003
[32] 王福瑞．单片微机测控系统设计大全．北京：北京航空航天大学出版社，1999
[33] 马忠梅，籍顺心，张凯，等．单片机的C语言应用程序设计．北京：北京航空航天大学出版社，1999
[34] 卢胜利，胡新宇，程森林．智能仪器设计与实现．重庆：重庆大学出版社，2003
[35] 王幸之，王雷，翟成，等．单片机应用系统抗干扰技术．北京：北京航空航天大学出版社，2000
[2] Law D P. Determination moisture content in wheat by NIR diffuse reflectance spectrophotometry. Cereal chem, 1977, 54(4): 874～881
[3] Williams P C, Norris K H, Sobering D C. Determination of protein and moisture in wheat and barley by NIR transmission. J. Agric Food Chem, 1985, 33: 239～244
[4] Hartwig R A, Hurburgh C R. Near Infrared Refletance Measurement of Moisture. Protein and Oil content of Ground Crambe Seed, JAOCS, 1990, 67(7):435～437
[5] 彭玉魁，李菊英，祁振秀．近红外光谱分析技术在小麦营养成分鉴定上的应用．麦类作物，1997，17(2)
[6] 吴建国，石春海，张海珍，等．应用近红外光谱法整粒测定小样品油菜籽含油量的研究．作物学报，2002，28(3)：421～425
[7] 金翀，耿燕茹，江明能，等．使用IM9100型近红外整粒谷物分析仪分析籼米直链淀粉含量．中国稻米2002，1：32
[8] 何东健，前川孝昭，森岛博．水果内部品质在线近红外分光检测装置及试验．农业工程学报，2001，17(1)：146～148
[9] 张广军，Suranjan Panigrahi．近红外透射式谷物蛋白质含量在线监测系统．光电工程，2001，28(2)：19～22
[10] 王文真，付翠真．利用近红外反射光谱快速测定大豆籽粒蛋白质、脂肪和部分氨基酸含量．作物品种资源．1994，1：31～32
[11] 石井良市，在线使用的近红外成分分析计，现代计量测试，1994，3：59～60
[12] 徐广通，袁洪福，陆婉珍．近红外光谱仪器概况与进展．现代科学仪器，1997，3：9～11
[13] Jouko Malinen, Markku Kansakoski, Raimo Rikola, et al. LED-based NIR spectrometer module for hand-held and process analyser applications. Sensor and Actuators B: Chemical Volume, 1998, 51: 220～226
[14] http://www.zeltex.com
[15] 严衍禄，高文淑，张录达，等．付里叶变换近红外漫反射光谱分析测定谷物粗蛋白质的规范化方法．北京农业大学学报，1990，16(增)：57～60
[16] Bochereau L, Bourgine P, Palagos B. A method for prediction by combining data analysis and neural networks: application to prediction of apple quality using near infra-red spectra. Journal of Agricultural Engineering Research. 1992, 51: 207～216
[17] Jasper W J, Kovacs E T. Using neural networks and NIR spectrophotometry to identify fibers. Textile Research Journal. 1994, 64(8): 444～448
[18] 陆婉珍，袁洪福，徐广通，等．现代近红外光谱分析技术．北京：中国石化出版社，2000
[19] 裴鑫德．多元统计分析及其应用．北京：北京农业大学出版社，1991，473～478
[20] 张伦，冯新强，吴常津。传感器与信号调理器件应用技术．北京：科学出版社，2002，123～126
[21] 扬振江，孙占彪，王曙梅，等．智能仪器与数据采集系统中的新器件及应用．西安：西安电子科技大学出版社，2001
[22] 北京光电器件厂．光电器件产品手册．1987
[23] 方佩敏，张国华．最新集成电路应用指南．北京：电子工业出版社．1996，343～344
[24] 扈啸，周旭升．单片机数据通信技术从入门到精通．西安：西安电子科技大学出版社，2002，233～236
[25] 廖焕霖．Protel99电路板设计者必读．北京：冶金工业出版社，2000
[26] 曾峰，侯亚宁，曾凡雨．印刷电路板(PCB)设计与制作．北京：电子工业出版社，2002，62～69
[27] 纪宗南．单片机外围器件实用手册输入通道器件分册．北京：北京航空航天大学出版社，1998，400～403
[28] 徐爱钧，彭秀华．单片机高级语言C51应用程序设计．北京：电子工业出版社，1998
[29] 赖麒文．8051单片机C语言彻底应用．北京：科学出版社，2002
[30] 林德杰，林均淳，许锦标，等．电气测试技术．北京：机械工业出版社，1999
[31] 林月芳．粮库温湿度测量与防火防盗监测系统：[硕士学位论文]．北京：中国农业大学，2003
[32] 王福瑞．单片微机测控系统设计大全．北京：北京航空航天大学出版社，1999
[33] 马忠梅，籍顺心，张凯，等．单片机的C语言应用程序设计．北京：北京航空航天大学出版社，1999
[34] 卢胜利，胡新宇，程森林．智能仪器设计与实现．重庆：重庆大学出版社，2003
[35] 王幸之，王雷，翟成，等．单片机应用系统抗干扰技术．北京：北京航空航天大学出版社，2000