饲料中肉骨粉含量的近红外反射光谱检测方法

收集28份鸡饲料,31份猪饲料,25份牛饲料和肉骨粉7份,在饲料中掺入不同比例(0.5%～6.0%)的肉骨粉,制成分析样本.采用偏最小二乘(PLS)法,分别对光谱进行散射校正、平滑、一阶导数和二阶导数预处理,建立了鸡饲料、猪饲料和牛饲料中肉骨粉含量的近红外定量分析模型.利用验证集样本对定标模型进行了检验,鸡饲料、猪饲料和牛饲料中肉骨粉含量的真值与预测值之间的决定系数分别为0.9694、0.9846和0.9788;标准差分别为0.279、0.252和0.287;相对分析误差分别为5.663、6.865和5.889.结果表明,利用近红外光谱法测定饲料中的肉骨粉含量具有较高的预测精度.     

基于近红外光谱技术的灵芝提取物掺假定量快速无损检测研究

为了探索快速测定灵芝提取物掺假(淀粉)含量的方法,采用近红外光谱扫描掺杂0%、5%、10%、20%、40%、60%、80%淀粉的灵芝提取物,对其光谱进行预处理和波段选择,并结合偏最小二乘法(PLS)建立灵芝提取物掺假定量快速无损检测方法。结果表明,使用多元散射校正(MSC)预处理方法,波数范围8 000~7 500、6 000~5 500和5 000~4 000cm~(-1),主因子数为8时,建立的偏最小二乘法模型的校正决定系数(R_(cal))、校正均方根差(RMSECV)、验证决定系数(R_(val))和验证均方根差(RMSEP)分别为0.9962、0.0249、0.9960和0.0241。运用该模型对验证集样品进行预测并统计分析,可知预测值与实际掺假值之间无显著差异。本研究为灵芝功能食品市场检测提供了方法基础。     

贮备饲料近红外光谱模型快速预测青绿饲料营养成分含量

为快速检测饲料的营养成分,该研究利用贮备饲料的近红处技术(near-infrared,NIR)快速分析模型预测青绿饲料的营养成分含量。基于贮备饲料的NIR定标模型,将建模优化模式转移应用到青绿饲料的营养成分定量检测,以判断模型转移能力。在实验室环境下扫描并记录新鲜的青绿饲料样本和储存的贮备饲料样本的近红外反射光谱,利用230个贮备饲料样本进行光谱定标训练,以修正偏最小二乘(modified-partial least squares,M-PLS)建模方法,结合随机局部样本、局部选参、局部非连续性可调、交叉检验等技术相结合的方式建立局部优化模型,分别测试120个贮备饲料样本和120个青绿饲料样本中的氮(nitrogen,N)、中性洗涤纤维(neutral detergent fiber,NDF)、酸性洗涤纤维(acid detergent fiber,ADF)含量。将贮备饲料的定标校正模型应用于贮备饲料验证样本的营养成分测定,其标准误差(square error of prediction,SEP):N为1.02、NDF为16.56和ADF为13.47,相关系数均在0.9以上,相对预测偏差(relative prediction derivation,RPD)均大于3;该模型具有对青绿饲料样本的营养成分预测能力,其预测SEP:N为0.90、NDF为14.11和ADF为9.98,预测相关系数均在0.9以上,预测RPD均大于3,达到快速检测误差标准。由于局部建模过程中考虑了数据的潜在非线性结构和具有近似光谱响应的样本之间的不均匀性,相对全局建模方式而言具有更好的数据驱动性质,其建模效果优于全局建模方法。结果表明,基于贮备饲料样本建立的NIR定标校正模型可以用于青绿饲料营养成分的预测,特别是局部分析模型的应用能够提高NIR快速分析的预测精度。     

烤烟烟叶钾含量的近红外光谱法快速测定

随机选取烤烟建模集样品(150个)和检验集样品(35个),利用傅里叶变换近红外光谱仪测定烤烟样品的近红外光谱,并用常规化学分析法测定烤烟样品的含钾量。采用偏最小二乘法(PLS)把测得的烤烟样品的光谱值与烤烟钾含量的数值拟合建立定标模型,经分析得出:预测模型分析烤烟钾含量的决定系数(R2)为0.909,预测标准差(RMSEP)为0.119%。近红外法测定结果与常规化学分析方法的结果具有较好的相关性,能够应用于烤烟钾含量的快速诊断。     

基于可见/近红外反射光谱的梨树叶片钾含量的快速测定研究

利用可见/近红外反射光谱定量分析技术对梨树鲜叶钾素含量进行快速测定研究。对150个梨树叶片样本进行光谱扫描,其中120个做建模集,30个做验证集。通过对样品的可见/近红外光谱进行多种预处理,并建立钾素预测模型,探讨了可见/近红外光谱数据预处理对预测精度的影响。结果表明,通过原始光谱与S-G(3)平滑相结合的预处理方法,用17个主成分建立的偏最小二乘法模型最好,其交叉验证集和预测集模型的决定系数(R2)分别为0.722 7和0.679 1,交叉验证均方根误差(RMSECV)为1.171,预测的平均相对误差为6.81%,能高效、快速地预测梨树叶片钾素含量,为梨树钾素快速测定提供了新的手段。     

近红外光谱法预测羊肉化学成分的研究

对3个品种、3个部位的106个羊肉样品进行近红外光谱扫描,并测定其蛋白质、水分、脂肪含量,采用Unscrambler软件建立基于偏最小二乘法的近红外光谱预测模型。结果显示:样品水分含量近红外光谱校正决定系数为0.94,验证决定系数是0.86;蛋白质含量近红外光谱预测模型的校正决定系数为0.90,验证决定系数为0.72;脂肪含量近红外光谱校正决定系数0.81,验证决定系数0.64,由此可知近红外光谱用于羊肉品质检测具有可行性。本研究为羊肉化学成分的快速检测提供了基础。     

苹果质地品质近红外无损检测和指纹分析

为了探索近红外光谱快速无损检测苹果质地品质的方法,采集240个苹果样本的近红外光谱( 波长 8002500 nm),通过解析光谱图和进行不同的预处理,利用偏最小二乘法(PLS)和多元线性回归(MLR)建立回归模型和确定特征指纹图谱.基于波长范围为1300～2500 nm,PLS结合多元散射校正(MSC)所建模型的预测效果最好,硬度模型的预测标准偏差(RMSEP)和决定系数(R2)分别为0.226 kg/cm2、96.52%,脆度模型的 RMSEP和R2分别为0.243 kg/cm2、97.15%.用权重法基于PLS模型选择的硬度特征波长为1657、1725、1790、2455、1929、2304 nm,脆度特征波长为1613、1725、1895、2304、2058、2087、2396 nm,经MLR模型检验,特征波长与苹果的硬度和脆度有很高的相关性,硬度的RMSEP和R2分别为0.271 kg/cm2、90.30%,脆度的RMSEP和R2分别为0.304kg/cn2、91.64%.结果表明,PLS模型和特征指纹光谱均能准确预测苹果的质地品质,为苹果质地品质的评价提供了快速、直观、简便、可行的新方法.     

傅立叶变换近红外光谱法检测白酒总酸和总酯

该文研究白酒总酸和总酯的快速检测技术，通过解析不同白酒样品的近红外光谱图，对光谱数据进行不同的处理，结果表明：用一阶导数预处理光谱，谱区选择6102～5446 cm^-1，利用人工神经网络与傅立叶变换近红外光谱相结合,采用内部交叉验证法建立模型，效果较好。其中，总酸模型的决定系数为96.73%，内部交叉验证均方根差为0.048 g/L；总酯模型的决定系数为99.58%，内部交叉验证均方根差为0.085 g/L；进一步对总酸和总酯的模型进行验证和评价，结果表明总酸模型验证集的相关系数为99.2%,预测标准偏差为0.074 g/L；总酯模型验证集的相关为99.7%,预测标准偏差为0.134 g/L，表明建立的模型可靠，预测效果好,能满足白酒生产中总酸和总酯的快速检测要求。     

基于近红外光谱的土壤全氮含量估算模型

土壤全氮是诊断土壤肥力水平和指导作物精确施肥所需的重要信息,建立土壤全氮的近红外光谱估测模型并对建模波段进行优化选择对于土壤养分信息快速获取和精确农业发展具有重要意义。该研究以中国中、东部地区5种主要类型土壤为研究对象,利用近红外光谱仪采集土壤样品的光谱信息,结合近红外区域分子振动特点选取全谱、合频、一倍频、二倍频和N-H基团及其组合的8个波段,采用多元散射校正等多种预处理方法组合进行处理,结合偏最小二乘法(PLS)对每个波谱区域进行定标建模。结果表明,利用4000～5500cm-1波谱区域结合附加散射校正处理过的原始光谱建立的模型精度表现最好,其内部互验证决定系数达到0.90,均方根误差为0.16。经不同类型土壤的观测资料检验,模型验证决定系数为0.91,均方根误差为0.15,相对分析误差RPD为3.40,表明模型具有极好的预测能力。因此,利用近红外光谱可以实现土壤全氮的快速估测,且以合频波段(4000～5500cm-1)为建模区域可以得到更好的预测效果。     

利用近红外光谱技术检测掺假豆浆

为了对豆乳内在营养指标及掺假豆乳进行快速检测,试验运用近红外光谱技术,利用偏最小二乘法进行回归分析,分别建立83个真伪豆浆样品的蛋白质和总固形物含量定标模型,并对模型的预测性能进行分析。结果表明:选取主成分数为12和14,蛋白质和总固形物含量的近红外光谱预测值与化学实测值之间的相关系数R分别为0.9756和0.9489,校正均方根误差分别为0.186和0.175,预测集样品的预测值和实测值之间的残差值均较小、接近零,残差之和分别为-0.074和-1.191,说明建立的定标模型可以准确预测豆浆中蛋白质和总固形物含量,且预测性能较好;通过对预测集样品的预测值与豆浆行业标准规定值相比较,确定预测集样品中掺假豆浆的正确判别率为100%,说明建立的蛋白质和总固形物含量定标模型可以应用于掺假豆浆的判别检测,且判别结果准确率高。本试验表明利用近红外光谱技术可实现对豆浆主要品质指标的快速无损检测,也可准确进行真伪豆浆的快速判别,本检测方法可为豆乳行业健康持续发展提供一定的技术支撑。     

Potential of Near- and Mid-infrared Spectroscopy in Biofuel Production

For many decades, near-infrared spectroscopy (NIRS) has been used to determine the composition of animal feedstuffs and grains. More recently, mid-infrared spectroscopy (mid-IR) has also been examined for similar determinations. These spectroscopic methods offer the potential for rapid and accurate determination of organic constituents, such as fiber components and protein, of forages, by-products, and grains at reduced cost and greatly increased speed (minutes instead of hours or days). Because they are nonchemical in nature, they result in a large reduction (90% or more) in the chemical wastes associated with standard chemical-based assay methods. The same components of interest for biofuel production (cellulose, hemicellulose, lignin, starch, protein, oil, etc.) are those that have already been determined by NIRS/mid-IR for evaluating grains and animal feedstuffs. Therefore, these techniques would appear to be a natural match for evaluating feedstocks for biofuels, and the literature shows that efforts in this direction are being successfully tested and instituted. For this discussion, an overview of where such efforts are and the potential for NIRS/mid-IR in producing biofuels will be covered. For example, while there are similarities between the needs of the biofuels industry and the analysis of animal feedstuffs, there are also both practical and technical differences between the two that will likely impact how NIRS/mid-IR is developed for biofuels. As an example, grain analysis for protein is performed on a large scale by government agencies such as the Canadian Grain Commission and U.S. Grain Inspection Service, while at least in the United States, animal feedstuff analysis is performed by state or independent laboratories for individual farmers. For biofuels, this might well result in most analysis being performed by the large corporations converting the feedstocks to biofuels, as opposed to the individual producer having analysis performed at an independent laboratory. Similarly for animal feedstuffs, measurements of fiber (neutral detergent fiber or NDF, acid detergent fiber or ADF, and lignin) and protein are carried out. These fiber measurements often consist of more than one type of fiber component with some being computed by difference (hemicellulose = NDF – ADF) and are empirical at best. Whether such empirical estimates will be sufficient for assessing biofuels or whether new spectroscopic methods for directly measuring the components of interest (cellulose, etc.) will need to be developed is a question to be answered when components other than starch for ethanol or oil for biodiesel become common.     

不同淀粉糊化及凝胶特性与粉条品质的关系

为了研究粉条加工过程中原料淀粉的糊化及凝胶特性对粉条品质的影响,该文对绿豆、红薯、马铃薯、大米和玉米等5种原料淀粉的糊化凝胶特性及其粉条品质进行了测定,并对淀粉糊化凝胶特性与淀粉粉条品质之间的关系进行探讨。结果表明:5种淀粉原料所制的粉条中,绿豆粉条的品质是较好,其次就是马铃薯粉条和红薯粉条,大米粉条和玉米粉条的品质较差;淀粉的糊化特性与粉条品质之间具有显著相关性,按显著程度的大小(P值大小)依次是:峰值黏度谷值黏度衰减值回生值、最终黏度;淀粉凝胶的硬度、弹性、黏性和咀嚼性对粉条品质的影响较大,按显著程度的大小(P值大小)依次是:硬度黏性咀嚼性弹性。在粉条加工原料选择及粉条品质改善中可以考虑用谷值黏度、回生值以及淀粉凝胶特性特征值回复性、咀嚼性和黏性作为考核衡量指标。研究结果为粉条生产中原料选择及品质改善提供参考依据。     

Effects of Phosphate Salts on the pH Values and Rapid Visco Analyser (RVA) Pasting Parameters of Wheat Flour Suspensions

Changes in pH and pasting properties of instant‐noodle formula dry‐mix suspensions containing each of 12 phosphate salts were investigated. The pH values of solutions alone and then solution and flour suspensions decreased as the level of phosphate salts increased, except that of trisodium phosphate, which increased the pH value. The changes in the Rapid Visco Analyser (RVA) pasting parameters of instant‐noodle formula suspensions were not consistent with the respective changes in pH, but the change trends of the RVA parameters for the two different wheat flours (hard red winter and soft white wheat) were similar. Five of the phosphate salts gradually increased the RVA peak viscosity (PV) as phosphate concentration increased. Seven other phosphate salts increased the PV at 0.05% and then decreased PV as the phosphate concentration increased. The change in trough viscosity owing to phosphate salt and concentration was similar to that of PV. The final viscosity (FV) gradually declined, to varying degrees, as the phosphate concentration increased for some of the phosphate salts. However, seven of the phosphate salts caused slight increases in FV as their concentrations increased. The response of starch gelatinization and pasting behavior, as measured by RVA, indicated that phosphate salts exert an influence on starch during heating in water. Because RVA parameters have been linked to instant‐noodle processing and textural properties, phosphate salt identity and concentration can likely be manipulated to affect end‐product quality.     

Molecular and Gelatinization Properties of Rice Starches from IR24 and Sinandomeng Cultivars

The differences in pasting properties involving gelatinization and retrogradation of rice starches from IR24 and Sinandomeng cultivars during heating‐cooling processes were investigated using a Rapid Visco Analyser (RVA)and a dynamic rheometer. The results were discussed in relation to the molecular structure, actual amylose content (AC), and concentration of the starches. Generally, both starches possessed a comparable AC (≈11 wt%), amylose average chain length (CL), iodine absorption properties, and dynamic rheological parameters on heating to 95°C at 10 wt% and on cooling to 10°C at higher concentrations. In contrast to Sinandomeng, IR24 amylose had a greater proportion of high molecular weight species and number‐average degree of polymerization (DP_n). IR24 amylopectin possessed a lower DP_n and greater CL, exterior CL (ECL), and interior CL (ICL). Comparing the results of RVA analysis and dynamic rheology, the gelatinization properties and higher retrogradation tendencies of IR24 starch can be related to the structural properties and depend on starch concentration. In addition, the exponent n of starch concentration for storage moduli at 25°C (G′₂₅ ∝ Cⁿ) increased linearly with increasing AC.     

Potential Sugar Reduction in Cookies Formulated with Sucrose Alternatives

Sugar reduction in low‐moisture cookies is a challenge for the baking industry, because detrimental gluten development and starch gelatinization/pasting increase as sugar concentration decreases. In this study, sucrose and two healthful carbohydrate oligomers (i.e., isomaltulose and Mylose 351 syrup/Glucodry 314 powder) were used to explore the effects of the sucrose alternatives on results from solvent retention capacity (SRC), differential scanning calorimetry (DSC), rapid viscoanalysis (RVA), and AACC International wire‐cut cookie baking. Wheat flour SRC results indicated lower swelling of solvent‐accessible arabinoxylans in isomaltulose solution, but higher swelling in Glucodry 314 and Mylose 351 solutions, compared with that in sucrose solution. DSC and RVA results showed retardation of wheat flour starch gelatinization and pasting onset, respectively, both in the order water < Glucodry 314 ≤ Mylose 351 < sucrose ≤ isomaltulose. Isomaltulose, when predissolved, exhibited cookie‐baking responses similar to those for sucrose, suggesting that this sugar could be used successfully as a sucrose alternative to produce wire‐cut cookies with lower glycemic impact. Cookie baking with blends of sucrose and Glucodry 314 alleviated a cookie‐geometry issue observed for 100% sucrose replacement by Glucodry 314, suggesting that Glucodry 314 or Mylose 351 could partially replace sucrose for sugar‐reduction purposes.     

Sources of Variation for Starch Gelatinization,Pasting, and Gelation Properties in Wheat

The starch of wheat (Triticum aestivum L.) flour affects food product quality due to the temperature-dependent interactions of starch with water during gelatinization, pasting, and gelation. The objective of this study was to determine the fundamental basis of variation in gelatinization, pasting, and gelation of prime starch derived from seven different wheat cultivars: Kanto 107, which is a partial waxy mutant line, and six near-isogenic lines (NILs) differing in hardness. Complete pasting curves with extended 16-min hold at 93°C were obtained using the Rapid Visco Analyser (RVA). Apparent amylose content ranged from 17.5 to 23.5%; total amylose content ranged from 22.8 to 28.2%. Starches exhibited significant variation in onset of gelatinization. However, none of the parameters measured consistently correlated with onset or other RVA curve parameters that preceded peak paste viscosity. Peak paste viscosity varied from 190 to 323 RVA units (RVU). Higher peak, greater breakdown, lower final viscosity, negative setback, and less total setback were associated with lower apparent and total amylose contents. Each 1% reduction in apparent or total amylose content corresponded to an increase in peak viscosity of about 22 and 25 RVU, respectively, at 12% starch concentration. Of the seven U.S. cultivars, the lower amylose cultivars Penawawa and Klasic were missing the granule-bound starch synthase (GBSS; ADPglucose starch glycosyl transferase, EC 2.4.4.21) protein associated with the Waxy gene locus on chromosome 4A (Wx-B1 locus). Kanto 107 was confirmed as missing both the 7A and 4A waxy proteins (Wx-A1 and Wx-B1 loci). The hardness NIL also were shown to be null at the 4A locus. Apparent and total amylose contents of prime starch generally corresponded well to the number of GBSS proteins; although the hardness NIL tended to have somewhat higher amylose contents than did the other GBSS 4A nulls. We concluded that reduced quantity of starch amylose due to decreased GBSS profoundly affects starch gelatinization, pasting, and gelation properties.     

Properties of Starches from Near‐Isogenic Wheat Lines with Different Wx Protein Deficiencies

To determine the effect of amylose content on the starch properties, the amylose content, pasting properties, swelling power, enzymatic digestibility, and thermal properties of partial and perfect waxy types along with their wild‐type parent were analyzed. As expected, amylose content decreases differently in response to the loss of each Wx gene, showing the least response to Wx‐A1a. Most of the characteristics, except the thermal properties of the amylose‐lipid complex in differential scanning calorimetry (DSC), differed significantly among the tested types. Furthermore, the breakdown, setback, and pasting temperatures from the Rapid Visco Analyser (RVA) and the enzymatic digestibility, swelling power, peak temperature, and enthalpy of starch gelatinization from DSC showed a correlation with the amylose content. The relationships between the peak viscosity from the RVA and the onset temperature of starch gelatinization determined by DSC with amylose content of the tested materials were not clear. Waxy starch, which has no amylose, showed a contrasting behavior in starch gelatinization compared with nonwaxy starches. Among the nonwaxy starches, lower setback, lower pasting temperature, higher enzyme digestibility, higher peak temperature, higher enthalpy of starch gelatinization, and higher swelling were generally associated with low amylose starches.     

Determination of Rapid Visco Analyser Parameters in Rice by Near‐Infrared Spectroscopy

The objective of these studies was to find alternative Rapid Visco Analyser (RVA) viscoelastic parameters that are predictable by near‐infrared spectroscopy (NIRS). Currently, RVA instruments are widely used in assessing cooking and processing characteristics in rice. The ability to predict RVA parameters by NIRS would be useful in rapidly determining rice pasting qualities, but NIRS does not correlate with the traditional parameters (peak viscosity, final viscosity, breakdown, consistency, and setback). Alternative RVA parameters were sought by collecting RVA and NIRS data for a total of 86 short, medium, and long grain rice cultivars. The amylose contents were 0.41–24.90% (w/w) and protein concentrations were 8.47–11.35% (w/w). Partial least squares (PLS) regression models generated for the entire NIR spectrum against the RVA curve showed viscosity at 212–228 sec (80°C ± 1) varied linearly with NIR spectra (1,100 to ‐2,500 nm). Regression coefficient values were R = 0.961 for 212 sec and R = 0.903 for 228 sec. The PLS correlation coefficient for the prediction of amylose at 212–228 sec decreases along with the NIRS correlation to the same time frame. An opposite trend was observed for the correlation with protein at 212–228 sec. This comparison suggests the importance of amylose and protein in water absorption during this time frame.     

Heat Treatments of Milled Rice and Properties of the Flours

The effects of autoclave and oven treatments on the gelatinization of rice flour and on the rheological characteristics of its pastes were studied by differential scanning calorimetry (DSC), rapid viscoanalysis (RVA), and rotational viscometry. Flours from autoclave‐treated rice (ATR) and oven‐treated rice (OTR) were prepared, respectively, by heating at 120°C for 60 min and 160°C for 60 min followed by drying (ATR sample), and grinding at 2.2–12.9% moisture content. The rice flour dispersions were adjusted between pH 6.3 and 2.8 using 0.2M citrate buffer. The retort processing of rice flour in water pastes were done at 120°C for 20 min either once or twice. The gelatinization peak temperature (PT and T_o) and the peak temperature corresponding to the amylose‐lipid complexes (T_p3) of ATR increased at pH 6.3 and 2.8 compared with OTR and UTR flour. This indicates that the internal structures of the starch granules in ATR became more stable to heat and acid, even though the damaged starch content of ATR was 23% compared with 16 and 7%, respectively, for untreated rice flour (UTR) and OTR. The OTR flour pastes showed a gel‐like behavior at pH 4.5 after retort processing in water at 120°C for 20 min; however, the ATR mixture behaved more like a liquid paste. Decreases in the reducing sugar content of OTR and ATR pastes suggested that enzymes in the heat‐treated rice were denatured, which retarded the hydrolysis of glucose chains and the rupture of starch granules during pasting.     

Evaluation of Liquid Nitrogen Freeze Drying and Ethanol Dehydration as Methods to Preserve Partially Cooked Starch and Masa Systems

Preservation of starch structure/properties, including structures formed during partial or complete cooking, are important when the impact of processing conditions is being studied. Two preservation techniques used to study changes in starch during thermal‐mechanical processing are commonly cited in the literature: 1) rapid freezing followed by lyophilization, and 2) a dehydration procedure using alcohols. A comparative determination on how these methods affect various starch structures has not been widely reported. Corn starch samples were collected from the Rapid Visco‐Analyser (RVA) at 3 min (swollen granules, 30°C), at the top of the pasting peak (gelatinized granules, 95°C), at the bottom of the trough (dispersed polymers, 95°C), and a completed RVA sample stored for 120 hr at 4°C (retrograded starch). Samples of masa were obtained by nixtamalizing corn. Differential scanning calorimetry (DSC) endotherms of starch and masa, and X‐ray diffraction (XRD) patterns of masa were evaluated after being preserved by alcohol‐ or freeze‐drying. No significant differences (P > 0.05) between methods were found for onset, end, and peak temperatures (°C), enthalpy (J/g) and % relative crystallinity in any of the samples analyzed. Liquid nitrogen freeze‐drying and ethanol dehydration are both effective methods of preserving various starch systems for structural changes detectible by DSC and XRD; freeze‐drying is generally less expensive and time‐consuming.