CN103728267A

Movatterモバイル変換

Info

Publication number: CN103728267A
Application number: CN201410021091.7A
Authority: CN
Inventors: 钱锋; 程辉; 杜文莉
Original assignee: East China University of Science and Technology
Current assignee: East China University of Science and Technology
Priority date: 2014-01-17
Filing date: 2014-01-17
Publication date: 2014-04-16

Abstract

本发明涉及一种汽油近红外分析过程中谱图的基线校正方法，该方法通过选取汽油近红外光谱中两点的波长、该两点附近点的平均光吸收度作为基准点，确立直线方程，并通过该直线方程对光谱中的其他点进行校正，本发明的有益效果在于，（1）基线校正后的谱图能够有效的消除谱图的基线漂移，有益于提高预测值的精度；（2）该方法可以非常方便的应用于汽油调和过程中汽油近红外谱图的在线校正。The invention relates to a method for correcting the baseline of a spectrogram in the near-infrared spectrum of gasoline. The method establishes a straight line equation by selecting the wavelengths of two points in the near-infrared spectrum of gasoline and the average light absorbance of points near the two points as reference points. And correct other points in the spectrum through the straight line equation, the beneficial effect of the present invention is that (1) the spectrum after baseline correction can effectively eliminate the baseline drift of the spectrum, which is beneficial to improve the accuracy of the predicted value; (2 ) This method can be very conveniently applied to the online calibration of gasoline near-infrared spectrum in the gasoline blending process.

Description

Translated fromChinese

汽油近红外分析过程中谱图的基线校正方法Baseline Correction Method of Spectrum in Gasoline Near Infrared Analysis Process

技术领域technical field

本发明涉及一种汽油近红外分析过程中谱图的基线校正方法。The invention relates to a method for correcting the baseline of spectrograms in the gasoline near-infrared analysis process.

背景技术Background technique

汽油调和是汽油出厂前的最后一道工序，关系到成品油的质量以及炼厂的效益。随着汽油质量指标的升级，传统的罐式调和方法已经不能够满足调和需求。汽油的管道调和工艺在节能减排、提高汽油质量指标上具有明显的优势。但是，汽油管道调和的优化控制过程依赖于近红外分析仪：分析仪的稳定性与精度是调和控制优化的基础。由于近红外分析仪在运行过程中容易受到基线漂移、温度、背景噪声等的影响，从而造成采集到的汽油近红外谱图发生漂移或变形影响到预测的精度；在利用近红外技术预测汽油属性的时候需要对谱图进行必要的预处理。目前常用的预处理方式有：导数、平滑、标准化、中心化等，但是对于汽油的近红外光谱的预处理均得不到理想的效果。如何在不增加算法的复杂度的前提下，对光谱进行适当的处理从而消除背景噪声以及谱图漂移的影响至关重要。Gasoline blending is the last process before gasoline leaves the factory, which is related to the quality of refined oil and the efficiency of refineries. With the upgrading of gasoline quality indicators, the traditional tank blending method can no longer meet the blending needs. The pipeline blending process of gasoline has obvious advantages in energy saving, emission reduction and improvement of gasoline quality indicators. However, the optimization control process of gasoline pipeline blending depends on the near-infrared analyzer: the stability and accuracy of the analyzer are the basis for the optimization of blending control. Since the near-infrared analyzer is easily affected by baseline drift, temperature, background noise, etc. during operation, the drift or deformation of the collected gasoline near-infrared spectrum will affect the prediction accuracy; when using near-infrared technology to predict gasoline properties Necessary preprocessing of the spectra is required. At present, the commonly used preprocessing methods include: derivative, smoothing, standardization, centering, etc., but the preprocessing of the near-infrared spectrum of gasoline cannot obtain ideal results. How to properly process the spectrum without increasing the complexity of the algorithm is very important to eliminate the influence of background noise and spectral drift.

发明内容Contents of the invention

为了解决上述技术问题，本发明提出一种可以在线实时对汽油近红外光谱进行校正的方法，该方法原理简单、计算方便，并能够有效的消除背景噪声的干扰。In order to solve the above-mentioned technical problems, the present invention proposes a method for correcting the near-infrared spectrum of gasoline online in real time. The method is simple in principle, convenient in calculation, and can effectively eliminate the interference of background noise.

本发明方法提供的近红外光谱建模的校正集选择包括以下步骤：The correction set selection of the near-infrared spectral modeling provided by the method of the present invention comprises the following steps:

步骤一：采集汽油样本，获取汽油样本的近红外光谱；Step 1: collecting a gasoline sample and obtaining the near-infrared spectrum of the gasoline sample;

步骤二：选取汽油近红外光谱中两点波长x₁，x₂处作为基准点；Step 2: Select two wavelengths x₁ and x₂ in the near-infrared spectrum of gasoline as reference points;

步骤三：针对步骤二中的基准点x_i，作以下计算：在该基准点波长附近选择多个波长点处吸光度的平均值记为y_i，其中i=1或2；Step 3: For the reference point x_i in step 2, do the following calculation: select the average value of absorbance at multiple wavelength points near the reference point wavelength as y_i , where i=1 or 2;

步骤四：通过步骤三、步骤四所获得的基准点点(x₁,y₁)与(x₂,y₂)确定的直线方程lStep 4: The straight line equation l determined by the reference point (x₁ , y₁ ) and (x₂ , y₂ ) obtained in Step 3 and Step 4

步骤五：根据方程l计算近红外光谱中所有波长点x_i处的y_i；Step five: calculate y_i at all wavelength points x_i in the near-infrared spectrum according to equation 1;

步骤六：近红外光谱对应波长点x_i处的吸光度减去对应的y_i。Step 6: Subtract the corresponding y_i from the absorbance at the corresponding wavelength point x_i in the near-infrared spectrum.

较佳的，所述步骤一中的汽油样品来自于汽油管道调和过程，样品不需要经过任何预处理。Preferably, the gasoline sample in the step 1 comes from the gasoline pipeline blending process, and the sample does not need any pretreatment.

较佳的，所述步骤一中近红外光谱获取是通过在线近红外分析仪采集汽油的近红外光谱。Preferably, the acquisition of the near-infrared spectrum in the first step is to collect the near-infrared spectrum of gasoline through an online near-infrared analyzer.

较佳的，所述步骤二中的基准点取在吸光度为0-0.02的波长段。Preferably, the reference point in the second step is taken in the wavelength band with an absorbance of 0-0.02.

较佳的，所述步骤三所述基准点波长附近选择多个波长点，为5-10个波长点。Preferably, multiple wavelength points are selected near the wavelength of the reference point in the third step, 5-10 wavelength points.

当近红外分析仪在线运行的时候，通过本方法可以自动的对采集到的每一个普通点做相同的处理，处理速度快，效果明显。When the near-infrared analyzer is running online, the method can automatically perform the same processing on each common point collected, and the processing speed is fast and the effect is obvious.

本发明具有以下有益效果：（1）基线校正后的谱图能够有效地消除谱图的基线漂移，有益于提高预测值的精度；（2）该方法可以非常方便地应用于汽油调和过程中汽油近红外谱图的在线校正。The present invention has the following beneficial effects: (1) The baseline-corrected spectrogram can effectively eliminate the baseline drift of the spectrogram, which is beneficial to improving the accuracy of the predicted value; (2) The method can be very conveniently applied to the gasoline blending process On-line calibration of NIR spectra.

附图说明Description of drawings

图1-1：93#汽油的近红外光谱原始谱图；Figure 1-1: The original NIR spectrum of 93# gasoline;

图1-2：93#汽油基线校正后的近红外光谱原始谱图。Figure 1-2: The original NIR spectra of 93# gasoline after baseline correction.

具体实施方式Detailed ways

以下实施例用来说明本发明，但不用来限制本发明的范围。The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

汽油近红外分析过程中谱图的基线校正方法，包括以下步骤：The baseline correction method of spectrogram in gasoline near-infrared analysis process comprises the following steps:

步骤一：利用在线近红外分析仪采集汽油管道调和过程中调和总管处的成品93#汽油近红外样本。采集过程中汽油不需要经过任何预处理。汽油近红外的原始谱图见附图1-1。Step 1: Use an online near-infrared analyzer to collect near-infrared samples of finished 93# gasoline at the blending main pipe during the gasoline pipeline blending process. Gasoline does not require any pretreatment during collection. The original near-infrared spectrum of gasoline is shown in Figure 1-1.

步骤二：选取1090nm与1310nm作为基准点，并令x₁=1090，x₂=1310Step 2: Select 1090nm and 1310nm as reference points, and set x₁ =1090, x₂ =1310

步骤三：计算基准点附近5个点的平均值。即：计算1088、1089、1090、1091、1092nm五个波长点处吸光度的平均值记为y₁，1308、1309、1310、1311、1312nm五个波长点处吸光度的平均值记为y₂。Step 3: Calculate the average value of 5 points near the reference point. That is: calculate the average value of absorbance at five wavelength points of 1088, 1089, 1090, 1091, and 1092 nm as y₁ , and record the average value of absorbance at five wavelength points of 1308, 1309, 1310, 1311, and 1312 nm as y₂ .

步骤四：根据点(x₁,y₁)和(x₂,y₂)，求出过这两点的直线方程l。Step 4: According to the points (x₁ , y₁ ) and (x₂ , y₂ ), find the equation l of the line passing through these two points.

步骤五：求出1000-1600nm，步长为1nm，的所有601个波长点对应的yStep 5: Find the y corresponding to all 601 wavelength points of 1000-1600nm with a step size of 1nm

步骤六：原始汽油近红外光谱的每个波长点的吸光度减去对应的y，即可得到校正后的谱图。校正前后的汽油近红外光谱谱图见附图1-1、图1-2，由图可见，原始近红外谱图基线发生严重漂移，尤其是1000-1100nm，1250-1320nm,1400-1600nm波长段附近。经过本发明方法校正之后，基线漂移基本消失，谱图更为平滑，噪声信号较少。Step 6: Subtract the corresponding y from the absorbance of each wavelength point of the original gasoline near-infrared spectrum to obtain the corrected spectrum. The near-infrared spectra of gasoline before and after correction are shown in attached drawings 1-1 and 1-2. It can be seen from the figures that the baseline of the original near-infrared spectra has drifted seriously, especially in the 1000-1100nm, 1250-1320nm, and 1400-1600nm wavelength bands nearby. After being corrected by the method of the invention, the baseline drift basically disappears, the spectrogram is smoother, and the noise signal is less.

本实施例以分析汽油研究法辛烷值（RON）属性为例，但也可以用于汽油其他属性的分析，如：马达法辛烷值，抗爆指数，雷德蒸汽压，硫含量，密度，烯烃含量，芳烃含量，或/和苯含量。This example takes the analysis of gasoline research method octane number (RON) properties as an example, but it can also be used for the analysis of other properties of gasoline, such as: motor method octane number, antiknock index, Reid vapor pressure, sulfur content, density , olefin content, aromatic content, or/and benzene content.

综上所述仅为发明的较佳实施例而已，并非用来限定本发明的实施范围。即凡依本发明申请专利范围的内容所作的等效变化与修饰，都应为本发明的技术范畴。In summary, the above are only preferred embodiments of the invention, and are not intended to limit the implementation scope of the invention. That is, all equivalent changes and modifications made according to the content of the patent scope of the present invention shall be within the technical scope of the present invention.

Claims

Translated fromChinese

1.汽油近红外分析过程中谱图的基线校正方法，主要包括以下步骤：1. The baseline correction method of the spectrum in the gasoline near-infrared analysis process mainly includes the following steps:

步骤一：获取汽油样本的近红外光谱；Step 1: Obtain the near-infrared spectrum of the gasoline sample;

步骤四：通过步骤二、步骤三所获得的基准点(x₁,y₁)与(x₂,y₂)确定直线方程lStep 4: Determine the line equation l through the reference points (x₁ , y₁ ) and (x₂ , y₂ ) obtained in Step 2 and Step 3

2.如权利要求1所述的校正方法，其特征在于，所述步骤一中的汽油样本来自于汽油管道调和过程，样本不需要经过任何预处理。2. The calibration method according to claim 1, wherein the gasoline sample in the step 1 comes from a gasoline pipeline blending process, and the sample does not need any pretreatment.

3.如权利要求1所述的校正方法，其特征在于，所述步骤一中近红外光谱获取是通过在线近红外分析仪采集汽油的近红外光谱。3. The calibration method according to claim 1, wherein the acquisition of the near-infrared spectrum in the step 1 is to collect the near-infrared spectrum of gasoline by an online near-infrared analyzer.

4.如权利要求1所述的校正方法，其特征在于，所述步骤二中的基准点取在吸光度为0-0.02的波长段。4. The calibration method according to claim 1, characterized in that, the reference point in said step 2 is taken in the wavelength band with an absorbance of 0-0.02.

5.如权利要求1或4所述的校正方法，其特征在于，所述步骤三所述基准点波长附近选择多个波长点，为5-10个波长点。5. The calibration method according to claim 1 or 4, characterized in that, in the third step, a plurality of wavelength points are selected near the wavelength of the reference point, which is 5-10 wavelength points.