阅读说明：本技术 一种管输原油的混油界面检测方法 (Oil mixing interface detection method for crude oil in pipeline transportation ) 是由 陈夕松 陶思琦 胡羽聪 梅彬 于 2021-08-10 设计创作，主要内容包括：本发明公开了一种原油管输的混油界面检测方法,该方法采用近红外光谱技术及导数加和技术,实时对混油界面进行检测与跟踪。该方法在线采集管输原油的近红外光谱,对光谱数据进行矢量归一化预处理后,计算连续测定的光谱样本序列在各波数点下的吸光度标准偏差及其标准偏差的一阶导数,计算光谱特征区内各波数点下的一阶导数加和值,以分析管输原油的混油程度。该方法建模简单,即使面对密度相近的原油,也可准确判断原油混油界面的开始、中间和结束阶段,以便指导收油站及时进行原油切割收油。这对于提高原油管输调度过程的灵活性,提升管输灵活性等具有重要价值。(The invention discloses a method for detecting a mixed oil interface of crude oil pipe transportation, which adopts a near infrared spectrum technology and a derivative addition technology to detect and track the mixed oil interface in real time. The method comprises the steps of collecting the near infrared spectrum of the crude oil in the pipeline transportation on line, carrying out vector normalization pretreatment on spectral data, calculating the absorbance standard deviation of a continuously measured spectral sample sequence at each wave number point and the first derivative of the standard deviation, and calculating the summation value of the first derivative at each wave number point in a spectral characteristic region to analyze the oil mixing degree of the crude oil in the pipeline transportation. The method is simple in modeling, and can accurately judge the starting, intermediate and ending stages of the oil mixing interface of the crude oil even for the crude oil with similar density so as to guide an oil receiving station to cut and receive the crude oil in time. The method has important values for improving the flexibility of the crude oil pipeline transportation scheduling process, improving the pipeline transportation flexibility and the like.)

1. A method for detecting a mixed oil interface of crude oil in a pipeline is characterized by comprising the following steps:

1) acquiring the near infrared spectrum of the crude oil sequentially conveyed by the pipeline in real time, and preprocessing the spectrum in the characteristic region;

2) continuously reading the spectral absorbance A of p moments forward by taking the current moment k as a starting point_ijAnd calculating the standard deviation S of the absorbance of the continuous p spectra aiming at each wave number point i_ik：

Where i is 1,2, and m is the total number of spectral points in the feature region, and A_ijThe spectral absorbance at the ith wave number point in the spectral sequence acquired at the time j, j being k-p +1, k-p + 2.

3) Calculating the standard deviation S of absorbance by adopting a Savitzky-Golay convolution derivation method_ikFirst derivative S of_ik′：

In the formula, g_tAs a weighting factor, the width of the smoothing window is 2 ω +1, ω is an integer, and Δ k is the differential width;

4) calculating the addition value of the first derivative of each wave number point in the spectral characteristic region:

5) according to the sum value S_kJudging the oil mixing stage of the crude oil conveyed by the pipeline;

6) and outputting the judgment result to a crude oil pipeline transportation scheduling control system, and cutting the crude oil pipeline transportation to receive oil in time.

2. The method for detecting the mixing interface of the crude oil transported by the pipeline as claimed in claim 1, wherein the spectral wave number range in the characteristic region in the step 1) is 5200cm^-1～6200cm^-1。

3. The method for detecting the oil mixing interface of the crude oil transported by the pipeline as claimed in claim 1, wherein the absorbance in the spectral feature region is preprocessed by a vector normalization method in the step 1).

4. The method for detecting the oil mixing interface of the crude oil transported by the pipeline as claimed in claim 1, wherein p in the step 2) is calculated according to the following formula:

p＝0.025πd²Lt^-1Q^-1；

in the formula, L is the length of the oil mixing segment, d is the inner diameter of the pipeline, Q is the volume flow of the oil product in the pipeline, t is the spectrum acquisition period, and p is rounded downwards.

5. The method as claimed in claim 1, wherein the smoothing window width in step 3) is set to 5, and the first derivative S is obtained by 5-point second smoothing_ik′：

6. The method for detecting the oil mixing interface of the crude oil transported by the pipeline as claimed in claim 1, wherein the oil mixing stage is judged in step 5) according to the following steps:

5-1) initializing an oil mixing flag bit tag to be 0;

5-2) acquiring the near infrared spectrum of the crude oil sequentially conveyed by the pipeline in real time;

5-3) calculating the addition value S of the first derivative of each wave number point of the spectrum in the characteristic region at the current moment_k′；

5-4) judgment of S_k' whether or not 0.005 is or less: if yes, turning to the step 5-5); otherwise, judging the oil mixing starting stage, setting tag to be 1, and turning to the step 5-7);

5-5) judgment of S_kWhether or not it is-0.005: if yes, turning to the step 5-6); otherwise, judging that the oil mixing is finished, setting tag to be 0, and turning to the step 5-7);

5-6) judging whether tag is 0: if yes, judging the crude oil stage as a single crude oil stage; otherwise, judging the oil mixing intermediate stage;

5-7) outputting a judgment result of the oil mixing interface of the crude oil;

5-8) to step 5-2).

Technical Field

The invention relates to the field of crude oil pipeline transportation, in particular to a method for quickly detecting a mixed oil interface of pipeline transportation crude oil.

Background

In the field of crude oil pipeline transportation, a sequential transportation method is generally adopted, namely a transportation mode of continuously transporting different oil types in a pipeline according to a certain batch and sequence. The sequential conveying can improve the utilization rate of pipelines, and has the advantages of large conveying capacity, good economical efficiency and convenient automatic control.

In the process of sequentially conveying the crude oil by the pipeline, the alternate positions of the two crude oils form an oil mixing section, so that an oil mixing interface needs to be detected in real time in an oil receiving station for receiving the crude oil, so that the crude oil is cut and received accurately in time. At present, the oil mixing interface detection in industry mostly adopts a density measurement method, and the method is simple to operate, has better resolution capability on crude oil with larger density difference and is widely applied. However, this method is difficult to distinguish between different crude oils with similar densities, thereby reducing the flexibility of pipeline transportation.

Near infrared spectroscopy analysis technology has begun to be applied to the field of rapid analysis of crude oil properties in recent years, and has the advantages of high measurement speed, high precision and the like compared with manual assay. At present, an analysis technology based on near infrared spectroscopy is applied to the research of oil mixing interface detection, but the adopted method is mainly machine learning methods such as principal component statistical analysis and the like, and the problems of more complex modeling, larger calculated amount, weaker real-time performance and the like exist.

Disclosure of Invention

Aiming at the problems in the background technology, the invention utilizes the near infrared spectrum analysis and the derivative addition technology to detect and track the mixed oil interface in real time. The method comprises the following steps:

1) collecting near infrared spectrum of crude oil conveyed by pipeline in sequence in real time, selecting 5200cm^-1～6200cm^-1The spectrum in the wave number range is used as a characteristic region spectrum, and vector normalization pretreatment is carried out on the spectrum in the characteristic region;

2) continuously reading the spectral absorbance A of p moments forward by taking the current moment k as a starting point_ijAnd calculating the standard deviation S of the absorbance of the continuous p spectra aiming at each wave number point i_ik：

Where i is 1,2, and m is the total number of wave number points of the spectrum in the characteristic region, a_ijThe spectral absorption at the ith wave point in the spectral sequence collected at time j, j being k-p +1, k-p + 2. p is 0.025 pi d²Lt^-1Q^-1Wherein L is the length of the oil mixing segment, d is the inner diameter of the pipeline, Q is the volume flow of the oil in the pipeline, t is the spectrum acquisition period, and p is rounded downwards.

3) Calculating the standard deviation S of absorbance by adopting a Savitzky-Golay convolution derivation method_ikFirst derivative S of_ik′：

In the formula, g_tAs a weighting factor, the width of the smoothing window is 2 ω +1, ω is an integer, and Δ k is the differential width; the width of the smoothing window is set to 5, and a first derivative S is obtained by a 5-point secondary smoothing method_ik′：

4) Calculating the addition value of the first derivative of each wave number point in the spectral characteristic region:

5) according to the sum value S_kThe size of the oil mixing stage of the pipeline transportation crude oil is judged according to the following steps:

5-1) initializing an oil mixing flag bit tag to be 0;

5-2) acquiring the near infrared spectrum of the crude oil sequentially conveyed by the pipeline in real time;

5-3) calculating the addition value S of the first derivative of each wave number point of the spectrum in the characteristic region at the current moment_k′；

5-4) judgment of S_k' whether or not 0.005 is or less: if yes, turning to the step 5-5); otherwise, judging the oil mixing starting stage, setting tag to be 1, and turning to the step 5-7);

5-5) judgment of S_kWhether or not it is-0.005: if yes, turning to the step 5-6); otherwise, judging that the oil mixing is finished, setting tag to be 0, and turning to the step 5-7);

5-6) judging whether tag is 0: if yes, judging the crude oil stage as a single crude oil stage; otherwise, judging the oil mixing intermediate stage;

5-7) outputting a judgment result of the oil mixing interface of the crude oil;

5-8) to step 5-2).

6) And outputting the judgment result to a crude oil pipeline transportation scheduling control system, and cutting the crude oil pipeline transportation to receive oil in time.

Has the advantages that:

the invention discloses a method for detecting a mixed oil interface of pipeline crude oil, which adopts near infrared spectrum analysis and derivative addition technology to detect and track a mixed oil product in real time. The method can analyze the change condition of the characteristic spectrum of the crude oil in real time even for crude oil with similar density, can accurately judge the starting, middle and ending stages of the oil mixing interface of the crude oil, has simple modeling and high operation speed, and can guide an oil receiving station to carry out crude oil cutting and oil receiving in time. The method has important values for improving the real-time performance of the crude oil pipeline transportation scheduling process, improving the pipeline transportation flexibility and the like.

Drawings

FIG. 1 is a flow chart of a method for detecting a mixing interface of crude oil transported by a pipeline according to the present invention;

FIG. 2 is a calculation result of standard deviation of absorbance of a spectrum sample of crude oil in a pipeline according to an embodiment of the present invention;

FIG. 3 is the calculation result of the first derivative sum of the standard deviation of the absorbance of crude oil in a pipeline according to the embodiment of the present invention;

FIG. 4 is a judgment result of the oil mixing interface for batch and fractional delivery of different types of crude oil by a pipeline according to an embodiment of the present invention.

Detailed Description

The invention is further described with reference to the following figures and specific examples. The implementation effect of the method in the process of detecting the mixed oil interface on line by pipeline transportation of crude oil is described by a specific operation flow. The present embodiment is implemented on the premise of the technical solution of the present invention, but the scope of the present invention is not limited to the following examples.

The Kasi crude oil and Rongaduo crude oil are sequentially conveyed by a long pipeline, an FTPA2000-260 online spectrometer is arranged on the pipeline, a 2mm optical path flow cell is adopted, and the pipeline is sent to a near-infrared spectrometer for scanning after being kept at a constant temperature of 40 ℃. In order to accurately judge the starting, middle and ending stages of the oil-mixing interface of the crude oil, the technology of the invention is decided to be adopted for analysis and detection. The specific process is as follows:

1) selecting the wave number band of 5200cm^-1～6200cm^-1And carrying out vector normalization pretreatment on the near infrared spectrum. Wherein, the table 1 is partial data of the near infrared spectrum of the pretreated pipeline crude oil.

TABLE 1 near-IR Spectrum of crude oil transported in pipe after vector normalization

2) At each wavenumber point, the standard deviation of absorbance of the spectral sample was calculated p consecutive times.

Calculated by a process theory, namely the prior actual measurement historical data, the length of the mixed oil section of the pipeline crude oil is about 3.0km, the flow rate is about 3.0ms, the spectrum acquisition period is set to 10s, and then p is 0.025 pi d²Lt^-1Q^-1When p is 10.2, p is 10.

As shown in Table 1, 5207.42cm for the wave number point^-1Calculating the absorbance A of 10 p in total of 01-10 with the spectrum number 01 as the starting point_1jStandard deviation S₁₁：

Similarly, the wave number point is 5207.42cm^-1The standard deviation S of absorbance at the 02-bit starting point and spectrum number of 02-11₁₂The value of (A) is 0.0001; similarly, the data of the standard deviation of absorbance calculated for other wavenumber points and different time series are shown in Table 2:

TABLE 2 standard deviation of absorbance of consecutive sequence spectra

The standard deviation data of absorbance of the sample was visualized, and the results are shown in fig. 2.

3) The first derivative of the absorbance standard deviation was calculated using the Savitzky-Golay convolution derivation method.

As shown in Table 2, 5207.42cm in wavenumber^-1For example, the standard deviation S of the absorbance of the time series is 16 to 20_1kRespectively is [0.0146,0.0135,0.0113,0.0085,0.0061]Calculating the standard deviation S of absorbance by a 5-point secondary smoothing method₁₍₂₀₎First derivative S of₁₍₂₀₎′：

Similarly, some data of other calculated first derivative values are shown in table 3:

TABLE 3 Absorbance standard deviation first derivative of convolution smoothing

4) And calculating the sum of the first derivative of each wave number point in the spectral characteristic region.

Taking the time series 15 in Table 3 as an example, the first derivative S of each wavenumber point in the spectral feature region_i(15)' is [0.0001,0.0001,... ], 0.0007]Then the time series 15 is the summation S of the first derivative of each wave number point in the spectral feature region₁₅' calculated as follows:

the calculation result of the summation value of the first derivative at each wavenumber point in the spectral feature region is shown in fig. 3.

5) According to the sum value S_kThe size of the' is that the oil mixing stage of the pipeline transportation crude oil is judged according to the following sequence:

5-1) initializing an oil mixing flag bit tag to be 0;

5-2) acquiring the near infrared spectrum of the crude oil sequentially conveyed by the pipeline in real time;

5-3) calculating the addition value S of the first derivative of each wave number point in the spectrum of the characteristic region at the current moment_k′；

5-4) judgment of S_k' whether or not 0.005 is or less: if yes, turning to the step 5-5); otherwise, judging the oil mixing starting stage, setting tag to be 1, and turning to the step 5-7);

5-5) judgment of S_kWhether or not it is-0.005: if yes, turning to the step 5-6); otherwise, judging that the oil mixing is finished, setting tag to be 0, and turning to the step 5-7);

5-6) judging whether tag is 0: if yes, judging the crude oil stage as a single crude oil stage; otherwise, judging the oil mixing intermediate stage;

5-7) outputting a judgment result of the oil mixing interface of the crude oil;

5-8) to step 5-2).

The process of the derivative value from 0 to the maximum, then from the maximum to the minimum, and then from the minimum to 0 is recorded as the mixed oil period of 1 oil mixing section. Fig. 3 clearly shows the change process of 1 mixed oil cycle, and it can be seen that the method can accurately judge the start, middle and end stages of the mixed oil interface, and the method does not need a complex modeling process, and has fast operation speed and good real-time performance.

6) And the result is output to a crude oil pipeline transmission scheduling control system, and a downstream oil receiving station can cut and receive oil in time.

In order to further demonstrate the reliability of the method, the condition that a plurality of oil mixing interfaces exist when different types of crude oil are conveyed in a pipeline for many times is detected. Similar to the foregoing process, the sum of the first derivative of the standard deviation of absorbance at each wave number point in the spectral feature region is continuously calculated, and the judgment result of the oil-mixing interface for transporting different types of crude oil for multiple times is shown in fig. 4. The figure clearly shows the curve of the 3-segment oil mixing period, and obviously, the method can accurately detect and track the mixed oil product in real time.