RU2441219C1 - Method of determining component composition of natural gas in real time - Google Patents

Abstract

FIELD: physics. ^ SUBSTANCE: method involves periodic or simultaneous measurement of radiation absorption by an analysed gas at different wavelength combinations (from eight to nine wavelengths in each combination) in the infrared range, corresponding to centres of absorption bands of natural gas components. Said combinations are selected based on the category of the natural gas, which in turn is determined by preliminary measurement of radiation absorption by the analysed gas at seven wavelengths in the infrared range. ^ EFFECT: invention enables analysis of the component composition of natural gas mixtures in real time, which is enough to perform component quantitative analysis of natural gas mixtures of different composition in accordance with Russian and international standards. ^ 7 dwg, 6 tbl

Description

The invention relates to the field of absorption spectroscopy and can be used for component analysis of natural gas and gas mixtures based on it in real time.

The determination of the component composition of natural gas and the composition of gas mixtures close to it in real time is one of the main tasks facing gas companies. This is due to the fact that the composition of natural gas can vary significantly depending on the field (in particular, the proportion of methane in natural gas can range from 99.2% to 63.7% for different fields [1]). It should be borne in mind that as gas mixtures spread in the main pipelines and when they are stored in underground storages, natural gas mixes from various fields. As a result of this, the component composition of natural gas entering the consumer can vary significantly over time, which leads to a change in the consumer properties of natural gas, in particular, a change in the heat of combustion. This, in turn, determines the differences in the cost of natural gas of different component composition.

Therefore, information on the component composition of natural gas is one of the most important parameters that significantly affect the economic efficiency of gas companies. This issue is particularly acute when gas is supplied through gas pipelines that run through several countries, since the lack of such information does not allow gas supply companies to defend their economic interests to the end consumer. In practice, there may be situations when the component composition of natural gas transmitted by the supplier to the final consumer may change significantly during transit through one or more countries.

The main components of natural gas are methane, ethane, propane, butane and pentane with a predominant methane content. Depending on the content of other components in natural gas, it seems reasonable to divide natural gas into three main categories. For the first category, the proportion of methane is 95% or more. Other categories include natural gas, in which the methane content is less than 95%, and the total share of any two or three other components, as a rule, is more than ten percent with the mass fraction of the remaining components not exceeding tenths of a percent. Moreover, the second category is characterized by a high content of ethane and propane, and the third category - by a high content of butane and pentane.

In addition, natural gas may contain some impurity components as impurities: hydrogen sulfide, carbon dioxide and molecular nitrogen, the total concentration of which does not exceed 1-1.5%.

As a rule, the main of these impurity components for most natural gas deposits are carbon dioxide and molecular nitrogen, and the proportion of nitrogen in most cases is several times greater than the proportion of carbon dioxide. However, it should be noted that for a nitrogen molecule that is homonuclear and does not have its own dipole moment, vibrational-rotational transitions are forbidden in the infrared region of the spectrum. As a result of this, the absorption cross sections for this molecule are too small for the presence of molecular nitrogen in the gas mixture to lead to any noticeable deterioration in the accuracy of detection of its main components. At the same time, vibrational-rotational transitions are allowed for a carbon dioxide molecule, and the contribution of carbon dioxide to the total absorption can be significant, i.e. without taking into account carbon dioxide absorption, the accuracy of hydrocarbon detection may be low.

From the above it follows that in practice, first of all, it is necessary to analyze the following components of natural gas: methane, ethane, propane, butane, pentane, carbon dioxide.

Currently, there are two GOSTs for determining the component composition of natural gas developed at VNIIGAZ and VNIINP: GOST 23781-87 [2] and GOST 14920-79 [3]. The following international standards comply with these GOSTs: ISO 6974-1-ISO 6974-6, ISO 6569 and ISO 6568 [4-6].

In 2008, the Interstate Council for Standardization, Metrology and Certification (Minutes No. 33 dated 06.06.2008) adopted a new set of GOSTs, which are the standard for determining the component composition of natural gas, developed at FSUE VNIIM im. D.I. Mendeleev: GOST 31371.1-2008-GOST 31371.6-2008 [7]. This standard is modified in relation to international standards ISO 6971-1: 2000 [8-12].

By order of the Federal Agency for Technical Regulation and Metrology dated 03.12.208 No. 340-st, this standard was put into effect as the national standard of the Russian Federation from 01.01.2010

All existing methods for determining the component composition of natural gas based on these standards [13] are chromatographic and vary depending on the amount of hydrocarbon components in natural gas, as well as compounds that are not natural gas components, but may be present as small impurities in industrial gas mixtures (oxygen, carbon monoxide and lower opheline). Chromatographic methods have a number of disadvantages, the main of which is the inability to carry out analysis in real time. Among other disadvantages, one can single out a rather large error in determining the molar fractions of the components of natural gas, as well as the high cost and complexity of the equipment used.

The listed disadvantages are associated with the fundamental features of chromatographic analysis and cannot be eliminated within the framework of this method.

Due to the above, chromatographic methods should be recognized as outdated and do not meet modern requirements, which determines the need to develop new methods for monitoring the gas composition, free from the disadvantages of the applied analysis methods.

From the point of view of eliminating the disadvantages of chromatographic methods, one of the most promising methods is optical spectroscopy, which are used in combination with automated means for processing measurement results.

There are methods [14–16] based on the determination of hydrocarbon concentrations in gaseous media by Raman scattering (Raman spectroscopy). In these methods, the radiation of a monochromatic source (laser) is directed to a measuring cell containing natural gas and excites Raman spectra of its constituent components. Selectivity is achieved due to the fact that, depending on the chemical formula of the natural gas component, the frequencies of Raman spectral lines differ. By registering the intensities of the Raman signals at the corresponding frequencies, a conclusion is drawn about the concentration of a component in the analyzed sample. Despite the fact that the measurements are carried out in real time (the time between measurements, taking into account the processing of the measured signals, does not exceed 2-3 s), the disadvantage of this method is that the intensities of the recorded Raman signals are small. This means that to measure the content of small impurities in a natural gas sample, it is necessary to use high-power lasers, which are expensive and structurally complex devices. In addition, to date, measurements have been carried out only for a small amount of natural gas components (methane, ethane, ethylene).

The known method [17], designed to detect gas leaks from underground gas pipelines. The authors of the work developed an absorption gas analyzer of hydrocarbons based on a He-Ne laser (“LGAU-02”), designed for this purpose. By measuring the attenuation of radiation on the generation line of a He-Ne laser with a wavelength of 3.3922 μm, it is concluded that the methane content is intensively absorbing at a given wavelength. The developed device is suitable only for qualitative estimates, suitable for detecting gas leaks, but not allowing it to be used both to determine the component composition of the gas and to detect the exact methane content in the gas pipeline, which is associated with the presence of a number of hydrocarbons included in the selected wavelength composition of natural gas.

In [18], the results of determining the component composition of natural gas mixtures by infrared absorption spectroscopy are presented. At the first stage, studies were carried out with artificial gas mixtures consisting of methane, ethane and propane. Molar fractions of methane varied in the range of 86-95% of the molecular weight, ethane in the range of 2-10% of the molecular weight, propane - from 0 and 4% of the molecular weight. Next, measurements were carried out using the described method for the concentrations of these three substances in natural gas samples taken from the pipeline. The results obtained were compared with the results obtained by the traditional chromatographic method. Satisfactory results were obtained for methane and ethane; the error in measuring propane was significant. The authors attribute this result to a significantly lower (several times) propane content in the studied gas mixtures compared to the other two components. Thus, it is not possible to use this method to solve the problem of component analysis of natural gas, in which the shares of various components can vary significantly.

Closest to the technical nature of the claimed invention is a method for detecting a mercaptan odorizing mixture of natural gas (odorant) in real time [19]. This method is based on measuring the absorption of probing radiation by odorized gas at four wavelengths, one of which λ ₁ corresponds to the minimum absorption of the odorant, and the other three λ ₂ , λ ₃ and λ ₄ correspond to the absorption maxima of the odorant. While λ ₁ support in the range from 7.6 to 7.8 μm, corresponding to the maximum absorption of non-ionized gas, λ ₂ support in the range from 9.0 to 9.2 μm, λ ₃ support in the range from 10.3 to 10, 5 μm, and λ _{4 is} maintained in the range from 15.6 to 15.8 μm, while the optical path of the probe radiation in the measuring cell is maintained in the range from 3 to 10 m. In this case, the radiation intensities are determined at the corresponding wavelengths before entering it into a measuring cell with a non-ionized mixture of natural gas and the intensity of cheniya passing through the measuring cell, when pumping therethrough neodorizovannoy natural gas mixture. The radiation intensities are measured before entering into the measuring cell with an already odorized mixture of natural gas and the intensities of the radiation transmitted through the measuring cell with an already odorized mixture of natural gas at the above four wavelengths. All signals are processed by the computer complex and then the concentrations of odorant and natural gas components are calculated from the solution of an inhomogeneous linear system of equations, in this case, by the least squares method with iterative refinement of the roots using the Householder transformations.

The technical task of the invention is the creation of a method for determining the component composition of natural gas and close to it by the composition of gas mixtures in real time.

The problem is solved by cyclically or simultaneously measuring the absorption of radiation by the analyzed gas at various wavelength combinations (from eight to nine wavelengths in each combination) of the infrared spectral range corresponding to the centers of the absorption bands of natural gas components. These combinations are selected based on which category the natural gas mixture belongs to, which, in turn, is determined by a preliminary measurement of the absorption of radiation by the analyzed gas at seven infrared wavelengths corresponding to the centers of the absorption bands of the components of natural gas.

According to the invention, a preliminary measurement of radiation absorption is carried out at seven wavelengths, while the wavelength λ ₁ support in the range from 7.6 μm to 7.8 μm, λ ₂ support in the range from 6.898 μm to 6.902 μm, λ ₃ support in the range from 6.87 to 6.89 μm, λ ₄ support in the range of 12.1 to 12.3 μm, λ ₅ support in the range of 15.2 to 15.4 μm, λ ₆ support in the range of 6.92 to 6 , 94 μm and λ ₇ support in the range from 6.908 μm to 6.912 μm. Moreover, for the first category of natural gas, for which the methane fraction is equal to or greater than 95% and for the second category of natural gas, for which the methane fraction is less than 95%, the relation n _e + n _pr > n _b + n _pent , where n _e , n _pr , n _b , n _pent are the concentrations of ethane, propane, butane and pentane in the analyzed gas, respectively, the final measurement of radiation intensities at eight wavelengths is carried out, where λ ₁ -λ ₇ are indicated above, and the wavelength λ _{8 is} maintained in the range from 10.518 microns to 10.522 microns for the first category of natural gas and in the range from 13.66 to 13.76 m m for the second category of natural gas. For the third category of natural gas, for which the methane fraction is less than 95% and the ratio n _e + n _pr > n _b + n _{pent is fulfilled} , the final measurement of radiation intensities at nine wavelengths is carried out, where λ ₁ -λ ₇ are indicated above; a wavelength of λ _{8 is} maintained in the range of 13.66 to 13.76 μm, and a wavelength of λ _{9 is} maintained in the range of 10.508 μm to 10.512 μm. Part of the luminous flux from the light source is sent to the meter (or meters) to control the values of the intensities at the entrance to the measuring cell. The radiation intensities J ₀₁ ^* , ..., J ₀₇ ^{* are} measured before entering it into the evacuated measuring cell and the radiation intensities J ₁ ^* , ..., J ₇ ^* passing through the evacuated measuring cell at the above wavelengths λ ₁ -λ ₇ and introduced the values of these intensities in the computing complex. Then pumping through the measuring cell of natural gas is started and the radiation intensities J ₀₁ , ..., J ₀₇ are measured before entering into the measurement cell and the radiation intensities J ₁ , ..., J ₇ passed through the measuring cell with natural gas at the above seven wavelengths λ ₁ -λ ₇ . The signals corresponding to the intensities J ₀₁ , ..., J ₀₇ , J ₁ , ..., J ₇ are routed to the computational complex, and the concentrations of the natural gas components are calculated from the solution of the heterogeneous linear system of equations, in this case, the least squares method with iterative refinement of the roots using Householder transformations:

where N is the number of wavelengths at which measurements are made,

n _m , n _e , n _pr , n _b , n _pent , n _co2 are the concentrations of methane, ethane, propane, butane, pentane and carbon dioxide in the analyzed natural gas, respectively,

L is the length of the measuring cell;

σ _m (λ _i ), σ _e (λ _i ), σ _pr (λ _i ), σ _b (λ _i ), σ _pent (λ _i ), σ _co2 (λ _i ) (i = 1 ... N) - sections absorption of methane, ethane, propane, butane, pentane and carbon dioxide, respectively, at wavelengths λ ₁ , ..., λ _N ;

k _i (i = 1 ... N) are the coefficients equal to the ratio of the intensities J _i ^* / J _0i ^* , where J _i ^* are the intensities of radiation at the corresponding wavelengths λ _i passed through the evacuated measuring cell, and J ₀ ^* is the radiation intensities at the appropriate wavelengths λ _i before entering it into the evacuated measuring cell;

J _0i (i = 1 ... N) - radiation intensity at the corresponding wavelengths λ _i , before entering them into the measuring cell with natural gas;

J _i (i = 1 ... N) - radiation intensity at the corresponding wavelengths λ _i passed through the measuring cell with natural gas.

Then, for the first and second categories of natural gas, an additional measurement of the radiation intensities J ₀₁ ^* , ..., J ₀₈ ^* and J ₁ ^* , ..., J ₈ ^* is carried out before and after the radiation is introduced into the evacuated measuring cell, respectively, at the above eight wavelengths λ ₁ -λ ₈ . For the third category of natural gas, an additional measurement of the radiation intensities J ₀₁ ^* , ..., J ₀₉ ^* and J ₁ ^* , ..., J ₉ ^* is carried out before and after the radiation is introduced into the evacuated measuring cell, respectively, at the above nine wavelengths λ ₁ - λ ₉ . Then these intensities are introduced into the computer complex, after which they pump natural gas through the measuring cell and measure the radiation intensities J ₀₁ , ..., J _0N before entering it into the measuring cell and the radiation intensities J ₁ , ..., J _N passed through the measuring cell with natural gas, where N = 8 for the first and second categories of natural gas, N = 9 for the third category of natural gas, at the above wavelengths λ ₁ -λ ₈ for the first and second categories of natural gas or λ ₁ -λ ₉ for the third category natural gas. The signals corresponding to the intensities J ₀₁ , ..., J _0N , J ₁ , ..., J _{N are sent} to the computer complex, and the final concentrations of the natural gas components are calculated by the formula (1), in this case, the least squares method with iterative refinement of the roots, s using householder transformations.

Figure 1 shows the absorption cross-sections of the components of natural gas - methane (1), ethane (2), propane (3), butane (4), pentane (5) and CO ₂ (6) - at atmospheric pressure.

Figure 2 shows the dependences of the absorption coefficients of the methane: ethane mixture on the fraction of ethane at wavelengths of 7.68 μm (1), 6.88 μm (2) and 12.24 μm (3) at full atmospheric pressure of the gas mixture.

Figure 3 shows the dependences of the absorption coefficients of a mixture of methane: ethane: propane with a ratio of ethane: propane = 1: 1 on the total fraction of ethane and propane at wavelengths of 7.68 μm (1), 6.90 μm (2), 6, 88 μm (3), 12.24 μm (4) and 13.64 μm (5) at a total atmospheric pressure of the gas mixture.

Figure 4 shows the experimental and calculated dependences of the error in determining the concentrations of methane, ethane and propane for the gas mixture of the Medvezhye, Jura deposit on the choice of measurement wavelengths.

Figure 5 shows the dependence of the error in determining the concentration of carbon dioxide from its concentration in the gas mixture of the field "Bear, Jura".

Figure 6 shows the experimental (points) and calculated (lines) dependences of the error in determining the concentration of carbon dioxide (1) and methane (2) on the proportion of CO ₂ in the mixture of the Medvezhye, Jura deposit.

Figure 7 shows the dependences of the detection error of butane (solid line) and pentane (dashed line) on the choice of a combination of wavelengths at which absorption measurements are made. The total proportion of butane and pentane with a ratio of 1: 1 in the mixture is 10 ^-3 %.

In Fig. 1, the absorption cross sections σ of methane (1), ethane (2), propane (3), butane (4), pentane (5) and CO ₂ (6) are plotted along the vertical axis, depending on the wavelength in the range from 6 to 17 microns.

The absorption cross sections were determined from the measured radiation intensities at a wavelength λ according to the following formula:

where σ is the absorption cross section of the natural gas component;

λ is the radiation wavelength of the source;

α is the absorption coefficient of the natural gas component;

n is the concentration of the component of natural gas;

L is the length of the measuring cell;

J ₀ (λ) is the intensity of the radiation transmitted through the cell in the absence of a natural gas component;

J (λ) is the intensity of the radiation transmitted through the cell in the presence of a natural gas component.

The concentration of the natural gas component n was determined from the ratio

where p is the pressure of the natural gas component in the measuring cell, equal to atmospheric;

k is the Boltzmann constant;

T is the temperature.

The analysis of figure 1 shows that the investigated components of natural gas have a number of absorption bands in the studied area. Methane has one absorption band centered at a wavelength of 7.68 microns. Ethane has two absorption bands with centers at wavelengths of 6.88 and 12.24 microns. In propane, 5 absorption bands with centers at wavelengths of 6.90 were detected; 8.77; 9.60; 11.16 and 13.64 microns, with the greatest absorption being achieved at wavelengths of 6.90 and 13.64 microns. Butane has three absorption bands with centers at wavelengths of 6.93; 10.51 and 13.49 microns. Three absorption bands with centers at wavelengths of 6.91 microns were found in pentane; 10.52 and 13.71 microns. In CO ₂ , one absorption band was found centered at a wavelength of 15.31 microns.

Despite the fact that the component analysis of natural gas is always carried out at the same total pressure of the gas mixture equal to atmospheric, the proportion of any component in the mixture can fluctuate over a wide range (for example, the proportion of ethane or propane is from 0 to 15 percent, and share of methane from 99 to 70 percent). Since the collisional broadening coefficients of natural gas components can vary, this can lead to nonlinear dependences of the absorption coefficients at fixed wavelengths for natural gas of different compositions.

To study the dependences of the absorption coefficients on the fractions of natural gas components in a mixture with each other, several mixtures of various compositions were prepared. Initially, a mixture of methane and ethane was studied, for which, in the studied spectral range, in steps of 1 μm, the dependences of the absorption coefficients on the ethane fraction in this mixture were measured, which varied from tenths of a percent to 15% at full atmospheric pressure of the whole mixture.

Figure 2 shows, as an example, the obtained dependences at wavelengths corresponding to the centers of the methane and ethane bands. As you can see, the dependencies are linear. At other wavelengths, linear relationships are also obtained.

Similar results were obtained for mixtures of methane-propane, methane-butane and methane-pentane.

Then, mixtures containing simultaneously all the components of natural gas in various proportions were investigated. In all cases, the dependences also turned out to be linear. As an example, Fig. 3 shows the dependences of the absorption coefficients for methane-ethane-propane mixtures on the total fraction of ethane and propane in the mixture at a ratio of 1: 1. The measurements were carried out at wavelengths corresponding to the centers of the absorption bands of ethane and methane, as well as at two wavelengths of 6.90 and 13.64 μm, corresponding to the maximum values of the absorption cross sections of propane.

Thus, the studies showed that in the entire spectral range studied, the dependences of the absorption coefficients on the fractions of natural gas components are linear for any possible ratios between these components. This makes it possible to unambiguously determine the proportions of natural gas components from the measured absorption coefficients.

The sensitivity of the detection of natural gas components, taking into account the measurement error based on the standard signal-to-noise ratio = 1, can be obtained from equation (2):

where n is the concentration of the component of natural gas;

σ (λ) is the absorption cross section of the component of natural gas;

λ is the radiation wavelength of the source;

L is the length of the measuring cell;

J ₀ (λ) is the intensity of the transmitted radiation in the absence of a natural gas component;

J (λ) is the intensity of transmitted radiation in the presence of a component of natural gas.

Obviously, the best sensitivities for all components of natural gas are achieved at wavelengths corresponding to the maximum values of their absorption cross sections in the studied range. However, it should be noted that these sensitivities are achievable provided that only one of the considered components is present in the test mixture, which in real conditions is almost never realized. To solve the problem of simultaneous determination of all components included in a gas mixture, it is necessary to use a technique that provides for the simultaneous measurement of the intensities of radiation transmitted through an absorbing cell at several wavelengths, the values and quantity of which are not obvious in advance and depend on which category the gas mixture under study belongs to.

The inventors conducted studies to determine these wavelengths.

Let σ _m (λ _i ), σ _e (λ _i ), σ _pr (λ _i ), σ _b (λ _i ), σ _pent (λ _i ), σ _co2 (λ _i ) be the absorption cross sections of methane, ethane, and propane , butane, pentane and carbon dioxide, respectively, at different wavelengths λ ₁ , ..., λ _N. Then, assuming that the intensity of the radiation transmitted through the cell changes in accordance with Bouguer’s law, we obtain the following system of equations for n _m , n _e , n _pr , n _b , n _pent , n _co2 :

where n _m , n _e , n _pr , n _b , n _pent , n _co2 are the concentrations of methane, ethane, propane, butane, pentane and carbon dioxide in the analyzed natural gas, respectively;

σ _m (λ _i ), σ _e (λ _i ), σ _pr (λ _i ), σ _b (λ _i ), σ _pent (λ _i ), σ _co2 (λ _i ) (i = 1 ... N) - sections absorption of methane, ethane, propane, butane, pentane and carbon dioxide, respectively, at wavelengths λ _i ;

λ _i (i = 1, ..., N) - absorption wavelengths;

L is the length of the measuring cell;

J _0i (i = 1, ..., N) - radiation intensity at the corresponding wavelengths λ _i before entering them into the measuring cell with natural gas;

J _i (i = 1, ..., N) - radiation intensity at the corresponding wavelengths λ _i passed through the measuring cell with natural gas;

J _0i * is the radiation intensity at the corresponding wavelengths λ _i before entering it into the evacuated measuring cell;

J _i ^* is the radiation intensity at the corresponding wavelengths λ _i passed through the evacuated measuring cell.

Since the intensities J _0i , J _i , J _0i ^* and J _i ^* are measured experimentally, this system is solvable when measuring at six wavelengths.

However, both the measured radiation intensities and the absorption cross sections for the components of natural gas are determined with errors, and we are dealing with an inhomogeneous system of equations of the form Ax≈b:

where σ _m (λ _i ), σ _e (λ _i ), σ _pr (λ _i ), σ _b (λ _i ), σ _pent (λ _i ), σ _co2 (λ _i ) (i = 1 ... N) - absorption cross-sections of methane, ethane, propane, butane, pentane and carbon dioxide, respectively, at wavelengths λ _i ;

λ _i (i = 1, ..., N) - absorption wavelengths;

n _m , n _e , n _pr , n _b , n _pent , n _co2 are the concentrations of methane, ethane, propane, butane, pentane, and carbon dioxide in the analyzed natural gas, respectively;

L is the length of the measuring cell;

k _i (i = 1 ... N) are the coefficients equal to the ratio of intensities J _i ^* / _0i ^* , where J _i ^* are the intensities of radiation at the corresponding wavelengths λ _i passing through the evacuated measuring cell, and J _oi ^* are the intensities of radiation at the corresponding wavelengths λ _i before entering it into the evacuated measuring cell;

J _oi (i = 1, ..., N) - radiation intensity at the corresponding wavelengths λ _i before entering them into the measuring cell with natural gas;

J _i (i = 1, ..., N) is the radiation intensity at the corresponding wavelengths λ _i passed through the measuring cell with natural gas.

Generally speaking, this system may turn out to be either incompatible or indefinite, as a result of which the error in calculating the concentrations of the studied substances may turn out to be large.

An increase in the accuracy of the obtained solutions in this case can be achieved by redefining the system of equations, i.e. taking measurements at more than six wavelengths.

The study of system (6) was carried out using the least squares method with iterative refinement of the roots, which used the QR decomposition of matrix A obtained by applying a sequence of orthogonal Householder transformations [20-23]. The accuracy criterion for the obtained solutions was minimization of the sum of squared residuals.

The first were studied mixtures of the second category of natural gas. Initially, mixtures containing methane, ethane and propane were investigated. A total of 18 mixtures with various proportions of these components were analyzed (table 1). Computational studies of system (6) showed that the smallest errors in determining the fraction of components for all cases are achieved on the wavelength combinations shown in Table 2. For each combination, the errors in determining the individual components increased as their content in the mixture decreased. Table 2 shows the error values for each combination corresponding to the maximum (upper row) and minimum (lower row) fraction of the component in the mixture.

As it turned out, the smallest errors correspond to measurements on a combination of 4 wavelengths (combination (3)). These wavelengths in their values correspond to the centers of the absorption bands of the components of the mixture.

As one would expect, the smallest error is achieved for methane (it practically does not change from the fraction of methane in the mixture) due to its significantly higher content in the mixture compared to other components and significantly larger absorption cross sections in the range of 7-8 μm. The propane measurement error is greatest, since its absorption band overlaps with the methane absorption band. At the same time, ethane has an absorption band centered at a wavelength of 12.24 microns, not overlapping with the methane absorption band. Errors in the determination of ethane are approximately the same for combinations of wavelengths (2) and (3) and are approximately an order of magnitude better compared to combination (1). However, for combination (3), the propane measurement error becomes slightly less.

Further studies were carried out with the gas mixture of the Medvezhye, Jura deposit, which was characterized by a high content of ethane and propane (the content of methane, ethane and propane was 76.58%; 10.2% and 12.6%, respectively). The shares of other components in this mixture are extremely small (the content of pentane and carbon dioxide is at the level of 0.01% for each substance; butane is completely absent), which allows us to speak with good approximation of this mixture as consisting of three components - methane, ethane and propane.

For this gas mixture, both computational and experimental studies were conducted to determine the error in determining the concentrations of its components depending on the choice of combinations of measurement wavelengths. The studies were carried out on combinations of wavelengths (1) - (3) of Table 2. Figure 4 shows the experimental and calculated dependences of the error in determining the concentrations of methane, ethane and propane on the choice of a combination of measurement wavelengths. Experimental studies have shown that the minimum errors are observed during measurements on a combination of wavelengths (3) and are 1.2% for methane; 1.3% for ethane and 1.6% for propane. The calculation results are in good agreement with experimental data.

Based on this, further studies used a combination of wavelengths of 7.68; 12.24; 6.90; 6.88 microns.

Further, carbon dioxide was added to this gas mixture in various proportions. The ratio of methane, ethane and propane, as well as the total pressure of the mixture equal to atmospheric, remained unchanged. The content of hydrocarbons and CO ₂ in the mixture was measured at a combination of wavelengths (3) from Table 2, as well as at a wavelength of λ ₅ = 15.31 μm, which coincides with the center of the carbon dioxide absorption band.

Figure 5 shows the experimentally obtained dependence of the detection error of carbon dioxide Δ on its concentration in the mixture. It can be seen that Δ remains small up to fractions of CO ₂ in the mixture of the order of 0.01%. In this case, the hydrocarbon detection error was at a level corresponding to the best values given in table 2 for a combination of wavelengths (3), and practically did not deteriorate even at a 10% CO ₂ fraction in the mixture due to the considerable remoteness of the carbon dioxide absorption band relative to methane absorption bands, ethane and propane.

In Fig.6. experimental (points) and calculated (lines) dependences of the error in determining the concentration of carbon dioxide and methane on the proportion of CO ₂ in the mixture of the Medvezhye, Jura deposit are shown.

As can be seen, the agreement between the experimental and calculated results in all the studies was quite good, as a result of which further studies were carried out using numerical simulation.

At the next stage, a mixture of methane, ethane, propane and CO ₂ was simulated in a ratio of 0.766: 0.102: 0.126: 0.01 (an analogue of the gas mixture of the Medvezhye, Jura deposit) to which were sequentially “added” (by modeling) as of small impurities of butane and pentane both individually and together in various ratios with each other, wavelengths were optimized in terms of achieving the best sensitivity of their detection and the best measurement accuracy of all components. The assessment was carried out during calculations on a combination of wavelengths (3) and a wavelength of 15.31 microns, as well as on one or several additional wavelengths that coincide with the centers of the absorption bands of butane and pentane.

First, calculations were carried out when “adding” butane to the initial mixture. At the first stage, the sensitivities of determining butane were calculated during studies at one of three additional wavelengths corresponding to the centers of its absorption bands. The best sensitivity for detecting butane was achieved when measuring at a wavelength of 10.51 μm and amounted to 1.4 · 10 ^-2 % of the volume. The worst sensitivity was obtained for a wavelength of 6.93 μm, which is obviously due to the proximity of this absorption band to the intense absorption band of methane, the concentration of which in the mixture is significantly higher. In addition, as the butane content increased, the detection accuracy of the remaining components decreased. This especially affected the accuracy of determining the concentration of propane for the same reason - the absorption bands of methane, propane and butane are in close proximity to each other.

At the next stage, the sensitivities were determined when the centers of any two or all three absorption lines of butane were selected as the additional wavelengths of measurements. The results are shown in table 3.

It can be seen that the best sensitivity is achieved while simultaneously detecting at additional wavelengths of 10.51 and 13.49 microns. Compared with measurements at a single absorption wavelength of butane, the obtained value is approximately 50 times better and amounts to 2.6 · 10 ^-4 %. The detection error of the other components of the mixture in all cases did not exceed two percent.

The next series of calculations was carried out when pentane was added to the initial mixture. As additional wavelengths of 6.91 were chosen; 10.52 and 13.71 microns, coinciding with the centers of its absorption lines. The studies were carried out in the same manner as in the case of butane. The final results are summarized in table 4.

It is seen that the best sensitivity is achieved with simultaneous detection at additional wavelengths of 10.52 and 13.71 microns. Comparison with the results obtained for butane with an appropriate choice of wavelengths shows that the sensitivities of detection of pentane and butane are quite close - for all combinations of wavelengths, the results obtained for pentane are 1.2-1.5 times better. In addition, the sensitivities for other components remain virtually unchanged. The result obtained has a fairly obvious explanation. A comparison of the absorption spectra of butane and pentane shows that these two substances have a similar structure of the absorption bands: in both cases, there are three absorption bands that are close in wavelength and have approximately the same absorption cross sections. Nevertheless, the fact that the absorption cross-sections of pentane is slightly higher than the absorption cross-sections of butane, and explains the slightly better detection sensitivity of this substance.

At the final stage of the research, butane and pentane were simultaneously added to the initial mixture of methane, ethane, propane and carbon dioxide in the ratios 1: 4, 1: 1 and 4: 1.

Figure 7 shows, by way of example, the errors in the detection of butane (solid line) and pentane (dashed line) Δ depending on the choice of wavelength combinations in the case where the ratio of butane and pentane is 1: 1 (the total fraction of butane and pentane in the mixture 10 ^-3 %).

It can be seen that these errors substantially depend on the values and the number of wavelengths at which measurements are made. So, for combinations (10) (additional wavelengths of 6.93; 6.91; 10.52 μm) and (18) (13.49; 10.52; 13.71 μm), the detection errors of both butane and pentane are at the level of 30-45%, while for combinations (11) (6.93; 6.91; 13.71 microns) and (20) (10.51; 6.91; 10.52; 13, 71 microns) the error in the determination of butane is at the level of 69-72%.

Similar calculation and experimental studies were conducted for the gas mixture from the Vuktylskoye field (methane content - 84.86%, ethane - 9.8%, propane - 3.8%, butane - 0.94%, pentane - 0.3% carbon dioxide - 0.3%).

The final results showed that for the second category of natural gas, measurements should be made on a combination of wavelengths of 7.68; 6.90; 6.88; 12.24; 15.31; 6.93; 6.91 and 13.71 microns. Table 5 shows the sensitivity values for the detection of butane and pentane for their various ratios on the indicated wavelength combinations for mixtures modeled on the basis of gas mixtures Medvezhye, Jura and Vuktylskoye.

Moreover, the measurement error of methane did not exceed 1.4%, ethane - 1.8%, propane - 2.6% and carbon dioxide - 1% for all considered ratios of butane and pentane.

For studies of the third category of natural gas, a gas mixture was selected from the Medvezhye, Cenoman field (methane - 86.9%; butane - 7.6%, pentane - 3.6%; ethane - 1.2%; CO ₂ - 0 , 7%). Research was conducted in a similar way. It was found that the best detection sensitivity of the components of this mixture are achieved on a combination of nine wavelengths of 7.68; 6.90; 6.88; 12.24; 15.31; 6.93; 6.91; 13.71 and 10.51 microns and comprise 6 · 10 ^-2 %; 8.2 · 10 ^-3 % and 8.1 · 10 ^-4 % of the volume for ethane, carbon dioxide and propane, respectively. Moreover, the measurement errors of the remaining components do not exceed 1.5%.

Further similar studies were carried out for thirteen gas mixtures of the first category from various Russian fields. The methane content in them ranged from 94.8% (Berezovskoye) to 99.2% (Bovanenkovskoye, Cenomanian). The content of other components in various mixtures was significantly different. Thus, ethane, propane, butane and pentane were practically absent in the mixture of the Urengoyskoye, Cenoman deposit (their total content was at the level of 0.01%), and the shares of carbon dioxide and nitrogen were 0.29 and 0.8%, respectively. In a number of other mixtures, the content of all these components could reach several percent (for example, the ethane content in the Bovanenkovskoye, deposits was 2.89%).

Calculations were performed for all these mixtures, and experimental studies were additionally carried out for the mixture from the Urengoyskoye, Senoman field (98.9% methane).

The best in terms of sensitivity wavelength combinations were determined. The results showed that a combination of wavelengths of 7.68 is optimal for all mixtures; 6.90; 6.88; 12.24; 15.31; 6.93; 6.91 and 10.52 microns. As an example, table 6 shows data on the sensitivity of detection of components of 5 different fields.

When detecting natural gas in advance of an indefinite composition, it is first necessary to find out which category it belongs to. For this, measurements should be made at seven wavelengths corresponding to the centers of the absorption bands of all its possible components, namely λ ₁ = 7.68 μm (methane); λ ₂ = 6.90 μm (propane); λ ₃ = 6.88 μm and λ ₄ = 12.24 μm (ethane); λ ₅ = 15.31 microns (carbon dioxide); λ ₆ = 6.93 μm (butane) and λ ₇ = 6.91 μm (pentane) and calculate the concentration of these substances by the formula (1), where N = 1. After identifying the category of natural gas, another measurement should be carried out at the above wavelengths λ ₁ -λ ₇ and additionally at wavelengths that depend on the category of natural gas. For the first category of natural gas at one additional wavelength, λ ₈ = 10.52 μm; for the second category of natural gas at one additional wavelength λ ₈ = 13.71 microns and for the third category of natural gas at two additional wavelengths λ ₈ = 13.71 microns and λ ₉ = 10.51 microns. Next, the final calculation of the concentrations of natural gas components is performed according to the formula (1), where N = 8 for the first and second categories of natural gas and N = 9 for the third category of natural gas.

The inventive method for determining the component composition of natural gas in real time is carried out in the following sequence.

The measuring cell is irradiated with radiation at 7 wavelengths λ ₁ , ..., λ ₇ . The wavelength λ ₁ support in the range from 7.6 μm to 7.8 μm, λ ₂ support in the range from 6.898 μm to 6.902 μm, λ ₃ support in the range from 6.87 to 6.89 μm, λ ₄ support in the range from 12.1 to 12.3 microns, λ ₅ support in the range from 15.2 to 15.4 microns, λ ₆ support in the range from 6.92 to 6.94 microns and λ ₇ support in the range from 6.908 microns up to 6.912 microns. Part of the luminous flux from the light source is sent to the meter (or meters) to control the values of the intensities at the entrance to the measuring cell. The radiation intensities J ₀₁ ^* , ..., J ₀₇ ^{* are} measured at the corresponding wavelengths before entering it into the evacuated measuring cell and the radiation intensities J ₁ ^* , ..., J ₇ ^* passed through the evacuated measuring cell, and these intensities are entered into the computer complex . Then, pumping through the measuring cell of natural gas is started and the radiation intensities J ₀₁ , ..., J ₀₇ are measured before entering it into the measuring cell and the radiation intensities J ₁ , ..., J ₇ passed through the measuring cell with natural gas at the above wavelengths. The signals corresponding to the intensities J ₀₁ , ..., J ₀₇ , J ₁ , ..., J _{7 are} sent to the computer complex, and the concentrations of natural gas components are calculated by the formula (1) (where N = 7), in this case, by the least squares method with iterative refinement of the roots, using the Householder transformations.

When the proportion of methane in the mixture is equal to or greater than 95%, a repeated measurement of radiation intensities J ₀₁ ^* , ..., J ₀₈ ^* and J ₁ ^* , ..., J ₈ ^{* is} carried out before and after the radiation is introduced into the evacuated measuring cell, respectively, on the indicated higher wavelengths λ ₁ -λ ₇ and at an additional wavelength λ ₈ , which is maintained in the range from 10.518 μm to 10.522 μm. Then these intensities are introduced into the computer complex, after which they pump natural gas through the measuring cell and measure the radiation intensities J ₀₁ , ..., J ₀₈ before entering it into the measuring cell and the radiation intensities J ₁ , ..., J ₈ passed through the measuring cell with natural gas at the above wavelengths. The signals corresponding to the intensities J ₀₁ , ..., J ₀₈ , J ₁ , ..., J ₈ are routed to the computer system, and the final concentrations of the natural gas components are calculated by the formula (1) (where N = 8), in this case, the least squares method with iterative refinement of the roots, using the Householder transformations.

When the proportion of methane is less _than 95%, and the ratio n _e + n _pr > n _b + n _{pent is fulfilled} , where n _e , n _pr , n _b , n _pent are the concentrations of ethane, propane, butane and pentane, respectively, an additional measurement of the intensities radiation J ₀₁ ^* , ..., J ₀₈ ^* and J ₁ ^* , ..., J ₈ ^* before and after introducing radiation into the evacuated measuring cell, respectively, at the above wavelengths λ ₁ -λ ₇ and at an additional wavelength λ ₈ , which support in the range from 13.66 to 13.76 microns. Then these intensities are introduced into the computer complex, after which they pump natural gas through the measuring cell and measure the radiation intensities J ₀₁ , ..., J ₀₈ before entering it into the measuring cell and the radiation intensities J ₁ , ..., J ₈ passed through the measuring cell, with natural gas at the above wavelengths. The signals corresponding to the intensities J ₀₁ , ..., J ₀₈ , J ₁ , ..., J ₈ are routed to the computer complex, and the final concentrations of the natural gas components are calculated by the formula (1), where N = 8, in this case, the least squares method with iterative refinement of the roots, using Householder transformations.

When the proportion of methane is less _than 95% and the ratio n _e + n _pr <n _b + n _{pent is fulfilled} , where n _e , n _pr , n _b , n _pent are the concentrations of ethane, propane, butane and pentane, respectively, an additional measurement of the intensities radiation J ₀₁ ^* , ..., J ₀₉ ^* and J ₁ ^* , ..., J ₉ ^* before and after introducing radiation into the evacuated measuring cell, respectively, at the above wavelengths λ ₁ -λ ₇ and at two additional wavelengths λ ₈ , which support in the range from 13.66 to 13.76 μm and λ ₉ , which is supported in the range from 10.508 μm to 10.512 μm. Then these intensities are introduced into the computer complex, after which they pump natural gas through the measuring cell and measure the radiation intensities J ₀₁ , ..., J ₀₉ before entering it into the measuring cell and the radiation intensities J ₁ , ..., J ₉ passed through the measuring cell with natural gas at the above wavelengths. The signals corresponding to the intensities J ₀₁ , ..., J ₀₉ , J ₁ , ..., J _{09 are} sent to the computer complex, and the final concentrations of the natural gas components are calculated by the formula (1), where N = 9, in this case, the least squares method with iterative refinement of the roots, using Householder transformations.

Thus, the proposed invention allows for the analysis of the component composition of natural gas mixtures in real time with sensitivities of measuring the proportions of their components at a level no worse than 3 · 10 ^-3 % of the total volume, which is sufficient to conduct a component quantitative analysis of natural gas mixtures of various compositions in accordance with accepted Russian and international standards.

Literature

1. To the operator of the main pipelines: Reference manual / A.V. Gromov, N.E. Guzanov, L.A. Khachikyan, etc. - M .: Nedra, 1987. 176 p.

2. GOST 23781-87. Combustible natural gases. Chromatographic method for determining the composition.

3. GOST 14920-79. The gas is dry. Method for determining the composition.

4. ISO / FDIS-6974-1-ISO / FDIS-6974-6. Natural gas. - Determination of composition with defined uncertainly by gas chromatography.

5. ISO 6569. Natural gas. - Rapid analysis by gas chromatography. International Standards Organization Central Secretariat. 1981.

6. ISO 6568-81. Natural gas. - Simple analysis by gas chromatography. International Standards Organization Central Secretariat. 1981.

7. GOST 31371.1-200 - GOST 31371.6-2008. Natural gas. Determination of composition by gas chromatography with an estimate of uncertainty.

8. ISO 6975: 1997. Natural gas - Extended analysis - Gas-chromatographic method.

9. ISO 6142: 2001. Gas analysis - Preparation of calibration gas mixture - Gravimetric method.

10. ISO 6143: 2001. Gas analysis - Determination of the composition of calibration gas mixture - Comprasion methods.

11. ISO 14111: 1997. Natural gas - Guidelines to traceability in analysis.

12. ISO 10723: 1995. Natural gas - Performance evaluation for on-line analytical systems.

13. Murin V.I., Kislenko H.H., Arystanbekova C.A. and other Gas chromatographic methods for determining the component composition of natural gas. M .: Gazprom. 2001.57 p.

14.V.E. Privalov, V.G. Shemanin. Remote laser sensing of hydrocarbons in the atmosphere. Letters to JETP, 2001, vol. 27, issue 1, pp. 71-75.

15. Mezheris R. Laser remote sensing. - M.: Mir, 1987, 550 p.

16. M.P. Buric, K.P. Chen, J.Falk, S.D. Woodruff. Enhanced spontaneous Raman scattering and gas composition analysis using a photonic crystal fiber. Applied Optics. 2008. Vol. 47. Number 23. P.4255-4261.

17. Bublicenko I.A., Mokhnonozhkin B.E. Laser gas analyzer to search for gas leaks from underground gas pipelines. Instruments and experimental equipment, 1999, No. 5, p.126-129.

18. N. Mahoukhi, E. Pere, R. Creff, C. Pouchan. Determination of the composition of a mixture of gases by infrared analysis and chemometric methods. Journal of Molecular Structure 744-747 (2005) 855-859.

19. Kireev S.V., Shnyrev S.L., Podolyako E.M., Usoshin V.A., Petrov N.G., Esin Yu.I., Klischevskaya V.M. RF patent for the invention No. 2267114 of 12/27/2005.

20. A. Bjorck. BIT. 1967. Vol. 7. p. 322-337.

21. A. Bjorck. BIT. 1968. Vol. 8. p. 8-30.

22. J. R. Bunch, and B. N. Parlett. SIAM J. of Numerical Analysis. 1979. Vol.16. P.368-375.

23. O.B. Bartenev. Fortran for professionals. M .: Dialog-MEPhI. 2000.448 s.

Table 1 No. Share of methane,% Ethane share,% The proportion of propane,% one 78 fifteen 7 2 79 5 16 3 84.9 fifteen 0.1 four 85.8 13.5 0.7 5 85.9 0.1 fourteen 6 86.5 11.5 2 7 87 10 3 8 87.4 0.6 12 9 87.5 6.5 6 10 88 6 6 eleven 88.3 1,2 10.5 12 88.5 2.5 9 13 88.5 four 7.5 fourteen 88.5 5.5 6 fifteen 88.5 7 4,5 16 90 3 7 17 90.5 2 7.5 eighteen 93 2 5

table 2 No. Combinations of wavelengths of measurement, microns Measurement error, % methane ethane propane one 7.68; 6.88; 6.90 0.9-1.0 eleven 12 0.9 19 49 2 7.68; 12.24; 6.90 0.9 1,0 2.2 0.9 17 45 3 7.68; 12.24; 6.90; 6.88 0.9-0.95 1,1 2.0 0.9 17 38

Table 3 Additional wavelengths, microns The sensitivity of the detection of butane (fraction in the mixture,%) The measurement error of the remaining components,% methane ethane propane CO ₂ 6.93; 10.51 5.1 · 10 ^-4 1,2 1,2 1.8 <1% 6.93; 13.49 8.7 · 10 ^-4 1,1 1.3 1,5 <1% 10.51; 13.49 2.6 · 10 ^-4 1,2 1.3 1,6 <1% 6.93; 10.51; 13.49 3.6 · 10 ^-4 1,1 1,2 1,6 <1%

Table 4 Additional wavelengths, microns Pentane detection sensitivity (fraction in the mixture,%.) The measurement error of the remaining components,% Methane Ethane Propane CO ₂ 6.91 2.6 · 10 ^-1 <1% 1,2 1,6 <1% 10.52 1.1 · 10 ^-2 <1% 1,2 1,6 <1% 13.71 1.9 · 10 ^-2 <1% 1,2 1,6 <1% 6.91; 10.52 3.6 · 10 ^-4 <1% 1,1 1,5 <1% 6.91; 13.71 5.4 · 10 ^-4 <1% 1,1 1.4 <1% 10.52; 13.71 2.2 · 10 ^-4 <1% <1% 1.4 <1% 6.91; 10.52; 13.71 2.5 · 10 ^-4 <1% <1% 1.3 <1%

Table 5 The ratio of butane: pentane Sensitivity (share in the mixture,%) for the Medvezhye, Jura deposit Sensitivity (fraction in the mixture,%) for the Vuktylskoye field butane pentane butane pentane 1: 4 1.4 · 10 ^-3 6.0 · 10 ^-4 4.8 · 10 ^-4 2.1 · 10 ^-4 1: 1 8.4 · 10 ^-4 7.5 · 10 ^-4 2.5 · 10 ^-4 2.3 · 10 ^-4 4: 1 7.5 · 10 ^-4 1.3 · 10 ^-3 2.4 · 10 ^-4 4.2 · 10 ^-5

Table 6 Deposit (methane: ethane: propane: butane: pentane) Detection Sensitivity,% ethane propane butane pentane Berezovskoe (94.8: 1.2: 0.3: 0.1: 0.1) 3.6 · 10 ^-5 2.1 · 10 ^-4 2.3 · 10 ^-4 2.3 · 10 ^-4 Komsomolskoe, Jura (95.4: 2.1: 0.5: 0.1: 0.1) 2.2 · 10 ^-5 1.6 · 10 ^-4 2.4 · 10 ^-4 2.3 · 10 ^-4 Bovanenkovskoe deposit (96.4: 2.89: 0.03: 0.02: 0.01) 2.1 · 10 ^-5 1.3 · 10 ^-3 9.6 · 10 ^-4 1.1 · 10 ^-3 Mesoyakhskoe (98.2: 0.1: 0.03: 0.01: 0.01) 9.3 · 10 ^-5 1.3 · 10 ^-3 1.2 · 10 ^-3 1.1 · 10 ^-3 Tazovsky, Cenomanian (98.6: 0.1: 0.05: 0.03: 0.02) 1.1 · 10 ^-4 8.1 · 10 ^-4 6.6 · 10 ^-4 7.2 · 10 ^-4

Claims (1)

A method for determining the component composition of natural gas in real time, which consists in measuring the absorption of radiation by the analyzed gas, characterized in that the preliminary measurement of the absorption of radiation is carried out at seven wavelengths, while the wavelength λ ₁ support in the range from 7.6 to 7.8 μm, λ ₂ support in the range from 6.898 to 6.902 μm, λ ₃ support in the range from 6.87 to 6.89 μm, λ ₄ support in the range from 12.1 to 12.3 μm, λ ₅ support in the range from 15 , 2 to 15.4 μm, λ ₆ support in the range from 6.92 to 6.94 μm and λ ₇ support in the range from 6.908 to 6.912 μm, while for the first category of natural gas for which the methane fraction is equal to or greater than 95%, and for the second category of natural gas for which the methane fraction is less than 95%, the ratio n _e + n _pr > n _b + n _pent , where n _e , n _pr , n _b , n _pent are the concentrations of ethane, propane, butane and pentane in the analyzed gas, respectively, the final measurement of radiation intensities at eight wavelengths is carried out, where λ ₁ -λ ₇ are indicated above, and the wavelength λ ₈ support in the range from 10.518 to 10.522 μm for the first category at native gas and in the range from 13.66 to 13.76 μm for the second category of natural gas, and for the third category of natural gas, for which the methane fraction is less than 95%, and the ratio n _e + n _pr > n _b + n _pent the final measurement of the radiation intensities at nine wavelengths is carried out, where λ ₁ -λ ₇ are indicated above, the wavelength λ _{8 is} maintained in the range from 13.66 to 13.76 μm, and the wavelength λ _{9 is} maintained in the range from 10.508 to 10.512 μm, at the same time, at the first stage of measurements, part of the light flux from the light source is sent to meters to control the values Intensity at the entrance to the measuring cell, both measured radiation intensity J ₀₁ ^*, ..., J ₀₇ ^* before putting it into an evacuated measuring cell and the radiation intensity J ₀₁ ^*, ..., J ₀₇ ^* passed through an evacuated measuring cell, at wavelengths above waves λ ₁ -λ ₇ , and enter the values of these intensities in the computer complex, then start pumping through the measuring cell of natural gas and measure the radiation intensity J ₀₁ , ..., J ₀₇ before entering it into the measuring cell and radiation intensity J ₁ , ..., J ₇ , passed through a measuring cell with natural gas at the above seven wavelengths λ ₁ -λ ₇ , send signals corresponding to the intensities J ₀₁ , ..., J ₀₇ , J ₁ , ..., J ₇ to the computer complex , and calculate the concentration of natural gas components according to the formula

where N is the number of wavelengths at which measurements are made,
n _m , n _e , n _pr , n _b , n _pent , n _co2 are the concentrations of methane, ethane, propane, butane, pentane and carbon dioxide in the analyzed natural gas, respectively,
L is the length of the measuring cell,
σ _m (λ _i ), σ _e (λ _i ), σ _pr (λ _i ), σ _b (λ _i ), σ _pent (λ _i ), σ _co2 (λ _i ) (i = 1 ... N) - sections absorption of methane, ethane, propane, butane, pentane and carbon dioxide, respectively, at wavelengths λ ₁ , ..., λ _N ,
k _i (i = 1 ... N) are the coefficients equal to the ratio of the intensities J _i ^* / J ₀₁ ^* , where J _i ^* are the intensities of radiation at the corresponding wavelengths λ _i passing through the evacuated measuring cell, and J _0i ^* are the intensities of radiation at the appropriate wavelengths λ _i before entering it into the evacuated measuring cell,
J _0i (i = 1 ... N) - radiation intensity at the corresponding wavelengths λ _i before entering them into the measuring cell with natural gas,
J _i (i = 1 ... N) - radiation intensity at the corresponding wavelengths λ _i passed through the measuring cell with natural gas,
then, for the first and second categories of natural gas, an additional measurement of radiation intensities J ₀₁ ^* , ..., J ₀₈ ^* and J ₁ ^* , ..., J ₈ ^* is carried out before and after the radiation is introduced into the evacuated measuring cell, respectively, at the above eight wavelengths λ ₁ -λ ₈ , and for the third category of natural gas, an additional measurement of radiation intensities J _0i ^* , ..., J ₀₉ ^* and J ₁ ^* , ..., J ₉ ^* is carried out before and after the radiation is introduced into the evacuated measuring cell, respectively, at the nine wavelengths indicated above λ ₁ -λ ₉ , then these intensities are introduced into the computer complex, after which pumping of natural gas through the measuring cell is started and radiation intensities J ₀₁ , ..., J _{0N are measured} before it is introduced into the measuring cell and radiation intensities J ₁ , ..., J _N passed through the measuring cell with natural gas, where N = 8 for the first and second categories of natural gas, N = 9 for the third category of natural gas, at the above wavelengths λ ₁ -λ ₈ for the first and second categories of natural gas or λ ₁ -λ ₉ for the third category of natural gas signals, respectively which compute the intensities J ₀₁ , ..., J _0N , J ₁ , ..., J _N into the computer system, and calculate the final concentrations of natural gas components according to formula (1).

Images