Raman analysis method of mixed gas content and ethylene cracking gas detection method and system

Abstract

The invention provides a Raman analysis method of mixed gas content and an ethylene cracking gas detection method and system, and belongs to the technical field of gas mixture online detection.The analysis method comprises the steps that mixed gas is split into n key components, and reference components are determined; performing spectral signal processing on the original Raman spectrum to obtain a normalized spectrum, and determining the relative Raman peak heights of the n key components; based on the relative Raman peak heights and the regression coefficients of the n-1 key components, determining the molar ratios of the n-1 key components relative to the reference component; the molar fractions of the n key components are determined based on the molar ratios of the n-1 key components relative to the reference component. The detection method comprises the following steps: taking an ethylene independent Raman peak near 1345cm <-1 > as a reference peak, and determining the content of each component of the ethylene cracking gas by adopting the analysis method. According to the method provided by the invention, the complex ethylene cracking gas composition can be accurately and rapidly detected.

Description

Technical Field

[0001] The present invention relates to the technical field of on-line detection of gas mixtures, and in particular, to a Raman analysis method for the content of a gas mixture, an on-line detection method for ethylene pyrolysis gas based on Raman spectroscopy, an on-line detection system for ethylene pyrolysis gas based on Raman spectroscopy, a Raman analysis device, and a computer-readable storage medium. Background Art

[0002] As a production device for important basic organic chemical raw materials such as ethylene, propylene, and butadiene, the material analysis of an ethylene pyrolysis furnace is very important. The outlet gas-phase material (pyrolysis gas) of the ethylene pyrolysis furnace contains light components such as H2, CH4, C2H6, C2H4, C3H8, C3H6, and C4H8. Measuring the content of these components in the pyrolysis gas can understand the pyrolysis depth and the distribution of pyrolysis products of the pyrolysis furnace, and help guide the optimization of process parameters.

[0003] Gas chromatography (GC) is one of the commonly used methods for gas composition analysis. For example, for the composition analysis of the pyrolysis gas of an ethylene pyrolysis furnace, a multi-dimensional gas chromatograph is mainly used. Through a switching system of multiple columns and multiple valves, and by using various techniques such as cutting, backflushing, and capillary columns, different components in the sample gas are separated and detected on different paths and different detectors (for example, a flame ionization detector FID is used to detect the hydrocarbon content in the gas, and a thermal conductivity detector TCD is used to detect the H2 content and N2 content in the gas), so as to obtain the full-component analysis results of H2 to toluene in the pyrolysis gas. Its advantages are comprehensive analysis results and a large amount of information, but the interval between two injections for full-component analysis (that is, the time required for sample analysis and instrument preparation and then re-injection) is about 20 minutes, which is time-consuming. In addition, on-line chromatographic analysis also has deficiencies such as a complex system, many auxiliary devices, and a large amount of on-site maintenance work, which limit its wide application.

[0004] Due to the increase in the production capacity of the pyrolysis furnace and the diversity of pyrolysis raw materials, maximizing production efficiency requires faster detection speeds for specific production indicators such as pyrolysis depth, so as to adjust process parameters and optimize product distribution in a timely manner. In recent years, spectral analysis methods such as near-infrared and Raman have been increasingly used for the rapid analysis of pyrolysis furnace raw materials and products. Laser Raman spectroscopy technology is a molecular structure characterization technology based on the Raman scattering effect, which can achieve non-destructive and rapid analysis of samples. Moreover, the spectral analysis system has a simple structure, low maintenance costs, and is easy to operate, and has been widely used in the fields of oil refining, chemical engineering, and pharmaceuticals.

[0005] However, in the prior art, when applying laser Raman spectroscopy to analyze gas composition, there are always problems such as low gas molecule density, small scattering cross section, and weak Raman spectrum. In particular, for the cracked gas actually produced by the ethylene cracking reaction, its pressure is close to normal pressure, resulting in a significant decrease in the Raman signal, and it is impossible to ensure the detection and analysis accuracy of the Raman spectrum for the components of the ethylene cracked gas. For example, a Chinese non-patent document entitled Raman spectroscopy analysis of ethylene cracked gas published in 2016 discloses a Raman spectroscopy analysis method for ethylene cracked gas, which obtains the Raman spectra of 6 key components in the mixed sample of ethylene cracked gas by detecting the laser Raman gas experimental prototype, thereby analyzing and obtaining the contents of the 6 key components. However, due to severe external interference during the detection process, this method has good analysis accuracy for H2, C2H4, and C3H6 with strong Raman signals, but the content detection error of alkane components such as CH4 is large.

[0006] In order to enhance the intensity of Raman signals, researchers at home and abroad have proposed various solutions, such as multiple reflections of excitation light, increasing laser power, and artificially pressurizing sample gas. For ethylene cracking gas, the multiple reflection method relies on the reflection efficiency of the reflector, which is easily contaminated by the sample gas. Although artificially pressurizing the sample gas can significantly increase the intensity of Raman signals, some components in the cracking gas (such as butadiene) may be partially hydraulically compressed, resulting in inconsistency between the Raman analysis object and the actual object.

[0007] In addition to the Raman signal intensity, the existing Raman quantitative analysis model is very cumbersome in the analysis process of the components of complex mixed gases (such as ethylene cracking gas), and it is impossible to achieve accurate and rapid detection. For example, the commonly used spectral quantitative analysis method is a chemometric method based on a training sample library, that is, first, a large number of typical ethylene cracking gas samples with different compositions need to be collected, and the corresponding sample gas composition is obtained by gas chromatography analysis, thereby obtaining a training sample library; then, based on the sample library, partial least squares (PLS) and other algorithms are used to establish a functional relationship between the cracking gas composition and the spectrum, thereby predicting the unknown mixed gas sample. Although this method is simple in calculation, the modeling workload is large; moreover, when the cracking gas composition changes significantly in the future, it is necessary to collect samples for model calibration, which is very time-consuming.

[0008] In summary, the existing Raman spectroscopy analysis systems and methods have technical defects such as weak Raman signal intensity, slow analysis speed, and low detection accuracy, and are unable to quickly and accurately detect the complex components of ethylene cracking gas. Summary of the invention

[0009] In view of the technical problems of weak Raman signal intensity, slow analysis speed and low detection accuracy in the existing Raman spectroscopy analysis system and method, the present invention provides a Raman analysis method for the content of mixed gas. By using this method, it is possible to quickly perform quantitative analysis on the composition of complex mixed gas (such as ethylene cracking gas), and improve the analysis speed and analysis accuracy of gas Raman spectroscopy detection.

[0010] To achieve the above object, the first aspect of the present invention provides a Raman analysis method for the content of mixed gas. The analysis method includes the following steps: splitting the mixed gas into n key components, and selecting one component from the n key components as the reference component; performing spectral signal processing on the original Raman spectrum of the mixed gas to obtain the normalized spectrum of the mixed gas; determining the relative Raman peak heights of the n key components based on the normalized spectrum of the mixed gas; determining the molar ratios of the n - 1 key components relative to the reference component based on the relative Raman peak heights of the n - 1 key components and the regression coefficients; and determining the molar fractions of the n key components respectively based on the molar ratios of the n - 1 key components relative to the reference component.

[0011] In an exemplary embodiment of the present invention, the performing spectral signal processing on the original Raman spectrum of the mixed gas to obtain the normalized spectrum of the mixed gas may include: performing background subtraction and baseline correction on the original Raman spectrum of the mixed gas to obtain the corrected Raman spectrum; and performing normalization processing on the corrected Raman spectrum with the independent characteristic peak of the reference component as the spectral normalization reference peak to obtain the normalized spectrum of the mixed gas.

[0012] In an exemplary embodiment of the present invention, the determining the relative Raman peak heights of the n key components based on the normalized spectrum of the mixed gas may include: combining the Raman spectra of pure component sample gases to determine the relatively independent characteristic spectral peaks of the n key components of the mixed gas; and determining the relative Raman peak heights of the n key components of the mixed gas in the normalized spectrum of the mixed gas according to the relatively independent characteristic spectral peaks of the n key components of the mixed gas.

[0013] In an exemplary embodiment of the present invention, the molar ratios of the n - 1 key components relative to the reference component can be calculated by the following formula:

[0014] z j (t) = k j0 + k j1 P j (t), (j = 2, …, n)

[0015] where z j (t) is the molar ratio of the jth component of the mixed gas relative to the reference component, P j (t) is the relative Raman peak height of the jth component in the mixed gas, k j0is the first regression coefficient of the j-th component of the mixed gas, k j1 is the second regression coefficient of the j-th component of the mixed gas.

[0016] In an exemplary embodiment of the present invention, the analysis method may further include: determining the first and second regression coefficients of n-1 key components of the mixed gas according to the training sample set of the mixed gas; wherein, the training sample set of the mixed gas includes: the normalized spectrum of the mixed gas and the molar ratio or volume concentration ratio of n-1 key components of the mixed gas relative to the reference component.

[0017] In an exemplary embodiment of the present invention, the determining the first and second regression coefficients of n-1 key components of the mixed gas according to the training sample set of the mixed gas may include: determining the molar ratio of n-1 key components of the mixed gas relative to the reference component according to the training sample set of the mixed gas; determining the relative Raman peak height of n-1 key components according to the normalized spectrum of the mixed gas and the characteristic peak positions of n-1 key components; and respectively determining the first and second regression coefficients of n-1 key components of the mixed gas by combining the statistical regression model according to the molar ratio of n-1 key components relative to the reference component and the relative Raman peak height of n-1 key components.

[0018] In an exemplary embodiment of the present invention, the molar fraction of the n components can be calculated by the following formula:

[0019]

[0020] c1 + c2 + … + c n = 1;

[0021] wherein, c j is the molar fraction of the j-th component in the mixed gas, z j is the molar ratio of the j-th component in the mixed gas relative to the reference component, and j = 1, 2, …… n.

[0022] The second aspect of the present invention provides an on-line detection method for ethylene cracking gas based on Raman spectroscopy. The detection method may include the following steps: detecting ethylene cracking gas by a Raman spectrometer to obtain the original Raman spectrum of the ethylene cracking gas; determining the molar fraction of each key component of the ethylene cracking gas by using the Raman analysis method for the content of the mixed gas as described above according to the original Raman spectrum of the ethylene cracking gas; wherein, the key components of the ethylene cracking gas include: hydrogen, methane, ethane, ethylene, acetylene, propane, propylene, propyne and other composite components, the other composite components are C4 olefins or butadiene, and the reference component of the ethylene cracking gas is ethylene.

[0023] In another exemplary embodiment of the present invention, the detection method may further include: at 1345 cm-1 The nearby ethylene independent Raman peak is used as the spectral normalization reference peak.

[0024] In another exemplary embodiment of the present invention, the detection method may further include: before detecting the ethylene cracking gas, configuring the laser power of the Raman spectrometer to be greater than or equal to 1 W, the line width to be less than or equal to 0.1 nm, and the spectral integration time to be set to 30 - 90 s.

[0025] The third aspect of the present invention provides a Raman analysis device, which includes a processor and a memory. At least one computer program is stored in the memory, and the at least one computer program is loaded and executed by one or more of the above processors, so that the processor executes the Raman analysis method for the mixed gas content as described above.

[0026] The fourth aspect of the present invention provides a computer-readable storage medium, which stores at least one program code. The program code is loaded and executed by a processor, so that a computer executes the Raman analysis method for the mixed gas content as described above.

[0027] The fifth aspect of the present invention provides an on-line detection system for ethylene cracking gas based on Raman spectroscopy. The detection system includes a sampling unit, a Raman spectrometer, and the Raman analysis device as described above; one end of the sampling unit is connected to the outlet pipeline of the ethylene cracking furnace, the other end of the sampling unit is connected to the Raman spectrometer, the Raman spectrometer is used to detect the sample gas extracted by the sampling unit to obtain the original Raman spectrum of the ethylene cracking gas, and the Raman analysis device is connected to the Raman spectrometer to analyze the original Raman spectrum of the ethylene cracking gas to obtain the mole fractions of the key components of the ethylene cracking gas.

[0028] Through the technical solution provided by the present invention, the present invention has at least the following technical effects:

[0029] (1) The Raman analysis method for the mixed gas content provided by the present invention forms a new gas Raman quantitative analysis model. Through this analysis model, not only can the composition of any mixed gas (such as ethylene cracking gas, carbon two hydrogenation mixed gas, etc.) be quantitatively analyzed, but also high analysis accuracy and fast analysis speed can be guaranteed;

[0030] (2) The Raman analysis method for the mixed gas content provided by the present invention directly determines the content of each component based on the relative Raman characteristic peak heights of each component. The model structure is simple and has strong adaptability, and can be applied to the on-line real-time analysis of the composition of cracking gas under different reaction feeds;

[0031] (3) In view of the characteristics of the ethylene pyrolysis gas, such as having many components and a wide content range, the on-line detection method for ethylene pyrolysis gas provided by the present invention divides the ethylene pyrolysis gas into several key components, and accurately calculates the mole fractions of each component of the ethylene pyrolysis gas by combining the above-mentioned gas Raman quantitative analysis model, thus making up for the deficiency of the existing Raman spectroscopic analysis method for ethylene pyrolysis gas, which has a large detection error for the content of alkane components such as CH4;

[0032] (4) The on-line detection method for ethylene pyrolysis gas provided by the present invention introduces background subtraction and baseline correction techniques, and uses the independent characteristic peak of a certain reference component as a benchmark to normalize the spectrum, which greatly improves the repeatability of the spectrum and significantly reduces the influence of the pressure fluctuation of the pyrolysis gas and the change of spectral detection conditions (such as laser power) on the spectrum;

[0033] (5) In view of the problem that the Raman signal intensity of some components in the ethylene pyrolysis gas is weak, the on-line detection method for ethylene pyrolysis gas provided by the present invention adopts direct Raman signal enhancement measures from the hardware perspective for the gas Raman spectrometer, including: increasing the laser power, increasing the collection angle of Raman scattered light, reducing the optical signal transmission loss, improving the detection sensitivity of the spectrometer (while reducing the detection noise level), appropriately increasing the integration time of the spectrometer, etc. The introduction of these measures can obtain a more satisfactory Raman spectral signal of the pyrolysis gas under process conditions;

[0034] (6) The on-line detection method for ethylene pyrolysis gas provided by the present invention can realize the on-line detection of the content of key components of ethylene pyrolysis gas, such as H2, CH4, C2H4, C3H6, etc., which can replace the traditional on-line chromatographic detection method to improve the detection speed and detection accuracy of the composition of ethylene pyrolysis gas; at the same time, combined with advanced control and real-time operation optimization techniques, it can significantly improve the operation level and economic benefits of the device.

[0035] Other features and advantages of the present invention will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification, and are used to explain the embodiments of the present invention together with the following specific implementation manners, but do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0037] Figure 1 It is a flow chart of the Raman analysis method for the content of the mixed gas provided by the embodiment of the present invention;

[0038] Figure 2 It is a schematic structural diagram of the Raman analysis device provided by the embodiment of the present invention;

[0039] Figure 3Hardware architecture diagram of the on-line detection system for ethylene pyrolysis gas based on Raman spectroscopy provided by the embodiment of the present invention;

[0040] Figure 4 Schematic diagram of the working principle of the on-line Raman spectrometer provided by the embodiment of the present invention;

[0041] Figure 5 Original Raman spectrum of ethylene pyrolysis gas provided by the embodiment of the present invention;

[0042] Figure 6 Raman spectrum of pyrolysis gas after signal correction provided by the embodiment of the present invention;

[0043] Figure 7 Normalized Raman spectrum of pyrolysis gas provided by the embodiment of the present invention;

[0044] Figure 8A Chromatographic analysis value and Raman estimation value of the molar ratio of H2 component to ethylene provided by the embodiment of the present invention;

[0045] Figure 8B Chromatographic analysis value and Raman estimation value of the molar ratio of methane component to ethylene provided by the embodiment of the present invention;

[0046] Figure 9 On-line Raman analysis result of key components of pyrolysis gas provided by the embodiment of the present invention.

[0047] Description of reference numerals

[0048] 201 - Processor, 202 - Memory. Detailed description of the specific implementation mode

[0049] The following details the specific implementation mode of the embodiment of the present invention with reference to the accompanying drawings. It should be understood that the specific implementation mode described herein is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0050] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0051] In the description of the present invention, it should also be noted that, unless otherwise clearly defined and limited, terms such as "installation" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, an indirect connection; it can be a wired connection, or a wireless connection. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0052] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0053] Example 1

[0054] Please refer to Figure 1 , the first embodiment of the present invention provides a Raman analysis method for the content of a mixed gas. The Raman analysis method for the content of the mixed gas includes the following steps:

[0055] Step S101: Split the mixed gas into n key components, and select one component from the n key components as a reference component.

[0056] Step S102: Perform spectral signal processing on the original Raman spectrum of the mixed gas to obtain the normalized spectrum of the mixed gas.

[0057] Here, it should be noted that in order to further improve the signal-to-noise ratio of the spectrum, for the original Raman spectrum of the cracked gas detected in real time by the Raman spectrometer, background subtraction and baseline correction are performed, and the original Raman spectrum is normalized with the independent characteristic peak of a certain reference component as the benchmark.

[0058] Specifically, in the embodiment of the present invention, the process of performing spectral signal processing on the original Raman spectrum of the mixed gas to obtain the normalized spectrum of the mixed gas may include but is not limited to the following sub-steps S1021 to sub-step S1022.

[0059] Sub-step S1021: Perform background subtraction and baseline correction on the original Raman spectrum of the mixed gas to obtain the corrected Raman spectrum.

[0060] Sub-step S1022: Use the independent characteristic peak of the reference component as the spectral normalization reference peak to perform normalization processing on the corrected Raman spectrum to obtain the normalized spectrum of the mixed gas.

[0061] Step S103: Based on the normalized spectrum of the mixed gas, determine the relative Raman peak heights of the n key components.

[0062] Specifically, in the embodiment of the present invention, the process of determining the relative Raman peak heights of the n key components based on the normalized spectrum of the mixed gas may include but is not limited to the following sub-steps S1031 to sub-step S1032.

[0063] Sub-step S1031: Combine the Raman spectra of pure component sample gases (with N2 as the balance gas) to determine the relatively independent characteristic spectral peaks of the n key components of the mixed gas.

[0064] It should be noted that in the above description, the "Raman spectrum of the pure component sample gas" refers to the Raman spectrum of each pure component gas measured separately. In actual analysis, by separately measuring the Raman spectra of each main component gas in the analysis sample, relatively independent characteristic spectral peaks of each component gas can be obtained. These characteristic spectral peaks refer to the unique Raman scattering peaks of each gas, which correspond to specific vibration modes of the molecules.

[0065] Sub-step S1032: Determine the relative Raman peak heights of the n key components in the mixed gas in the normalized spectrum of the mixed gas according to the relatively independent characteristic spectral peaks of the n key components of the mixed gas.

[0066] Step S104: Determine the molar ratios of the n - 1 key components relative to the reference component based on the relative Raman peak heights of the n - 1 key components and the regression coefficients.

[0067] Specifically, in the embodiments of the present invention, if it is assumed that the molar fractions of the components of the mixed gas are c j (j = 1,..., n), and the Raman peak heights of the components of the mixed gas are P j (j = 1,..., n), according to the Raman spectroscopy principle, for different components in the same mixture, the ratio of the Raman peak height of the jth component relative to the reference component should be proportional to the molar ratio of these two components, and is independent of the detection conditions and the sample gas pressure. Thus, the following formula can be obtained:

[0068]

[0069] In formula (1), P j (t) is the relative Raman peak height of the jth component in the mixed gas, P1(t) is the relative Raman peak height of the reference component in the mixed gas, K j is the Raman intensity factor of the jth component in the mixed gas, K1 is the Raman intensity factor of the reference component in the mixed gas (usually assuming that the Raman intensity factor of the reference component is 1), c j (t) is the molar fraction of the jth component in the mixed gas, and c1(t) is the molar fraction of the reference component in the mixed gas.

[0070] For the normalized spectrum of any sample, let the characteristic peak height (i.e., the relative Raman peak height) P1(t) of the reference component be 1. From formula (1), it can be obtained:

[0071]

[0072] To improve the calculation accuracy of the analysis model, formula (2) above can be further expanded as:

[0073] z j (t) = k j0 + k j1 P j(t), (j = 2, …, n) (3)

[0074] where z j (t) is the molar ratio of the j-th component of the mixed gas to the reference component, and P j (t) is the relative Raman peak height of the j-th component in the mixed gas, and k j0 is the first regression coefficient of the j-th component of the mixed gas, and k j1 is the second regression coefficient of the j-th component of the mixed gas.

[0075] That is to say, the molar ratios of the n - 1 key components to the reference component can be obtained by calculating with the above formula (3). It should be noted that if the composition of the mixed gas is relatively simple (only composed of several components, and the characteristics between the components are obvious), then the molar ratios of the key components of the mixed gas to the reference component can also be obtained by calculating with the above formula (2).

[0076] Step S105: Based on the molar ratios of the n - 1 key components to the reference component, determine the molar fractions of the n key components respectively.

[0077] Specifically, in the embodiments of the present invention, for any mixed gas with unknown composition, the sum of the molar ratios of the above n key components is 1, that is:

[0078] c1 + c2 + … + c n = 1 (4)

[0079] Among them, z1 = 1, so formula (4) can be further expanded to:

[0080] c1(z1 + z2 + … + z n ) = 1 (5)

[0081] And the molar fractions of the components of the mixed gas are:

[0082]

[0083] In the formula, c j is the molar fraction of the j-th component in the mixed gas, z j is the molar ratio of the j-th component in the mixed gas to the reference component, and z i is the molar ratio of the i-th component in the mixed gas to the reference component.

[0084] That is to say, the molar fractions of the n components can be obtained by calculating with the above formula (6).

[0085] It should be noted that the above formulas (3) and (6) can jointly constitute the gas Raman quantitative analysis model of the present invention. For any mixed gas composed of the above n components, its original spectrum can be detected by an on-line Raman spectrometer, and then its normalized spectrum can be obtained through spectral signal processing, thereby obtaining the Raman peak heights at the characteristic peak positions of each component. Once the Raman peak heights and regression coefficients at the characteristic peak positions of the n components are determined and substituted into the gas Raman quantitative analysis model, the molar concentrations of the components in the mixed gas can be obtained, so as to realize the quantitative analysis of any mixed gas composed of these components.

[0086] Furthermore, in a possible implementation manner, in step S104, before determining the molar ratios of the n - 1 key components relative to the reference component, it is necessary to determine the first and second regression coefficients of the n - 1 key components of the mixed gas according to the training sample set of the mixed gas. The training sample set of the mixed gas includes: the normalized spectra of several groups of mixed gases and the molar ratios (or volume concentration ratios) of the n - 1 key components of the mixed gas relative to the reference component.

[0087] Specifically, the process of determining the regression coefficients of the n - 1 key components of the mixed gas can be achieved through the following methods:

[0088] (1) According to the training sample set of the mixed gas, determine the molar ratios of the n - 1 key components of the mixed gas relative to the reference component;

[0089] (2) According to the normalized spectrum of the mixed gas and the characteristic peak positions of the n - 1 key components, determine the relative Raman peak heights of the n - 1 key components;

[0090] (3) According to the molar ratios of the n - 1 key components relative to the reference component and the relative Raman peak heights of the n - 1 key components, respectively determine the first and second regression coefficients of the n - 1 key components of the mixed gas in combination with the statistical regression model.

[0091] In addition, the implementation environment of this embodiment includes at least one terminal and a server, and this method is executed on the terminal or the server respectively. The terminal and the server can be communicatively connected to realize the interactive transmission of information.

[0092] Among them, the terminal can be any electronic product that can perform human-computer interaction with the user through one or more methods such as a keyboard, a touchpad, a touch screen, and voice interaction, such as a PC (Personal Computer), a PPC (Pocket Personal Computer), a tablet computer, etc.

[0093] The server can be a single server, a server cluster composed of multiple servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0094] Embodiment 2

[0095] The second embodiment of the present invention further provides a Raman analysis device. Refer to Figure 2 , the Raman analysis device includes a processor 201 and a memory 202. At least one computer program is stored in the memory, and the at least one computer program is loaded and executed by one or more of the above-mentioned processors so that the processor implements the Raman analysis method for the mixed gas content as described above.

[0096] Of course, the Raman analysis device may also have components such as a wired or wireless network interface, a keyboard, and an input / output interface for input and output. The Raman analysis device may further include other components for implementing various functions of the device, which will not be elaborated here.

[0097] The second embodiment of the present invention further provides a computer-readable storage medium. At least one program code is stored in the computer-readable storage medium, and the program code is loaded and executed by a processor so that the computer implements the Raman analysis method for the mixed gas content as described above.

[0098] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical disc data storage device, etc. Those skilled in the art can understand that all or part of the steps of implementing the method in the above embodiments can be completed by a program instructing relevant hardware. The program is stored in a storage medium, including several instructions for causing a single-chip microcomputer, a chip, or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage media include: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, etc., which can store program codes.

[0099] Embodiment 3

[0100] The third embodiment of the present invention provides an on-line detection system for ethylene cracking gas based on Raman spectroscopy. The on-line detection system for ethylene cracking gas includes: an on-line Raman spectroscopy analyzer and a remote monitoring PC.

[0101] As Figure 3 shown, one end of the on-line Raman spectroscopy analyzer is connected to an on-line sampling device through a dedicated Raman optical cable to obtain a sample to be measured and perform Raman spectroscopy detection and analysis. The other end of the on-line Raman spectroscopy analyzer is connected to the remote monitoring PC through a communication optical cable to obtain the Raman analysis results of each component of the ethylene cracking gas.

[0102] The on-line sampling device (such as Figure 3 the 1# on-site sampling unit in

[0103] is directly installed on the evaluation device and consists of a cracking gas sampler, a cracking gas Raman detection unit, etc. Since the material flow rate of the evaluation device is small and it is an intermittent reaction device, after its own quench and three-stage water-cooled separation and recovery system, it can already meet the subsequent on-line analysis requirements. Therefore, the on-line Raman sampler of the evaluation device is installed after the three-stage water cooling, and the material temperature is close to room temperature. The sampler passes the complex cracking gas through a filter reflux section and a cooling and dehydration section, leaving solid particles, heavy components and moisture in the process pipeline, taking out the light components, and sending them to the Raman detection unit uniformly at a certain flow rate. The purified sample gas mainly consists of H2, C1-C4 hydrocarbons. The sampling cell is connected to the external sample tube by a ferrule. The Raman detection probe connected to the sampling cell is connected to the on-line Raman spectroscopy analyzer by a dedicated Raman optical cable. Figure 4 shown, the monochromatic excitation light emitted by the laser is irradiated on the sample to be measured in the sampling tube (also called the flow cell) through a dedicated optical fiber and the Raman probe. The excited Raman scattered light is collected by the probe and transmitted to the spectrometer through a dedicated optical fiber for spectral splitting and analog-to-digital conversion. Finally, the Raman spectral data is transmitted to the computer for preprocessing and analysis model calculation to obtain the corresponding composition content or quality index of the sample to be measured.

[0104] That is to say, the main body of the on-line Raman spectroscopy analyzer adopts a positive pressure explosion-proof form, and its internal key components include: a 532nm laser, an optical fiber spectrometer, a programmable logic controller and an embedded PC. The embedded PC is configured with control software for each optical component, Raman spectroscopy detection and preprocessing software, and quantitative analysis software for the composition of the mixed gas.

[0105] The third embodiment of the present invention provides an on-line detection method for ethylene cracking gas based on Raman spectroscopy. The on-line detection method for ethylene cracking gas may include the following steps:

[0106] Step S301: Use a Raman spectrometer to detect ethylene cracking gas and obtain the original Raman spectrum of the ethylene cracking gas.

[0107] Here, it should be noted that in view of the technical defect that the Raman spectrometer has good analysis accuracy for H2, C2H4, and C3H6 with strong Raman signals when detecting ethylene cracking gas, but has a large detection error for the content of alkane components such as CH4, before the detection of the on-line detection method for ethylene cracking gas in the embodiments of the present invention, some direct Raman signal enhancement measures are implemented on the Raman spectrometer from a hardware perspective, including: increasing the laser power, increasing the collection angle of Raman scattered light, reducing the optical signal transmission loss (such as adopting integrated molding by fusion), improving the detection sensitivity of the spectrometer while reducing the detection noise level (such as using CCD cooling), and appropriately increasing the integration time of the spectrometer (such as setting the spectrometer integration time to 30 - 90 s), etc. The introduction of these measures helps to obtain a relatively satisfactory Raman spectral signal of ethylene cracking gas under process conditions.

[0108] For example, in order to ensure that the Raman spectrometer meets the needs of on-line gas detection and ensure the stability of operation under on-line conditions, the spectral range of the selected Raman spectrometer should include 400 - 3100 cm -1 , to ensure coverage of the Raman characteristic spectral regions of the key components of ethylene cracking gas, and the spectral resolution should be ≤ 6 cm -1 , to ensure the measurement accuracy of the spectrum.

[0109] The Raman spectrometer should be equipped with a high-power 532 nm semiconductor pumped laser, the laser power should be above 1 W, and the line width should be ≤ 0.1 nm, to meet the actual application needs of low pressure of ethylene cracking gas and high detection accuracy requirements.

[0110] To reduce the noise level of the CCD detector and improve the signal-to-noise ratio at the same time, a high-end fiber optic spectrometer is selected, and its detector is a back-illuminated CCD detector array with TEC cooling, and the cooling temperature is ≤ -15 °C.

[0111] Based on the above hardware selection and parameter design, both the Raman spectral detection signal is improved and the noise level is reduced, and at the same time, future engineering implementation and application are very convenient.

[0112] Step S302: Perform spectral signal processing on the original Raman spectrum of ethylene cracking gas to obtain the normalized spectrum of ethylene cracking gas.

[0113] Here, it should be noted that ethylene cracking gas may contain many single-component substances (up to hundreds), and the embodiments of the present invention mainly focus on the following 9 components, namely: H2, methane, ethane, ethylene, acetylene, propane, propylene, propyne, and other composite components (mainly C4 olefins or butadiene), and focus on the on-line detection of the content of H2, methane, ethylene, and propylene components. For each ethylene cracking gas, it can be considered that the sum of the contents of the above 9 components is approximately 100%.

[0114] In addition, other composite components are virtual composite components. The purpose of introducing these components is to meet the molar fraction normalization condition. The reference component of the ethylene pyrolysis gas is ethylene.

[0115] For an actual ethylene pyrolysis unit, the original Raman spectrum of the pyrolysis gas detected in real time by an on-line Raman spectrometer is as Figure 5 shown. The spectral sampling interval is 1 min. The spectrum after background subtraction and baseline correction is as Figure 6 shown. Among them, P0, P1, and P2 are the characteristic peaks of N2, ethylene, and H2 respectively. The Raman spectrum obtained after normalization with the P1 peak as the reference peak is as Figure 7 shown.

[0116] Combined with Figure 5 、 Figure 6 and Figure 7 it can be seen that for the original Raman spectrum of the pyrolysis gas detected in real time by the on-line Raman spectrometer, background subtraction and baseline correction techniques are introduced, and the spectrum is normalized with the independent characteristic peak of a certain reference component as the benchmark, which greatly improves the repeatability of the spectrum and significantly reduces the influence of the pyrolysis gas pressure fluctuation and the change of spectral detection conditions (such as laser power) on the spectrum.

[0117] Step S303: Determine the relative Raman peak heights of n key components according to the normalized spectrum of the ethylene pyrolysis gas.

[0118] Step S304: Substitute the Raman peak heights and regression coefficients at the peak positions of the characteristic peaks of the n components of the ethylene pyrolysis gas into the gas Raman quantitative analysis model to calculate the molar concentrations of the components of the ethylene pyrolysis gas.

[0119] To establish and solve the Raman quantitative analysis models described by Equations (3) and (6), the core problems include: (1) selecting the characteristic peak positions of the components of the ethylene pyrolysis gas; (2) estimating the regression coefficients of each component.

[0120] For the above core problem (1), since the ethylene pyrolysis gas has the characteristics of many detected components and a wide content range, in order to improve the detection accuracy, for the ethylene pyrolysis gas, it is advisable to select ethylene with a relatively large content as the reference component, and select an independent Raman peak near 1345 cm -1 as the spectral normalization reference peak.

[0121] Taking the ethylene pyrolysis gas in a certain process pipeline as an example, the characteristic peak positions of its components are shown in Table 1.

[0122] Table 1: Characteristic peaks of the main components of the pyrolysis gas

[0123] Component <![CDATA[H2]]> <![CDATA[CH4]]> <![CDATA[C2H6]]> <![CDATA[C2H4]]> <![CDATA[C2H2]]> <![CDATA[C3H8]]> <![CDATA[C3H6]]> Propyne Others Characteristic peak position 586 2921 2904 1345 1975 866 919 2142 1655

[0124] For the above core problem (2), before applying the gas Raman quantitative analysis model, it is necessary to estimate the first regression coefficient k of each component by using the statistical regression method based on a number of training samples with known Raman spectra and molar ratios. j0 and the second regression coefficient k j1 .

[0125] The following takes components H2 and methane as examples to illustrate how to determine the regression coefficients of each component based on a number of training samples.

[0126] Based on 27 training samples collected manually under different times, different feeds and different reaction conditions, the volume concentration ratio of component H2 relative to the ethylene component can be obtained; and the relative Raman peak height of component H2 can be obtained from the normalized spectrum and the characteristic peak position of H2 (the relative Raman peak height of the ethylene component in each sample is 1). Combining this with the least squares algorithm, the corresponding model regression coefficients can be estimated as k 11 = 1.008, k 10 = 0.012.

[0127] Similarly, from the training sample set, the volume concentration ratio of component methane relative to the ethylene component can be obtained; and from the corresponding Raman normalized spectrum, the relative Raman peak height of the characteristic peak position of the component can be obtained. Combining this with the least squares algorithm, the corresponding model regression coefficients can be estimated as k 21 = 0.5473, k 20 = 0.0126.

[0128] As Figure 8A shows, the chromatographic analysis value of the relative volume concentration of component H2 and the corresponding Raman regression value; as Figure 8B shows, the chromatographic analysis value of the relative volume concentration of component methane and the corresponding Raman regression value. Combining Figure 8A and Figure 8B it can be seen that the Raman analysis models of the relative volume concentrations of these two components have high estimation accuracy.

[0129] Similarly, the characteristic peaks of other components and the corresponding model regression coefficients can be obtained, as shown in Table 2 specifically.

[0130] Table 2: Characteristic peaks and model coefficients of the main components of cracked gas

[0131] Component <![CDATA[H2]]> <![CDATA[CH4]]> <![CDATA[C2H6]]> <![CDATA[C2H4]]> <![CDATA[C2H2]]> <![CDATA[C3H8]]> <![CDATA[C3H6]]> Propyne Others Characteristic peak position 586 2921 2904 1345 1975 866 919 2142 1655 Characteristic peak gain 1.008 0.5473 0.6327 1.0 0.6163 0.5077 2.364 0.4557 1.0291 Intercept 0.012 0.0126 -0.0123 0 -0.0003 0.0043 -0.020 0.0027 0.1086

[0132] For any normalized spectrum of cracked gas, by combining the characteristic peak positions of each component, the effective Raman characteristic peak height of each component can be calculated. Substituting the effective Raman characteristic peak height and regression coefficient of each component into the Raman quantitative analysis model, the volume percentage of each component can be calculated.

[0133] Taking an experiment of a device evaluated in the ethylene cracking mode as an example, the on-line detection and analysis results of the key components of the cracked gas are as Figure 9 shown.

[0134] The preferred embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various simple modifications can be made to the technical solutions of the present invention, and these simple modifications all fall within the protection scope of the present invention.

[0135] In addition, it should be noted that, in the case of no contradiction, the various specific technical features described in the above specific embodiments can be combined in any appropriate manner. To avoid unnecessary repetition, the present invention will not separately describe various possible combination manners.

[0136] Furthermore, any combination can be made between various different embodiments of the present invention, as long as it does not violate the idea of the present invention, and it should also be regarded as the content disclosed by the present invention.

Claims

1. A Raman analysis method for the content of a mixed gas, characterized in that, The analysis method includes: Splitting the mixed gas into n key components and selecting one component from the n key components as the reference component; Performing spectral signal processing on the original Raman spectrum of the mixed gas to obtain the normalized spectrum of the mixed gas; Based on the normalized spectrum of the mixed gas, determining the relative Raman peak heights of the n key components; Based on the relative Raman peak heights and regression coefficients of n - 1 key components, determining the molar ratios of the n - 1 key components relative to the reference component; Based on the molar ratios of the n - 1 key components relative to the reference component, respectively determining the molar fractions of the n key components.

2. The Raman analysis method for the content of the mixed gas according to claim 1, characterized in that, The performing spectral signal processing on the original Raman spectrum of the mixed gas to obtain the normalized spectrum of the mixed gas includes: Performing background subtraction and baseline correction on the original Raman spectrum of the mixed gas to obtain the corrected Raman spectrum; Using the independent characteristic peak of the reference component as the spectral normalization reference peak to perform normalization processing on the corrected Raman spectrum to obtain the normalized spectrum of the mixed gas.

3. The Raman analysis method for the content of the mixed gas according to claim 1, characterized in that, The determining the relative Raman peak heights of the n key components based on the normalized spectrum of the mixed gas includes: Combining the Raman spectra of the pure component sample gases to determine the relatively independent characteristic spectral peaks of the n key components of the mixed gas; According to the relatively independent characteristic spectral peaks of the n key components of the mixed gas, determining the relative Raman peak heights of the n key components of the mixed gas in the normalized spectrum of the mixed gas.

4. The Raman analysis method for the content of the mixed gas according to claim 1, characterized in that, The molar ratios of the n - 1 key components relative to the reference component are obtained by calculating with the following formula: z j (t) = k j0 + k j1 P j (t), (j = 2, …, n) where z j (t) is the molar ratio of the j-th component of the mixed gas relative to the reference component, P j (t) is the relative Raman peak height of the j-th component in the mixed gas, k j0 is the first regression coefficient of the j-th component of the mixed gas, k j1 is the second regression coefficient of the j-th component of the mixed gas.

5. The Raman analysis method for the mixed gas content according to claim 4, characterized in that, The analysis method further includes: According to the training sample set of the mixed gas, determining the first and second regression coefficients of the n - 1 key components of the mixed gas; Wherein, the training sample set of the mixed gas includes: the normalized spectrum of the mixed gas and the molar ratio or volume concentration ratio of the n - 1 key components of the mixed gas relative to the reference component.

6. The Raman analysis method for the content of the mixed gas according to claim 5, characterized in that, The determining the first and second regression coefficients of the n - 1 key components of the mixed gas according to the training sample set of the mixed gas includes: According to the training sample set of the mixed gas, determining the molar ratios of the n - 1 key components of the mixed gas relative to the reference component; According to the normalized spectrum of the mixed gas and the characteristic peak positions of the n - 1 key components, determining the relative Raman peak heights of the n - 1 key components; According to the molar ratios of the n - 1 key components relative to the reference component and the relative Raman peak heights of the n - 1 key components, respectively determining the first and second regression coefficients of the n - 1 key components of the mixed gas by combining with the statistical regression model.

7. The Raman analysis method for the content of the mixed gas according to claim 1, characterized in that, The molar fractions of the n components are obtained by calculating with the following formula: c1 + c2 + … + c n = 1; where c j is the mole fraction of the j-th component in the mixed gas, and z j is the molar ratio of the j-th component in the mixed gas relative to the reference component, where j = 1, 2,..., n.

8. An on-line detection method for ethylene cracking gas based on Raman spectroscopy, characterized in that, The detection method includes: Using a Raman spectrometer to detect the ethylene cracking gas and obtaining the original Raman spectrum of the ethylene cracking gas; According to the original Raman spectrum of the ethylene cracking gas, using the Raman analysis method for the content of the mixed gas described in any one of claims 1 to 7 to determine the molar fractions of the key components of the ethylene cracking gas; Wherein, the key components of the ethylene cracking gas include: hydrogen, methane, ethane, ethylene, acetylene, propane, propylene, propyne and other composite components, the other composite components are C4 olefins or butadiene, and the reference component of the ethylene cracking gas is ethylene.

9. The on-line detection method of ethylene cracking gas based on Raman spectroscopy according to claim 8, characterized in that The detection method further includes: using the independent Raman peak of ethylene near 1345 cm -1 as the spectral normalization reference peak.

10. The on-line detection method of ethylene cracking gas based on Raman spectroscopy according to claim 8, characterized in that, The detection method further includes: before detecting the ethylene cracking gas, configuring the laser power of the Raman spectrometer to be greater than or equal to 1 W, configuring the line width to be less than or equal to 0.1 nm, and setting the spectral integration time to 30 - 90 s.

11. A Raman analysis device, characterized in that, The Raman analysis device includes a processor and a memory, and at least one computer program is stored in the memory. The at least one computer program is loaded and executed by one or more of the above processors, so that the processor executes the Raman analysis method for the mixed gas content according to any one of claims 1 - 7.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one program code, and the program code is loaded and executed by a processor, so that a computer executes the Raman analysis method for the mixed gas content according to any one of claims 1 - 7.

13. An on-line detection system for ethylene pyrolysis gas based on Raman spectroscopy, characterized in that, The detection system includes a sampling unit, a Raman spectrometer, and the Raman analysis device according to claim 11; One end of the sampling unit is connected to the outlet pipeline of the ethylene cracking furnace, the other end of the sampling unit is connected to the Raman spectrometer, the Raman spectrometer is used to detect the sample gas extracted by the sampling unit to obtain the original Raman spectrum of the ethylene cracking gas, and the Raman analysis device is connected to the Raman spectrometer and is used to analyze the original Raman spectrum of the ethylene cracking gas to obtain the mole fractions of the key components of the ethylene cracking gas.

Images