CN117828397B - Tunnel rock mineral identification method and system with cooperative multi-element spectrum - Google Patents

Abstract

The present invention proposes a tunnel rock and mineral identification method and system based on coordinated multivariate spectroscopy, which relates to the technical field of rock and mineral qualitative and quantitative identification. It includes collecting rock samples in the tunnel site area, obtaining accurate mineral identification results, establishing a rock spectrum library and an end-member mineral spectrum library in the tunnel site area; obtaining tunnel rock spectrum data; using the overall waveform of the fused spectrum data to perform a preliminary match with the data in the rock spectrum library in the tunnel site area; using the characteristic peak band position and characteristic peak characteristics of the rock spectrum to perform a fine identification of the mineral type based on the end-member mineral spectrum library; establishing an association between spectral variation characteristics and mineral types, and performing auxiliary identification of the tunnel rock mineral types that cannot be determined by fine identification, and completing the hierarchical qualitative identification of the tunnel rock minerals; selecting strongly correlated bands, and continuously iterating the unmixing model to achieve accurate quantitative inversion of minerals, and completing quantitative identification. The present invention realizes the qualitative and quantitative identification of minerals based on coordinated multivariate spectroscopy.

Description

Tunnel rock and mineral identification method and system based on collaborative multivariate spectroscopy

Technical Field

The present invention belongs to the technical field of qualitative and quantitative identification of rocks and minerals in tunnels, and in particular to a method and system for identifying rocks and minerals in tunnels using a coordinated multivariate spectrum.

Background technique

The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

In recent years, my country's tunnel construction has gradually moved to areas with complex geological conditions. The inventors have found that complex geological conditions often lead to the development of unfavorable geological conditions such as faults, karst, and alteration zones. If they are not identified in time, they are likely to cause geological disasters, such as sudden water and mud, landslides, and large deformation of surrounding rocks, which seriously affect the safety and progress of on-site construction.

The mineral anomalies of tunnel surrounding rocks are often used as signs of faults and other unfavorable geological conditions, and can identify the nature of unfavorable geology, and have been widely used in related fields. However, existing mineral testing methods mainly include microscopic thin section method, X-ray diffraction, infrared spectroscopy and other methods. Microscopic thin section and X-ray diffraction methods require complex sample preparation (grinding or slicing), and microscopic thin section testing has high environmental requirements. Dust and darkness in the tunnel greatly affect the accuracy of the identification results. Therefore, the above methods cannot meet the requirements of rapid on-site construction.

Infrared spectroscopy has fast testing speed and strong environmental adaptability, and can obtain spectral information of surrounding rocks in the harsh environment of tunnels. However, the existing spectral interpretation methods have low accuracy and poor efficiency. They cannot accurately identify mineral types and invert mineral content through the spectral characteristics of surrounding rocks, making it difficult to provide technical guidance for actual engineering construction.

Summary of the invention

In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a tunnel rock mineral identification method and system based on collaborative multivariate spectroscopy, which realizes the qualitative and quantitative identification of minerals based on collaborative multivariate spectroscopy, and is of great significance for improving the efficiency and accuracy of tunnel surrounding rock mineral identification and ensuring rapid and safe tunnel construction.

To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

The first aspect of the present invention provides a tunnel rock mineral identification method based on collaborative multivariate spectroscopy.

The tunnel rock mineral identification method based on collaborative multivariate spectroscopy includes the following steps:

Collect rock samples in the tunnel site area, obtain accurate mineral identification results, establish a rock spectral library and an end-member mineral spectral library in the tunnel site area, and obtain the spectral variation characteristics of different minerals mixed in the rock;

Acquire tunnel rock spectral data along the tunnel excavation direction, fuse the spectral data of the two bands, and use the overall waveform of the fused spectral data to make a preliminary match with the data in the rock spectral library of the tunnel site to determine the approximate mineral composition of the tunnel rock;

Using the characteristic peak band position of rock spectra and the fine and combined characteristics of characteristic peaks, the mineral types are precisely identified based on the end-member mineral spectral library.

Establish the association between spectral variation characteristics and mineral types, assist in identifying the types of tunnel rock minerals that cannot be determined through fine identification, and complete the hierarchical qualitative identification of tunnel rock minerals;

The tunnel rock spectrum is unmixed, and the pure spectrum of the end-member mineral is used for reverse inference. The strongly correlated bands are selected and the unmixing is iterated again using the strongly correlated bands until the number of strongly correlated bands remains unchanged, thus completing the quantitative identification of the minerals.

The second aspect of the present invention provides a tunnel rock mineral identification system based on collaborative multivariate spectroscopy.

The tunnel rock and mineral identification system based on collaborative multi-spectral analysis includes:

The spectrum library establishment module is configured to: collect rock samples in the tunnel site area, obtain accurate mineral identification results, establish a rock spectrum library and an end-member mineral spectrum library in the tunnel site area, and obtain the spectrum variation characteristics of different minerals mixed in the rock;

The qualitative identification module is configured to: obtain tunnel rock spectral data along the tunnel excavation direction, fuse the spectral data of the two bands, use the overall waveform of the fused spectral data to perform preliminary matching with the data in the rock spectral library of the tunnel site area, and determine the approximate mineral composition of the tunnel rock;

Using the characteristic peak band position of rock spectra and the fine and combined characteristics of characteristic peaks, the mineral types are precisely identified based on the end-member mineral spectral library.

Establish the association between spectral variation characteristics and mineral types, assist in identifying the types of tunnel rock minerals that cannot be determined through fine identification, and complete the hierarchical qualitative identification of tunnel rock minerals;

The quantitative identification module is configured to: unmix the tunnel rock spectrum, reversely infer using the pure spectrum of the end-member minerals, select the strongly correlated bands, and re-iterate the unmixing using the strongly correlated bands until the number of strongly correlated bands remains unchanged, thus completing the quantitative identification of the minerals.

The third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the collaborative multivariate spectroscopy tunnel rock mineral identification method as described in the first aspect of the present invention.

The fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the collaborative multivariate spectroscopy tunnel rock mineral identification method as described in the first aspect of the present invention are implemented.

One or more of the above technical solutions have the following beneficial effects:

(1) The present invention provides a tunnel rock mineral identification method and system based on collaborative multi-spectral analysis. By combining the visible light-near infrared band and the thermal infrared band, full coverage of mineral identification types is achieved, and multi-spectral collaborative qualitative and quantitative identification of minerals is realized. It provides more abundant mineral anomaly signs for poor geological identification, improves the accuracy of poor geological identification, improves the efficiency of mineral identification in tunnels, and eliminates the influence of subjective factors on the poor geological identification results through quantitative data.

(2) The present invention fully exploits the spectral characteristics of the surrounding rock and combines multiple characteristics in depth for mineral identification. The spectral criteria for mineral qualitative identification are progressive and mutually verified. The mineral spectral knowledge is fully integrated into the identification process. The variability of the spectral characteristics of mixed minerals is taken into account, which effectively avoids the multi-solution of the spectrum and improves the solvability and accuracy of rock spectral data.

(3) The quantitative method proposed in the present invention is based on reverse thinking, and the reasoning process has a solid theoretical foundation. It reduces the subjectivity of band selection through quantitative errors, ensures the objective accuracy of the model, and achieves the optimal unmixing effect through continuous iteration of unmixed strongly correlated bands, thus realizing the rapid quantitative identification of minerals in the tunnel and the transition from qualitative to quantitative identification of adverse geological conditions.

Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

FIG. 1 is a flow chart of a method according to a first embodiment.

Figure 2 is a characteristic diagram of the variation of the thermal infrared band after mixing albite and kaolinite.

Figure 3 is a map of measuring points on the tunnel face using the ground feature spectrometer and the Fourier transform infrared spectrometer.

Figure 4 is the spectrum measured data of the tunnel surrounding rock.

Figure 5 is a band error diagram of the measured rock spectrum and the calculated rock spectrum.

Detailed ways

It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

The overall idea proposed by the present invention is:

The present invention provides a tunnel rock mineral identification method and system based on collaborative multivariate spectroscopy. The method provides geological constraints for the qualitative interpretation of surrounding rock spectra through preliminary geological surveys, makes full use of the characteristic information of the spectrum, selects indicators such as "overall spectrum waveform-characteristic peak absorption band-characteristic peak fine characteristics-characteristic peak combination characteristics-mixed spectrum variation characteristics", forms a hierarchical mineral qualitative method to realize mineral type identification, and then realizes accurate quantitative inversion of minerals by selecting strongly correlated bands and continuously iterating the unmixing model, thereby providing mineral anomaly marks for poor geological identification.

Embodiment 1

This embodiment discloses a tunnel rock and mineral identification method using collaborative multivariate spectroscopy.

As shown in FIG1 , the tunnel rock mineral identification method based on collaborative multivariate spectroscopy includes:

(1) Conduct preliminary geological surveys to fully investigate the geological characteristics of the tunnel site and provide geological information constraints for qualitative identification of minerals;

(2) Collect rock samples from the tunnel site, use precise mineral identification methods in the laboratory to obtain accurate mineral composition information, and perform spectral testing using a ground feature spectrometer and a Fourier transform infrared spectrometer to establish a rock spectral library for the tunnel site. Based on the rock mineral identification results, an end-member mineral spectral library is established;

(3) Using the rock mineral information, spectral data and spectral library obtained in step (2), the spectral variation characteristics of different minerals mixed in the rock are explored to assist in identifying the mineral types;

(4) Using a ground object spectrometer and a Fourier transform infrared spectrometer, a large number of tests were carried out along the tunnel excavation direction to obtain spectral data of the tunnel rock;

(5) Preprocess the acquired rock spectral data to improve the data quality, highlight the spectral features, and improve the accuracy of spectral analysis;

(6) Performing qualitative identification of minerals on the acquired data: a) First, the spectral data of the two bands are fused, and the overall waveform of the spectrum is used to match the rock spectrum library with known rock mineral components established in advance to perform preliminary identification of the mineral symbiosis; b) Then, based on other characteristics of the rock spectrum, such as "characteristic peak band position (wavelength at the peak) → fine characteristics of the characteristic peak (characteristic peak shape, characteristic peak symmetry, whether the characteristic peak is double-peaked or multi-peaked, etc.) → combination characteristics of the characteristic peak (combination form of multiple characteristic peaks)", the mineral type is finely identified; c) For minerals that are difficult to identify based on the above characteristics, auxiliary identification is performed based on the knowledge of mineral mixing variation rules discovered in step (3) (the characteristic peak band shifts or the characteristic peak shape changes after a certain mineral is mixed), so that the atypical spectral characteristics after variation (the shifted characteristic peak position or the characteristic peak shape, etc.) are related to the mineral type for auxiliary identification;

(7) After obtaining the mineral composition information of the tunnel surrounding rock through step (6), quantitative identification of minerals is performed: a) first unmixing is performed to preliminarily obtain the contribution ratio of each mineral spectrum in the mixed spectrum, that is, the mineral content; b) using the pure spectrum of the end-member mineral, according to the content after unmixing in step a), reverse inference is performed according to the mixed spectrum unmixing rules to calculate and obtain the rock spectrum, which is called the calculated rock spectrum; c) solving the error value of each band of the measured rock spectrum and the calculated rock spectrum; d) according to the error size, the band that meets the error range is selected as the strong correlation band; e) the strong correlation band is used again for unmixing; f) the above method is iterated continuously until the number of strong correlation bands remains unchanged.

More specifically, they include:

(1) Conduct preliminary geological survey along a tunnel to investigate the geological information on the tunnel surface and inside the tunnel to provide geological information constraints for qualitative identification of minerals;

The geological information mainly includes the lithology of the ground strata along the tunnel, the types of geological actions that occurred in the past, the location of unfavorable geology, etc.

The methods for conducting preliminary geological surveys include longitudinal drilling, surface sampling, airborne electromagnetic geophysical exploration, etc.;

Among them, an investigation was conducted along the tunnel, and it was found that the strata along the tunnel were granite strata, and no major geological tectonic movements had occurred in history. However, there was a small-scale fault about 200m in front of the tunnel entrance, where clay minerals were produced.

(2) Collect rock samples from the tunnel site, use precise mineral identification methods in the laboratory to obtain accurate mineral composition information, and perform spectral testing using a ground feature spectrometer and a Fourier transform infrared spectrometer to establish a rock spectral library for the tunnel site. Based on the rock mineral identification results, an end-member mineral spectral library is established;

Collect demonstration samples from the tunnel site, i.e., collect and number the cores, rock blocks and other demonstrative samples collected in the geological survey in step (1);

Use accurate laboratory mineral identification methods, including but not limited to thin section microscopy, X-ray diffraction and other methods;

When testing with a ground object spectrometer and a Fourier transform infrared spectrometer, the sample surface must be flat to prevent light leakage and other possible factors that may reduce data quality.

Establish an end-member spectral library, that is, after obtaining accurate mineral identification results, collect the spectra of these minerals in the visible light-near infrared and thermal infrared bands to establish an end-member mineral spectral library at the tunnel site;

Methods for establishing an end-member mineral spectral library include collecting from existing spectral libraries such as the USGS spectral library, the JHU spectral library, and the ASU spectral library, or purchasing pure minerals and then collecting them with a spectrometer to establish an end-member mineral spectral library.

In this embodiment, the samples collected by geological survey are numbered according to mileage, coordinates and other information, and the minerals are identified by X-ray diffraction. The results show that most rocks contain minerals such as potassium feldspar, albite, muscovite, kaolinite, illite, etc., and the above pure minerals are spectrally tested using two spectral techniques to establish an end-member mineral spectral library.

(3) Using the rock mineral information, spectral data and spectral library obtained in step (2), the spectral variation characteristics of different minerals mixed in the rock are explored to assist in identifying the mineral types;

Mineral mixture variation characteristics refer to the phenomenon that after multiple minerals are mixed, their mixed spectral characteristics will be offset, annihilated, and a series of other phenomena that are inconsistent with the spectral characteristics of pure minerals;

Exploring the spectral variation characteristics of different mineral mixtures plays a positive role in establishing a connection between the atypical characteristics of mineral mixtures and mineral types.

In this embodiment, taking the mixture of albite and kaolinite as an example, there is no obvious variation phenomenon in the near-infrared band. In the thermal infrared band, the characteristic peak of albite is originally located at 9616nm, and it shifts to the short-wave direction after mixing, and the remaining characteristics are obliterated. Kaolinite can be confirmed from the characteristics at 9899nm and 10950nm/10972nm, as shown in Figure 2.

(4) Using a ground object spectrometer and a Fourier transform infrared spectrometer, a large number of tests were carried out along the tunnel excavation direction to obtain spectral data of the tunnel rock;

A ground feature spectrometer and a Fourier transform infrared spectrometer are used to test along the tunnel excavation direction. During the test, the instrument probe needs to be in full contact with the flat surface of the rock or surrounding rock and pressed tightly to prevent light leakage and other phenomena that affect data accuracy.

In this embodiment, a comprehensive test is performed on the tunnel wall and the tunnel face along the tunnel by manual labor or by placing the instrument on a robot. The test points should be representative points, and the test range covers the tunnel face, as shown in FIG3 .

(5) Preprocess the acquired rock spectral data to improve the data quality, highlight the spectral features, and improve the accuracy of spectral analysis;

Preprocessing the acquired rock spectrum data, wherein the preprocessing methods include but are not limited to multivariate scattering correction, differentiation, smoothing, envelope removal and other methods;

In this embodiment, after step (4) obtains the full-band spectral data of the surrounding rock, SG smoothing is first performed to remove noise, and then multivariate dispersion correction is used to eliminate the influence of volume scattering. At the same time, the envelope of the near-infrared data is removed, and the baseline correction is performed on the thermal infrared data to highlight its absorption and reflection characteristics.

(6) Qualitative identification of minerals on the acquired data:

a) First, the spectral data of the two bands are fused, and the overall waveform of the spectrum is used to match the rock spectrum library with known rock mineral components established in the early stage to make a preliminary identification of the mineral paragenesis combination;

b) Then, according to other characteristics of the rock spectrum, such as "characteristic peak band position (wavelength at the peak) → fine characteristics of characteristic peaks (characteristic peak shape, characteristic peak symmetry, whether the characteristic peak is double-peaked or multi-peaked, etc.) → combination characteristics of characteristic peaks (combination form of multiple characteristic peaks)", the mineral type is finely identified;

c) For minerals that are difficult to identify using the above characteristics, auxiliary identification is performed using the knowledge of the mineral mixing variation rules explored in step (3) (such as the shift of the characteristic peak spectral band or the change of the characteristic peak morphology after a certain mineral is mixed), so that a relationship is established between the atypical spectral characteristics after variation (such as the shifted characteristic peak position or the characteristic peak morphology) and the mineral type for auxiliary identification.

In step a), the spectral data of the two bands are fused, and the fusion methods include but are not limited to cumulative fusion (CF), equal weight fusion (ERF), outer product fusion (OPF) and the like.

In this example, we:

a) The rock spectral data of the two bands after preprocessing were fused by equal weight fusion method. Compared with other methods, the absorption and reflection features were not missed while reducing the data dimension. The fused data were preliminarily matched with the established rock spectral library of the tunnel site to determine the approximate mineral combination. It was preliminarily judged to be a mixture of potassium feldspar, kaolinite and montmorillonite. The spectral data is shown in Figure 4;

b) First, comprehensive identification is performed through the characteristic peak absorption position and the end member spectrum library. The 9480nm reflection peak can be used to determine the presence of potassium feldspar, while the 10972nm reflection peak cannot identify kaolinite and montmorillonite. The reflection characteristic positions of the three mineral spectra in the thermal infrared band are shown in Table 1;

Table 1: Reflection characteristic positions of the three mineral spectra in the thermal infrared band

The asymmetric double peaks near 2200nm in the near-infrared spectrum indicate the presence of kaolinite. The absence of slope-shaped absorption features at 1400 and 1900nm and absorption features at 2210nm, combined with the absence of the characteristic reflection peak position of montmorillonite in the thermal infrared band, indicates that montmorillonite does not exist in the rock.

c) Through the variation law, the reflection peak at 9881nm is confirmed to be kaolinite, verifying the identification results in steps a) and b).

(7) After obtaining the mineral composition information of the tunnel surrounding rock through step (6), quantitative identification of minerals is performed:

a) firstly perform unmixing to preliminarily obtain the contribution ratio of each mineral spectrum in the mixed spectrum, that is, the content of the mineral;

b) using the pure spectrum of the end-member minerals, according to the content after unmixing in step a), reverse inference is performed according to the rules of mixed spectrum unmixing, and the rock spectrum is calculated, which is called calculated rock spectrum;

c) Solve the measured rock spectrum and calculate the error value of each band of the rock spectrum;

d) According to the error size, select the band that meets the error range as the strong correlation band;

e) Reusing strongly correlated bands for unmixing;

f) The above method is iterated continuously until the number of strongly correlated bands remains unchanged.

The unmixing methods of steps a) and e) include linear unmixing and nonlinear unmixing methods.

In this embodiment:

We determined through (6) that the rock contains two minerals: potassium feldspar and kaolinite;

Following step a) in (7), spectra of two minerals, potassium feldspar and kaolinite, are selected from the end-member mineral spectrum library, and the rock spectra are linearly unmixed to obtain the relative contents of the two minerals;

b) The spectra of the two end-member minerals and their relative contents are linearly superimposed according to the following formula to obtain the calculated spectrum of the rock: where W represents the rock spectrum, a _n represents the content of the nth end-member mineral, and f _n represents the pure spectrum of the nth end-member mineral.

W＝ _{a1f1 + a2f2 +} ......+a _n f _n

(c) After obtaining the calculated spectrum of the rock, the error value of each band between it and the measured spectrum of the rock is calculated (Figure 5);

(d) Considering the error value of each band, we found that the error between 8um and 12um is relatively high, so it is not suitable as a strong correlation band. We initially selected 5um to 8um as a strong correlation band;

e) Linearly unmixing the spectral data in the 5um to 8um band to obtain the content of each end-member mineral again;

(f) The above process is iterated continuously until the number of selected correlation bands remains unchanged.

Embodiment 2

This embodiment discloses a tunnel rock and mineral identification system based on collaborative multivariate spectroscopy.

The tunnel rock and mineral identification system based on collaborative multi-spectral analysis includes:

The spectrum library establishment module is configured to: collect rock samples in the tunnel site area, obtain accurate mineral identification results, establish a rock spectrum library and an end-member mineral spectrum library in the tunnel site area, and obtain the spectrum variation characteristics of different minerals mixed in the rock;

The qualitative identification module is configured to: obtain tunnel rock spectral data along the tunnel excavation direction, fuse the spectral data of the two bands, use the overall waveform of the fused spectral data to perform preliminary matching with the data in the rock spectral library of the tunnel site area, and determine the approximate mineral composition of the tunnel rock;

Using the characteristic peak band position of rock spectra and the fine and combined characteristics of characteristic peaks, the mineral types are precisely identified based on the end-member mineral spectral library.

Establish the association between spectral variation characteristics and mineral types, assist in identifying the types of tunnel rock minerals that cannot be determined through fine identification, and complete the hierarchical qualitative identification of tunnel rock minerals;

The quantitative identification module is configured to: unmix the tunnel rock spectrum, reversely infer using the pure spectrum of the end-member minerals, select the strongly correlated bands, and re-iterate the unmixing using the strongly correlated bands until the number of strongly correlated bands remains unchanged, thus completing the quantitative identification of the minerals.

Embodiment 3

The purpose of this embodiment is to provide a computer-readable storage medium.

A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the collaborative multivariate spectroscopy tunnel rock mineral identification method as described in Example 1 of the present disclosure.

Embodiment 4

The purpose of this embodiment is to provide an electronic device.

An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the collaborative multivariate spectroscopy tunnel rock mineral identification method as described in Example 1 of the present disclosure are implemented.

The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method of the present invention.

Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Technical personnel in the relevant field should understand that various modifications or variations that can be made by technical personnel in the field without creative work on the basis of the technical solution of the present invention are still within the scope of protection of the present invention.

Claims (10)

1. A tunnel rock mineral identification method based on collaborative multivariate spectroscopy, characterized in that it comprises the following steps: Collect rock samples in the tunnel site area, obtain accurate mineral identification results, establish a rock spectral library and an end-member mineral spectral library in the tunnel site area, and obtain the spectral variation characteristics of different minerals mixed in the rock; Acquire tunnel rock spectral data along the tunnel excavation direction, fuse the spectral data of the two bands, and use the overall waveform of the fused spectral data to make a preliminary match with the data in the rock spectral library of the tunnel site to determine the approximate mineral composition of the tunnel rock; The fusion of the spectral data of the two bands is to achieve full coverage of mineral identification types by combining the visible light-near infrared band and the thermal infrared band; Using the characteristic peak band position of rock spectra and the fine and combined characteristics of characteristic peaks, the mineral types are precisely identified based on the end-member mineral spectral library. Establish the association between spectral variation characteristics and mineral types, assist in identifying the types of tunnel rock minerals that cannot be determined through fine identification, and complete the hierarchical qualitative identification of tunnel rock minerals; The tunnel rock spectrum is unmixed, and the pure spectrum of the end-member mineral is used for reverse inference. The strongly correlated bands are selected and the unmixing is iterated again using the strongly correlated bands until the number of strongly correlated bands remains unchanged, thus completing the quantitative identification of the minerals. 2. The tunnel rock and mineral identification method based on collaborative multivariate spectroscopy as claimed in claim 1, characterized in that: The rock samples obtained from the tunnel site are identified using precise mineral identification methods in the laboratory to obtain precise mineral identification results; Use ground object spectrometer and Fourier transform infrared spectrometer to carry out spectrum test on rock samples in tunnel site, and establish rock spectrum library in tunnel site; Collect the spectral data of pure minerals from the accurate mineral identification results from the existing spectral library, or establish an end-member mineral spectral library by purchasing pure minerals and collecting spectral data with a spectrometer. 3. The tunnel rock mineral identification method based on collaborative multivariate spectroscopy as claimed in claim 1 is characterized in that the tunnel rock spectral data is obtained along the tunnel excavation direction, specifically: Comprehensive tests are conducted on the tunnel walls and faces along the tunnel, representative points are selected as test points, the test range covers the face, and tunnel rock spectral data is obtained. 4. The collaborative multivariate spectroscopy tunnel rock mineral identification method as described in claim 1 is characterized in that the characteristic peak fine features include characteristic peak morphology, characteristic peak symmetry, whether the characteristic peak is bimodal or multimodal, and the characteristic peak combination features are the combination form of multiple characteristic peaks. 5. The tunnel rock mineral identification method based on collaborative multivariate spectroscopy as claimed in claim 1 is characterized in that the quantitative identification of minerals is specifically: a) Unmix the rock spectra and obtain the preliminary mineral content; b) Use the end-member mineral spectral library to obtain the pure spectrum of the end-member minerals, and based on the obtained mineral content, perform reverse inference according to the rules of mixed spectrum unmixing to obtain the calculated rock spectrum; c) Solve the measured rock spectrum and calculate the error value of each band of the rock spectrum; d) According to the error size, select the band that meets the error range as the strong correlation band; e) Use strongly correlated bands for unmixing to obtain new mineral content; f) Iterate the above steps b) - e) until the number of strongly correlated bands remains unchanged. 6. The tunnel rock mineral identification method based on collaborative multivariate spectroscopy as described in claim 5 is characterized in that the unmixing method includes a linear unmixing method and a nonlinear unmixing method. 7. The tunnel rock and mineral identification method based on collaborative multivariate spectroscopy as claimed in claim 1, characterized in that before collecting rock samples in the tunnel site area, it also includes: Conduct preliminary geological surveys along the tunnel to investigate the geological information on the tunnel surface and inside the tunnel, and provide geological information constraints for qualitative identification of minerals. 8. A collaborative multi-spectral tunnel rock and mineral identification system, characterized in that: when executed, the collaborative multi-spectral tunnel rock and mineral identification method according to any one of claims 1 to 7 is implemented, comprising: The spectrum library establishment module is configured to: collect rock samples in the tunnel site area, obtain accurate mineral identification results, establish a rock spectrum library and an end-member mineral spectrum library in the tunnel site area, and obtain the spectrum variation characteristics of different minerals mixed in the rock; The qualitative identification module is configured to: obtain tunnel rock spectral data along the tunnel excavation direction, fuse the spectral data of the two bands, use the overall waveform of the fused spectral data to perform a preliminary match with the data in the rock spectral library of the tunnel site area, and determine the approximate mineral composition of the tunnel rock; Using the characteristic peak band positions, fine features and combination features of the rock spectra, the mineral types are precisely identified based on the end-member mineral spectral library. Establish the association between spectral variation characteristics and mineral types, assist in identifying the types of tunnel rock minerals that cannot be determined through fine identification, and complete the hierarchical qualitative identification of tunnel rock minerals; The quantitative identification module is configured to: unmix the tunnel rock spectrum, reversely infer using the pure spectrum of the end-member minerals, select the strongly correlated bands, and re-iterate the unmixing using the strongly correlated bands until the number of strongly correlated bands remains unchanged, thus completing the quantitative identification of the minerals. 9. A computer-readable storage medium having a program stored thereon, wherein when the program is executed by a processor, the program implements the steps of the collaborative multivariate spectroscopy tunnel rock mineral identification method as described in any one of claims 1 to 7. 10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the collaborative multivariate spectroscopy tunnel rock mineral identification method as described in any one of claims 1 to 7 are implemented.