A method and system for stage difference calibration of near-infrared spectroscopy equipment based on an external reference plate

Abstract

This invention belongs to the field of spectral equipment calibration technology, specifically providing a method and system for calibration of stage differences in near-infrared spectral equipment based on an external reference plate. The method includes measuring the internal reference plate of the near-infrared spectral equipment to acquire raw spectral data; measuring a standard external reference plate to acquire standard spectral data; establishing a spectral conversion model for the near-infrared spectral equipment based on the raw and standard spectral data; acquiring the raw spectral signal; performing real-time conversion on the raw spectral signal based on the spectral conversion model; and outputting the converted standard spectral signal. By introducing a standard external reference plate as a unified physical benchmark, a conversion model from the internal reference signal to the external benchmark signal is established for each near-infrared spectral equipment, enabling spectral measurements of different near-infrared spectral equipment to be calibrated to the same benchmark.

Description

Technical Field

[0001] This invention belongs to the field of spectral equipment calibration technology, and particularly relates to a stage difference calibration method and system for near-infrared spectral equipment based on an external reference plate. Background Technology

[0002] Near-infrared spectroscopy, with its advantages of being rapid, non-destructive, efficient, requiring no complex sample pretreatment, and capable of simultaneously detecting multiple components, has become a key tool for process analysis and quality control in agriculture, food, pharmaceuticals, and petrochemicals. As this technology expands from laboratory analysis to online detection and large-scale industrial applications, the need to deploy the same analytical model across multiple devices is becoming increasingly urgent. However, in practical applications, even near-infrared spectroscopy devices of the same model and batch can exhibit significant systematic deviations in their measurement results; this phenomenon is known as "inter-device variability," or simply "device difference."

[0003] The root causes of performance discrepancies lie in several aspects. First, the core optical components in the equipment, such as light sources, beam splitters, and detectors, have inherent performance tolerances during manufacturing. For example, different light sources have slight differences in the energy distribution of their emitted spectra, and different detectors have different response efficiencies. Second, during equipment assembly, there are micron-level tolerances in the relative positions and angles of optical elements. These mechanical differences alter the optical path, leading to systematic deviations in the optical signals that ultimately reach the detector. Third, the different operating environments and intensities of each piece of equipment result in varying rates and degrees of decay of the internal optical components. This independent and asynchronous aging further exacerbates the performance divergence between equipment.

[0004] Currently, to ensure the measurement stability of individual near-infrared spectroscopy instruments, a built-in reference plate is commonly used for calibration. The basic procedure is as follows: before and after sample measurement, the instrument measures its built-in reference plate (usually a ceramic plate, Teflon plate, or the inner wall of an integrating sphere). The built-in reference plate is used to correct for light source fluctuations and system drift, obtaining relatively stable reflectance or absorbance data, which improves the measurement repeatability and short-term stability of a single instrument. However, this existing technology has the following drawbacks: the internal reference plates of each instrument differ in material, manufacturing process, initial reflectance, and decay characteristics over time. Using different internal reference plates as calibration references is equivalent to calibrating each instrument to a different reference system. Although the short-term stability of a single instrument is improved, measurement data between instruments cannot be directly compared, and analytical models cannot be directly shared across instruments. When the measurement results of a certain instrument are abnormal or questionable, it is impossible to quickly locate the root cause of the problem through comparison with other instruments or a unified benchmark. Repair and calibration work often relies on troubleshooting one by one, which is inefficient. Summary of the Invention

[0005] To address the problems existing in the prior art, the present invention provides a stage difference calibration method for near-infrared spectroscopy equipment based on an external reference plate, comprising the following steps: Step S1: Measure the internal reference plate of the near-infrared spectroscopy device and acquire raw spectral data; Step S2: Measure the standard external reference plate and collect standard spectral data; Step S3: Based on the original spectral data and standard spectral data, establish the spectral conversion model of the near-infrared spectral device; Step S4: Obtain the original spectral signal, perform real-time conversion on the original spectral signal based on the spectral conversion model, and output the converted standard spectral signal.

[0006] Preferably, step S3, establishing the spectral conversion model, specifically includes: S31: For each wavelength point λ in the spectrum, calculate the conversion slope Slope(λ) and conversion intercept Intercept(λ): , , in, For the raw spectral data, Standard spectral data; S32: Using the conversion slope and conversion intercept as calibration parameters for the near-infrared spectrometer, establish a spectral conversion model: ; in, The converted standard spectral signal, This is the original spectral signal.

[0007] Based on the above scheme, there are multiple near-infrared spectrometers, each with its own calibration parameters. The calibration parameters are associated with the identifiers of the corresponding near-infrared spectrometers and stored in the memory of the near-infrared spectrometers.

[0008] Based on the above scheme, step S4 specifically includes: The calibration parameters of the near-infrared spectrometer are retrieved from its memory. The original spectral signal is converted based on the calibration parameters, and the converted standard spectral signal is output.

[0009] Specifically, the multiple near-infrared spectroscopy devices use the same standard external reference plate, which is a standard white board made of Spectralon material.

[0010] Preferably, the spectral conversion model established in step S3 is a polynomial conversion model, specifically including: S31': For each wavelength point λ in the spectrum, calculate the model coefficients of the polynomial transformation model based on the original spectral data and the standard spectral data: ; in, For the raw spectral data, For standard spectral data, , and To calculate the model coefficients at each wavelength point using the least squares method; S32': The model coefficients , and A polynomial transformation model is established as the calibration parameter for the near-infrared spectrometer: ; in, The converted standard spectral signal, This is the original spectral signal.

[0011] Preferably, the method further includes: Dark current spectral data were collected under conditions of no light. The dark current spectral data is subtracted from the original spectral data and the standard spectral data respectively to obtain the corrected original spectral data and standard spectral data; A spectral conversion model for the near-infrared spectrometer is established based on the corrected original spectral data and standard spectral data.

[0012] Preferably, step S4 further includes: The standard spectral signal obtained from the external reference plate of the measurement standard; The converted standard spectral signal is compared with the standard spectral signal obtained from the external reference plate of the measurement standard to determine whether it meets the expected conditions. If it does, the corresponding near-infrared spectroscopy equipment stage difference calibration is completed.

[0013] On the other hand, the present invention also provides a stage difference calibration system for near-infrared spectroscopy equipment based on an external reference plate, the system comprising: Multiple near-infrared spectrometers of the same model to be calibrated, each with an integrated internal reference plate; A standard external reference plate is used as a physical benchmark; The calibration processing unit is used to implement the near-infrared spectroscopy equipment stage difference calibration method as described above, and includes a measurement control module, a conversion module and a parameter storage module; The measurement control module is used to control the near-infrared spectroscopy equipment to perform spectral measurements on the internal reference plate and the standard external reference plate, respectively. The conversion module is used to obtain the calibration parameters of the near-infrared spectroscopy equipment based on the spectral data measured by the measurement control module; The parameter storage module is used to store the calibration parameters in the memory of the near-infrared spectroscopy device or in a cloud database.

[0014] Furthermore, the calibration processing unit connects to the near-infrared spectrometer to be calibrated via a standard communication interface, aligning the measurement window of the near-infrared spectrometer to be calibrated with the measurement surface of the standard external reference plate.

[0015] Compared with the prior art, the present invention has the following beneficial effects: By introducing a standard external reference board as a unified physical benchmark, a conversion model from the internal reference signal to the external benchmark signal is established for each near-infrared spectrometer, enabling the spectral measurements of different near-infrared spectrometers to be calibrated to the same benchmark. This solves the "data silo" problem caused by hardware differences and lays the foundation for establishing a portable and universal analysis model. The linear transformation model is computationally simple and stable, does not rely on a large amount of training data, and enables real-time, online correction of spectral measurements. It avoids problems such as overfitting, convergence difficulties or computational delays that may exist in complex nonlinear models, ensuring the high reliability and ease of implementation of the method. A large number of devices of the same model can be calibrated using a single standard external reference board, achieving high-precision calibration at a low cost, which is suitable for large-scale industrial production and application. The initial stage difference of the equipment is compensated by factory calibration, and long-term performance drift caused by light source attenuation, detector aging, etc. can be effectively tracked and corrected by periodic return-to-factory calibration. Attached Figure Description

[0016] Figure 1 This is a flowchart of the calibration method of the present invention; Figure 2 This is a schematic diagram of the system composition of the present invention. Detailed Implementation

[0017] In the mass production, quality control, and subsequent maintenance of multiple devices, due to differences in manufacturing tolerances, assembly tolerances, and device aging, there are significant systematic deviations in the response of each device to the same tested sample. This results in measurement data that cannot be directly compared and analytical models that cannot be transferred across devices. Existing technologies rely on the internal reference boards of each device for calibration, but the internal reference boards of different devices have individual differences and lack a unified absolute physical benchmark, forming "data silos" and making long-term quality traceability impossible.

[0018] The present invention aims to provide a calibration method that is cost-effective, easy to operate, and scalable. By introducing an independent standard external reference plate, a unified physical benchmark is established for all near-infrared spectroscopy equipment, fundamentally eliminating inter-equipment differences.

[0019] The invention will be further described below with reference to specific embodiments. Those skilled in the art should understand that these embodiments are only for explaining the invention and not for limiting its scope.

[0020] Example 1 like Figure 1 As shown, this embodiment provides a stage difference calibration method for near-infrared spectroscopy equipment based on an external reference plate, including the following steps: Step S1: Measure the internal reference plate of the near-infrared spectroscopy device and acquire raw spectral data; For each near-infrared spectrometer to be calibrated, before leaving the factory or at the start of the specified calibration cycle, the near-infrared spectrometer is controlled to measure its internal reference plate, and the raw spectral data is collected and recorded as follows: .

[0021] Step S2: Measure the standard external reference plate and collect standard spectral data; Under the same environmental and equipment conditions as in step S1, using the same standard external reference plate, standard spectral data were collected from multiple near-infrared spectral devices, denoted as... The standard external reference board is a standard white board made of Spectralon material.

[0022] Preferably, in this embodiment, steps S1 and S2 involve continuously collecting raw spectral data and standard spectral data multiple times, and then averaging the collected raw spectral data and standard spectral data to reduce random errors and improve data stability.

[0023] Step S3: Based on the original spectral data and standard spectral data, establish a spectral conversion model for the near-infrared spectral device; wherein, the spectral conversion model is a linear conversion model, and the specific steps for establishing it include: S31: For each wavelength point λ, calculate the conversion slope Slope(λ) and conversion intercept Intercept(λ): , , in, For the raw spectral data, Standard spectral data; S32: Use the obtained conversion slope and conversion intercept as calibration parameters for the near-infrared spectrometer to establish a spectral conversion model: ; in, The converted standard spectral signal, This is the original spectral signal measured during the actual measurement.

[0024] By transforming the model, when the input is the original spectral signal based on the internal reference board, the output is the standard spectral signal based on the standard external reference board, thereby unifying the internal reference of each near-infrared spectrometer to the external reference reference.

[0025] Furthermore, each of the multiple near-infrared spectrometers has a corresponding spectral conversion model. The spectral conversion model is associated with the identifier of the corresponding near-infrared spectrometer and stored in the hardware memory or cloud database of the near-infrared spectrometer. Specifically, the calculated Slope(λ) and Intercept(λ), which vary with wavelength, are used as the calibration parameter set of the device and are strongly associated with the unique identifier of the device.

[0026] After establishing the above conversion model, the near-infrared spectrometer can be calibrated using the conversion model when measuring any actual sample: Step S4, acquire the original spectral signal, convert the original spectral signal in real time based on the spectral conversion model, and output the converted standard spectral signal. Specifically: Retrieve the calibration parameters Slope(λ) and Intercept(λ) of the near-infrared spectrometer from its memory; Based on calibration parameters, the original spectral signal Perform real-time conversion and output the converted standard spectral signal. .

[0027] According to this embodiment, the converted standard spectral signal output is comparable spectral data calibrated to the same reference as other spectral devices of the same model. Therefore, a complex model established in one spectral device can be ported to other spectral devices, which is suitable for application scenarios that require large-scale collaborative operation.

[0028] Preferably, step S4 further includes: The converted standard spectral signal is verified by comparing it with the standard spectral signal obtained from the external reference board of the measurement standard to determine whether it meets the expected conditions. If it does, the corresponding near-infrared spectral equipment stage error calibration is completed. The expected conditions are set based on the equipment design specifications and can be evaluated using the root mean square error (RMSE) or the cosine of the spectral angle. In this embodiment, the RMSE is 0.005. If the expected conditions are not met, the following anomaly handling steps are executed: The interactive interface or indicator light will be triggered to prompt the operator to clean the standard external reference plate and the measurement window of the near-infrared spectroscopy measurement equipment, and to re-execute the stage difference calibration method of the near-infrared spectroscopy equipment. If the expected conditions are not met in N consecutive verifications, it is determined that the internal optical system of the current device has experienced severe physical drift or hardware aging, such as excessive energy attenuation of the light source or abnormal deformation of the drive structure. At this time, it is outside the calibration range, an error log with band difference characteristics is generated, the calibration status of the device is locked, and it is indicated that it needs to be returned to the factory for repair or the optical module needs to be replaced.

[0029] Furthermore, after calibrating one near-infrared spectroscopy device, the next device is calibrated using the same steps until all near-infrared spectroscopy devices are calibrated.

[0030] Example 2 The optical components of the equipment may exhibit nonlinear response attenuation. Therefore, this embodiment provides a stage difference calibration method for near-infrared spectroscopy equipment based on an external reference plate. The spectral conversion model established in step S3 is a polynomial conversion model. By acquiring multiple standard external reference plates with different reflectance gradients, including 99%, 75%, and 50% standard reference plates, multiple sets of standard spectral data are obtained, specifically including: S31': For each wavelength point λ in the spectrum, calculate the model coefficients of the polynomial transformation model based on the original spectral data and standard spectral data: ; in, For the raw spectral data, For standard spectral data, , and To calculate the model coefficients at each wavelength point using the least squares method; S32': Model coefficients , and A polynomial transformation model is established as the calibration parameter for the near-infrared spectrometer: ; in, The converted standard spectral signal, This is the original spectral signal.

[0031] Furthermore, each of the multiple near-infrared spectrometers has a corresponding spectral conversion model. This model is associated with the identifier of the corresponding near-infrared spectrometer and stored in the hardware memory of the near-infrared spectrometer or in a cloud database. Specifically, the calculated wavelength-varying... , and The calibration parameter set of the device is strongly associated with the device's unique identifier.

[0032] After establishing the above conversion model, the near-infrared spectrometer can be calibrated using the conversion model when measuring any actual sample: Step S4, acquire the original spectral signal, convert the original spectral signal in real time based on the spectral conversion model, and output the converted standard spectral signal. Specifically: Retrieve the calibration parameters of the near-infrared spectrometer from its memory. , and ; Based on calibration parameters, the original spectral signal Perform real-time conversion and output the converted standard spectral signal. .

[0033] According to this embodiment, the original spectral signal is transformed by a polynomial transformation model to obtain a standard spectral signal based on a standard external reference plate, thereby unifying the internal reference of each near-infrared spectroscopy device to the external reference reference.

[0034] According to another preferred embodiment, considering the correlation and optical crosstalk between adjacent wavelength points, a segmented direct normalization method is adopted. Based on obtaining the original spectral data for different wavelength points, an adjacent window (of size ) is also introduced. w The set of wavelength points within the w window is used as the input matrix, and the original spectral data and the set of wavelength points within the w window are used to establish a local multivariate transformation matrix through partial least squares method to reconstruct the standard spectral signal.

[0035] Based on the above embodiments, those skilled in the art can establish and select a better spectral conversion model according to different influencing factors to obtain the required standard spectral signal.

[0036] Example 3 This embodiment provides a stage difference calibration method for near-infrared spectroscopy equipment based on an external reference plate. Building upon Embodiment 1, it further eliminates the influence of electrical noise, specifically including: Under no-light conditions, dark current spectral data of the near-infrared spectral equipment are collected, and the average value of the dark current spectral data is calculated for subsequent spectral data correction. The dark current spectral data refers to the raw light intensity energy data output by the detector when the light source of the near-infrared spectral equipment is completely turned off or the optical path is completely blocked by the light shutter. It is the background noise signal caused by the thermionic emission of the detector and the system circuit.

[0037] To eliminate the interference of random thermal noise, the dark current spectral signal was acquired multiple times and the average value at each wavelength was calculated, denoted as . ; Based on the multiple raw spectral data collected in Example 1, an average processing was performed, and dark current spectral data were subtracted: This yields a stable and clean internal reference signal (corrected original spectral data); similarly, multiple standard spectral data are averaged, and dark current spectral data are subtracted. This yields an external reference signal (corrected standard spectral data).

[0038] Furthermore, a spectral conversion model for the near-infrared spectral device is established based on the corrected original spectral data and standard spectral data to convert the corrected original spectral data into corrected standard spectral data.

[0039] The calculated calibration parameters, device identifier, calibration date, and other metadata are packaged into a complete device-specific calibration file, which is then burned and stored in the non-volatile memory of the near-infrared spectroscopy device.

[0040] Other specific implementation methods of this embodiment can be found in Embodiment 1, and will not be repeated here.

[0041] Based on the same technical concept, this invention also provides a near-infrared spectroscopy equipment stage difference calibration system based on an external reference plate, used for spectral stage difference calibration of near-infrared spectroscopy equipment, such as... Figure 2 As shown, the system includes: Multiple near-infrared spectrometers of the same model are to be calibrated. Each near-infrared spectrometer integrates an internal reference plate. The internal reference plate can ensure that the results are repeatable when the same instrument measures the same sample at different times, thereby maintaining the measurement stability of a single instrument.

[0042] A standard external reference plate, a standard white plate made of Spectralon material, physically covers the measurement spot of the near-infrared spectroscopy equipment to be calibrated; this external reference plate has high reflectivity, excellent Lambertian properties and excellent long-term chemical and physical stability, and is used as a physical reference. The calibration processing unit is used to implement the near-infrared spectroscopy equipment stage difference calibration method as described in Examples 1-3, and includes a measurement control module, a conversion module and a parameter storage module; preferably, the calibration processing unit can be integrated into the control system of a single spectroscopy equipment, or a separate computer, industrial computer or server, etc. The measurement and control module is used to send commands to the near-infrared spectroscopy equipment to control the near-infrared spectroscopy equipment to perform spectral measurements on the internal reference plate and the standard external reference plate, and to collect raw spectral data and standard spectral data. Based on the spectral data measured by the measurement and control module, the conversion module calculates the conversion relationship between the original spectral data and the standard spectral data for each near-infrared spectrometer, and obtains the calibration parameters of the near-infrared spectrometer. The parameter storage module is used to store the calculated calibration parameters in the non-volatile memory of the near-infrared spectroscopy device or in an associated cloud database.

[0043] In this embodiment, the calibration processing unit is an industrial control computer. The computer is connected to the near-infrared spectroscopy equipment to be calibrated through a standard communication interface, and the computer and the spectroscopy equipment can exchange control signals and data. After the industrial control computer establishes a connection with the near-infrared spectroscopy equipment to be calibrated, it controls the near-infrared spectroscopy equipment to collect multiple sets of dark current spectral data under no-light conditions and calculates the average value for subsequent spectral data correction. The industrial control computer controls the near-infrared spectroscopy equipment to measure the internal reference plate, collect multiple sets of raw spectral data, average the multiple sets of data, and subtract the average value of the dark current spectral data to obtain the corrected raw spectral data. Under the same equipment parameters, the industrial control computer controls the near-infrared spectroscopy equipment to measure the standard external reference plate, collect multiple sets of standard spectral data, average the data, and subtract the average value of the dark current spectral data to obtain the corrected original spectral data. In use, the standard external reference plate is stably placed on a fixed platform, and the measurement window of each near-infrared spectroscopy equipment to be calibrated is aligned with the measurement surface of the standard external reference plate for measurement.

[0044] The industrial control computer performs calculations and analysis based on the corrected original spectral data to obtain the quantitative conversion relationship between the two and generates calibration parameters. Then, it packages the calibration parameters, device unique identifier, calibration date and other metadata into a device-specific calibration file and burns and stores the device-specific calibration file into the non-volatile memory of the near-infrared spectroscopy device through the communication interface.

[0045] When measuring actual samples, near-infrared spectroscopy equipment reads calibration parameters from the memory to convert the original spectrum in real time and outputs a standard spectral signal after stage difference calibration.

[0046] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or portion of code containing at least one executable instruction for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0047] In general, various exemplary embodiments of the present invention can be implemented in hardware or dedicated circuitry, software, firmware, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software that can be executed by a controller, microprocessor, or other computing device. When aspects of embodiments of the present invention are illustrated or described as block diagrams, flowcharts, or represented using certain other images, it will be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or certain combinations thereof.

[0048] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0049] While the specific embodiments of the present invention have been described above, they are not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. An external reference plate based near infrared spectroscopy equipment bench difference calibration method, characterized in that, Includes the following steps: Step S1: Measure the internal reference plate of the near-infrared spectroscopy device and acquire raw spectral data; Step S2: Measure the standard external reference plate and collect standard spectral data; Step S3: Based on the original spectral data and standard spectral data, establish the spectral conversion model of the near-infrared spectral device; Step S4: Obtain the original spectral signal, perform real-time conversion on the original spectral signal based on the spectral conversion model, and output the converted standard spectral signal.

2. The external reference plate based near infrared spectroscopy instrument bench difference calibration method according to claim 1, wherein, The spectral conversion model established in step S3 is a linear conversion model, specifically including: S31: For each wavelength point λ in the spectrum, calculate the conversion slope Slope(λ) and conversion intercept Intercept(λ): ， ， wherein is the original spectral data, is the standard spectral data; S32: Using the conversion slope and conversion intercept as calibration parameters for the near-infrared spectrometer, establish a spectral conversion model: ； wherein, is the converted standard spectral signal, is the original spectral signal.

3. The near-infrared spectroscopy equipment stage difference calibration method based on an external reference plate according to claim 2, characterized in that, There are multiple near-infrared spectrometers, each with its own calibration parameters. These calibration parameters are associated with the identifiers of the corresponding near-infrared spectrometers and stored in the memory of the near-infrared spectrometers.

4. The near-infrared spectroscopy equipment stage difference calibration method based on an external reference plate according to claim 3, characterized in that, Step S4 specifically involves: The calibration parameters of the near-infrared spectrometer are retrieved from its memory. The original spectral signal is converted based on the calibration parameters, and the converted standard spectral signal is output.

5. The near-infrared spectroscopy equipment stage difference calibration method based on an external reference plate according to claim 3, characterized in that, The multiple near-infrared spectroscopy devices use the same standard external reference board, which is a standard white board made of Spectralon material.

6. The near-infrared spectroscopy equipment stage difference calibration method based on an external reference plate according to claim 1, characterized in that, The spectral conversion model established in step S3 is a polynomial conversion model, specifically including: S31': For each wavelength point λ in the spectrum, calculate the model coefficients of the polynomial transformation model based on the original spectral data and the standard spectral data: ； in, For the raw spectral data, For standard spectral data, , and To calculate the model coefficients at each wavelength point using the least squares method; S32': The model coefficients , and A polynomial transformation model is established as the calibration parameter for the near-infrared spectrometer: ； in, The converted standard spectral signal, This is the original spectral signal.

7. The near-infrared spectroscopy equipment stage difference calibration method based on an external reference plate according to claim 1, characterized in that, The method further includes: Dark current spectral data were collected under conditions of no light. The dark current spectral data is subtracted from the original spectral data and the standard spectral data respectively to obtain the corrected original spectral data and standard spectral data; A spectral conversion model for the near-infrared spectrometer is established based on the corrected original spectral data and standard spectral data.

8. The near-infrared spectroscopy equipment stage difference calibration method based on an external reference plate according to claim 1, characterized in that, Step S4 further includes: The standard spectral signal obtained from the external reference plate of the measurement standard; The converted standard spectral signal is compared with the standard spectral signal obtained from the external reference plate of the measurement standard to determine whether it meets the expected conditions. If it does, the corresponding near-infrared spectroscopy equipment stage difference calibration is completed.

9. A stage difference calibration system for near-infrared spectroscopy equipment based on an external reference plate, characterized in that, The system includes: Multiple near-infrared spectrometers of the same model to be calibrated, each with an integrated internal reference plate; A standard external reference plate is used as a physical benchmark; A calibration processing unit is used to implement the near-infrared spectroscopy equipment stage difference calibration method as described in claim 1, and includes a measurement control module, a conversion module, and a parameter storage module; The measurement control module is used to control the near-infrared spectroscopy equipment to perform spectral measurements on the internal reference plate and the standard external reference plate, respectively. The conversion module is used to obtain the calibration parameters of the near-infrared spectroscopy equipment based on the spectral data measured by the measurement control module; The parameter storage module is used to store the calibration parameters in the memory of the near-infrared spectroscopy device or in a cloud database.

10. The near-infrared spectroscopy equipment stage difference calibration system based on an external reference plate according to claim 9, characterized in that, The calibration processing unit connects to the near-infrared spectrometer to be calibrated via a standard communication interface, aligning the measurement window of the near-infrared spectrometer to be calibrated with the measurement surface of the standard external reference plate.

Images