Intelligent fitting method and system for molecular weight distribution of lentinan

Abstract

The application discloses a lentinan molecular weight distribution intelligent fitting method and system, and particularly relates to the field of analysis and detection and intelligent sensing technology, and is used for solving the problem that the existing combined technology cannot quantize the reliability of the molecular weight distribution result; by acquiring the concentration and light scattering original signal, evaluating the uncertainty sequence and analyzing the correlation, a signal uncertainty correlation matrix is constructed, the contribution degree of different elution times to the uncertainty is determined, and a molecular weight distribution uncertainty envelope is synthesized; then, the envelope is compared with the expected elution behavior to perform credibility grading, and finally, a molecular weight distribution fitting atlas with graded visual information is output; the quantization of the measurement uncertainty propagation process and the intuitive evaluation of the result reliability are realized.

Description

Technical Field

[0001] This invention relates to the field of analytical detection and intelligent sensing technology, and more specifically, to a method and system for intelligent fitting of the molecular weight distribution of lentinan. Background Technology

[0002] Gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector (GPC-MALLS-RI) is currently the main method for determining the molecular weight and distribution of biopolymers such as lentinan. It obtains the concentration information of the eluted components through differential refractive index detector and the light scattering signal through multi-angle laser light scattering detector. Based on relevant theoretical models, the molecular weight and molecular size are directly calculated without relying on calibration curves of standards. It is considered an absolute measurement method. Existing technical solutions usually treat the concentration and scattering signals collected by the above-mentioned independent detectors as deterministic input data and perform direct calculations through predetermined mathematical models to finally output the molecular weight distribution results.

[0003] However, in actual measurement, the raw signals output by each detector contain their inherent measurement uncertainties, such as baseline noise fluctuations, detector response nonlinearity, and slight sensitivity drift. Existing data processing methods, when fusing detector data from different physical principles, fail to consider the correlation between uncertainties in each signal channel and their propagation and amplification effects in subsequent cascade calculations. As a result, although the final molecular weight distribution spectrum has a clear numerical form, the overall reliability of the result, especially the confidence level in the low signal-to-noise ratio regions on both sides of the distribution, cannot be effectively assessed and quantified. This hidden uncertainty poses potential risks and controversies when conducting detailed comparisons and quality assessments of the molecular weight distribution of lentinan from different batches of samples or obtained through different processes. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent fitting method and system for the molecular weight distribution of lentinan to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for intelligently fitting the molecular weight distribution of lentinan includes: S1. Obtain the original concentration signal and original light scattering signal output by the gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector system when analyzing lentinan samples; S2. Evaluate the uncertainty of the original concentration signal and the original light scattering signal respectively to obtain the corresponding uncertainty sequence; S3. Analyze the correlation between uncertainty sequences to obtain the signal uncertainty correlation matrix; S4. Based on the signal uncertainty correlation matrix, determine the contribution of different elution time periods to the uncertainty of molecular weight distribution, and synthesize the uncertainty envelope of molecular weight distribution. S5. Compare the uncertainty envelope of molecular weight distribution with the expected elution volume-molecular weight relationship of lentinan sample in the coupled system. Based on the comparison results, classify the confidence of uncertainty envelope in different elution volume ranges to obtain uncertainty envelope with confidence classification information. S6. Based on the uncertainty envelope with confidence level information, output the molecular weight distribution fitting spectrum of the shiitake polysaccharide sample.

[0006] Furthermore, S1 includes: Simultaneously acquire the concentration simulation signal output by the differential refractive index detector and the multi-angle light scattering simulation signal output by the multi-angle laser light scattering detector; The concentration simulation signal and the multi-angle light scattering simulation signal were converted into digital signals, which were used as the original concentration signal and the original light scattering signal, respectively.

[0007] Furthermore, converting the concentration analog signal and the multi-angle light scattering analog signal into digital signals includes: performing synchronous analog-to-digital conversion on the concentration analog signal and the multi-angle light scattering analog signal at a sampling rate of not less than twice the highest signal frequency of the detector.

[0008] Furthermore, S2 includes: Based on the noise statistical characteristics of the original concentration signal in the baseline region without sample elution, the concentration measurement uncertainty of each data point of the original concentration signal is evaluated, and a concentration signal uncertainty sequence is formed. Based on the noise statistics of the original light scattering signal in the baseline section without sample elution, the measurement uncertainty of light scattering intensity at each data point of the original light scattering signal is evaluated, forming a light scattering signal uncertainty sequence.

[0009] Furthermore, S3 includes: Align the concentration signal uncertainty sequence with the light scattering signal uncertainty sequence on the time axis; Calculate the degree of matching between the fluctuation patterns of the concentration signal uncertainty sequence and the fluctuation patterns of the light scattering signal uncertainty sequence at different time offsets; Based on the maximum matching degree and its corresponding time offset, the correlation strength and association pattern between the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence are determined. Based on the determined correlation strength and association pattern, a correlation matrix of signal uncertainty is generated.

[0010] Furthermore, S4 includes: Based on the signal uncertainty correlation matrix, the entire chromatographic elution process is divided into multiple consecutive elution time periods; For each elution time period, the contribution weight of the elution time period to the molecular weight distribution uncertainty is calculated by combining the values ​​of the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence within the elution time period. Based on the contribution weight of each elution time period, the global uncertainty distribution is mapped to the corresponding molecular weight range; The uncertainty envelope of the molecular weight distribution is synthesized by superimposing the uncertainty distributions obtained from mapping across all molecular weight ranges.

[0011] Furthermore, S5 includes: Obtain the expected elution volume-molecular weight relationship curve for lentinan samples; Calculate the degree of deviation between the central trend line of the uncertainty envelope of the molecular weight distribution and the expected elution volume-molecular weight relationship curve in each elution volume interval; Based on the magnitude of the deviation, a corresponding confidence level is assigned to the uncertainty envelope of each elution volume range; The confidence level is combined with the uncertainty envelope of the molecular weight distribution as additional information to form an uncertainty envelope with confidence level information.

[0012] Furthermore, obtaining the expected elution volume-molecular weight relationship curve for lentinan samples includes: establishing a linear calibration curve of elution volume versus the logarithm of molecular weight by analyzing the elution behavior of lentinan standards with known molecular weights in the coupled system.

[0013] Furthermore, S6 includes: The central trend line of molecular weight distribution is determined based on the uncertainty envelope with confidence level information. The uncertainty envelope with confidence level information is combined with the central trend line of the molecular weight distribution in the form of a boundary line; Based on the credibility rating information, the boundary lines are visually labeled; Generate and output a molecular weight distribution fitting map of lentinan samples, including a central trend line of molecular weight distribution, boundary lines with visual annotations, and a legend.

[0014] On the other hand, the present invention provides an intelligent fitting system for the molecular weight distribution of lentinan, comprising: The signal acquisition module is used to acquire the original concentration signal and original light scattering signal output by the gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector system when analyzing lentinan samples; The evaluation and processing module is used to evaluate the uncertainty of the original concentration signal and the original light scattering signal respectively, and obtain the corresponding uncertainty sequence. The correlation analysis module is used to analyze the correlation between uncertainty sequences and obtain the signal uncertainty correlation matrix; The envelope synthesis module is used to determine the contribution of different elution time periods to the uncertainty of molecular weight distribution based on the signal uncertainty correlation matrix, and to synthesize the uncertainty envelope of molecular weight distribution. The grading module is used to compare the uncertainty envelope of molecular weight distribution with the expected elution volume-molecular weight relationship of lentinan sample in the coupled system. Based on the comparison results, the confidence of the uncertainty envelope in different elution volume ranges is graded to obtain uncertainty envelope with confidence grading information. The spectrum output module is used to output a molecular weight distribution fitting spectrum of shiitake polysaccharide samples based on the uncertainty envelope with confidence level information.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By systematically introducing and processing the uncertainty information of multi-detector signals in the traditional molecular weight calculation process, the metrological quality of the final distribution results is significantly improved. First, the measurement uncertainties of each detector are evaluated and their interrelationships are analyzed. Then, based on the signal uncertainty correlation matrix, the temporal uncertainty contribution is mapped to the molecular weight dimension, thereby synthesizing a molecular weight distribution uncertainty envelope that can objectively reflect the error accumulation of the entire measurement chain. This achieves quantitative control over the propagation and amplification of uncertainty, so that the output molecular weight distribution spectrum not only includes the central trend line, but also includes an envelope that characterizes the reliability range of each point. This solves the problem that the reliability of the results cannot be quantitatively evaluated in the existing technology, and provides a solid metrological foundation for high-precision comparison of the differences in molecular weight distribution of samples from different batches.

[0016] 2. The uncertainty envelope of the synthesis is compared with the expected elution behavior of the sample in the chromatographic system, and the confidence level of the envelope is graded according to the degree of deviation. The output fitted spectrum is visually labeled to intuitively distinguish the high, medium and low confidence intervals, so that the analyst can clearly identify the reliable part of the result and the part that may have potential problems. This not only enhances the practicality and interpretability of the molecular weight distribution results, but also provides refined data support that can directly guide decision-making for quality control, process optimization and structure-activity relationship studies based on molecular weight distribution. This reflects the deep application value of intelligent sensing and data analysis technology in the field of complex biopolymer characterization. Attached Figure Description

[0017] Figure 1 This is a flowchart of an intelligent fitting method for the molecular weight distribution of lentinan according to the present invention; Figure 2 This is a schematic diagram of the structure of an intelligent fitting system for the molecular weight distribution of lentinan according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Figure 1 This invention provides an intelligent fitting method for the molecular weight distribution of lentinan, comprising: S1. Obtain the original concentration signal and original light scattering signal output by the gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector system when analyzing lentinan samples; S2. Evaluate the uncertainty of the original concentration signal and the original light scattering signal respectively to obtain the corresponding uncertainty sequence; S3. Analyze the correlation between uncertainty sequences to obtain the signal uncertainty correlation matrix; S4. Based on the signal uncertainty correlation matrix, determine the contribution of different elution time periods to the uncertainty of molecular weight distribution, and synthesize the uncertainty envelope of molecular weight distribution. S5. Compare the uncertainty envelope of molecular weight distribution with the expected elution volume-molecular weight relationship of lentinan sample in the coupled system. Based on the comparison results, classify the confidence of uncertainty envelope in different elution volume ranges to obtain uncertainty envelope with confidence classification information. S6. Based on the uncertainty envelope with confidence level information, output the molecular weight distribution fitting spectrum of the shiitake polysaccharide sample.

[0020] S1. Obtain the original concentration signal and original light scattering signal output by the gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector system when analyzing lentinan samples. The specific implementation is as follows: The prepared lentinan sample solution was injected into a gel permeation chromatography system via an injector and, driven by the mobile phase, entered the chromatographic column for separation. At the column outlet, the eluent was split into the flow cells of a differential refractive index detector (DRID) and a multi-angle laser light scattering detector (MLS). During data acquisition, a shared timing control unit simultaneously sent an electrical signal to both the DRID and MLS to initiate operation, enabling them to measure the eluent components flowing through their respective flow cells at the same time. The DRID generates a continuous voltage signal based on the difference in refractive index between the eluent and the pure mobile phase. The magnitude of the voltage signal is proportional to the change in refractive index, and the voltage signal is a concentration simulation signal. The MLS simultaneously measures the intensity of scattered light generated by the eluent at multiple preset scattering angles and outputs a set of continuous voltage signals corresponding to the intensity of scattered light at each angle. The continuous voltage signals from each channel of the MLS output together constitute the multi-angle light scattering simulation signal. Both the concentration simulation signal and the multi-angle light scattering simulation signal are transmitted to the data acquisition equipment in real time via shielded cables.

[0021] The data acquisition equipment is equipped with multiple analog input channels and a high-precision clock circuit. The concentration analog signal from the differential refractive index detector is connected to a designated analog input channel, while each angle signal from the multi-angle laser light scattering analog signal is connected to an independent analog input channel. The high-precision clock circuit of the data acquisition equipment generates a periodic electrical pulse signal as the sampling clock. This sampling clock is simultaneously connected to the clock input ports of the analog-to-digital converters (ADCs) of all analog input channels, ensuring that all ADCs perform sampling operations on the rising edge of the same electrical pulse, achieving synchronous acquisition of all analog signals. Before performing analog-to-digital conversion, the sampling rate needs to be set. Referring to the differential refractive index detector's technical manual, the highest frequency of the output signal is 5 Hz; referring to the multi-angle laser light scattering detector's technical manual, the highest frequency of the output signal for each channel is 10 Hz. The maximum value of 10 Hz is selected from the highest frequencies of the differential refractive index detector and the multi-angle laser light scattering detector. According to the Nyquist sampling theorem, the sampling rate must be higher than twice the highest frequency of the signal. When setting the sampling rate, a value no lower than 20 Hz should be chosen. For example, the sampling rate can be set to 50 Hz, and the sampling interval can be 0.02 seconds. The data acquisition device, at a sampling rate of 50 Hz, synchronously acquires the instantaneous voltage values ​​of all analog input channels every 0.02 seconds, triggered by a sampling clock pulse. The analog-to-digital converter (ADC) connected to each analog input channel converts the acquired analog voltage value into a digital value. The number of bits in the ADC determines the precision of the digital value. For example, using a 16-bit ADC, when the analog voltage range is ±5 volts, a +1 volt will be converted into a digital value of approximately +6553, and a -1 volt will be converted into a digital value of approximately -6553. The analog-to-digital conversion process is completed simultaneously for all channels of the concentration analog signal and the multi-angle light scattering analog signal.

[0022] After analog-to-digital conversion, a series of time-ordered digital values ​​from the differential refractive index detector channel are stored; this series of digital values ​​represents the original concentration signal. A series of time-ordered digital values ​​from each channel of the multi-angle laser light scattering detector are stored separately. Multiple sets of time-aligned digital values ​​from the multi-angle laser light scattering detector together constitute the original light scattering signal. Both the original concentration signal and the original light scattering signal are arranged with time as the x-axis and the converted digital values ​​as the y-axis, forming time-series data. The time-series data is written to a data file in the computer system for storage. Each data point in the data file is recorded with precise time stamp information derived from the timing value of the high-precision clock circuit of the data acquisition device. By sharing the same high-precision clock circuit to provide a sampling clock for all analog-to-digital converters and recording uniform timing values, a strict time synchronization relationship between the original concentration signal and the original light scattering signal is ensured. The entire signal acquisition process continues from the start of sample elution to the end of the preset analysis time. The preset analysis time should be long enough to ensure complete capture of the elution process of all components of the lentinan sample.

[0023] S2. Evaluate the uncertainty of the original concentration signal and the original light scattering signal respectively to obtain the corresponding uncertainty sequence. The specific implementation is as follows: Uncertainty assessment was performed on the raw concentration signal to identify the baseline segment without sample elution. The baseline segment without sample elution corresponds to the signal segment in the chromatogram where the sample component has either not eluted or has been completely eluted, and the differential refractive index detector responds only to the pure mobile phase. The method for identifying the baseline segment without sample elution is to observe the raw concentration signal curve and select a continuous time period with relatively stable signal values ​​and small fluctuations. The average value of the signal within this continuous time period represents the baseline value. The baseline segment without sample elution can be found in the initial stage before sample injection and in the final stage after all chromatographic peaks have eluted. After determining the baseline segment without sample elution, the raw concentration signal values ​​corresponding to all data points within the baseline segment without sample elution are extracted to form a baseline noise dataset.

[0024] The statistical characteristics of noise are calculated based on the baseline noise dataset. The mean of all values ​​in the baseline noise dataset is calculated to obtain the baseline noise mean. The difference between each value in the baseline noise dataset and the baseline noise mean is calculated to obtain a series of deviation values. The standard deviation of this series of deviation values ​​is calculated to obtain the baseline noise standard deviation. The baseline noise standard deviation characterizes the random fluctuation amplitude of the original concentration signal under sample-free elution conditions. For example, if the baseline noise dataset contains 1000 data points, the calculated baseline noise mean is 1000 numerical values, and the baseline noise standard deviation is 5 numerical values. After obtaining the baseline noise standard deviation, the concentration measurement uncertainty of each data point in the original concentration signal is evaluated. The concentration measurement uncertainty of any data point in the original concentration signal is assigned the value of the baseline noise standard deviation. The concentration measurement uncertainty of all data points in the original concentration signal is evaluated to the same value, i.e., the baseline noise standard deviation. The concentration measurement uncertainty values ​​corresponding to each data point in the original concentration signal are arranged in chronological order of the original concentration signal to form a concentration signal uncertainty sequence. The concentration signal uncertainty sequence has the same number of data points and time coordinates as the original concentration signal.

[0025] Uncertainty assessment of the raw light scattering signal requires independent evaluation for each scattering angle channel within the signal. Taking a specific scattering angle channel as an example, the baseline segment without sample elution is identified from the raw light scattering signal of that channel. The method for identifying this baseline segment is to select a continuous time period where the raw light scattering signal value of that channel is relatively stable and has minimal fluctuations; this continuous time period corresponds to the state where the sample has not been eluted or has been completely eluted. After determining the baseline segment without sample elution for that channel, the raw light scattering signal values ​​corresponding to all data points within this baseline segment are extracted to form the baseline noise dataset for that channel.

[0026] The statistical characteristics of the noise are calculated based on the baseline noise dataset of this channel. The average value of all values ​​in the baseline noise dataset of this channel is calculated to obtain the baseline noise average value of this channel. The difference between each value in the baseline noise dataset of this channel and the average baseline noise value of this channel is calculated to obtain a series of deviation values ​​for this channel. The standard deviation of this series of deviation values ​​of this channel is calculated to obtain the baseline noise standard deviation of this channel. The baseline noise standard deviation of this channel characterizes the random fluctuation amplitude of the original light scattering signal of this scattering angle channel under sample elution conditions. For example, for a 90-degree scattering angle channel, its baseline noise dataset contains 1000 data points, and the calculated baseline noise average value of this channel is 800 digital values, and the baseline noise standard deviation of this channel is 8 digital values. After obtaining the baseline noise standard deviation of this channel, the measurement uncertainty of the light scattering intensity of each data point in the original light scattering signal of this channel is evaluated. The measurement uncertainty of the light scattering intensity of any data point in the original light scattering signal of this channel is assigned the value of the baseline noise standard deviation of this channel. The measurement uncertainty of light scattering intensity for all data points in the original light scattering signal of this channel is evaluated as the same value, i.e., the baseline noise standard deviation of this channel. The measurement uncertainty values ​​of light scattering intensity for each data point in the original light scattering signal of each scattering angle channel are arranged in chronological order of the original light scattering signal, forming the light scattering signal uncertainty sequence for each channel. The light scattering signal uncertainty sequence for each channel has the same number of data points and time coordinates as the original light scattering signal for that channel. The light scattering signal uncertainty sequences for all scattering angle channels together constitute the complete uncertainty evaluation result of the original light scattering signal.

[0027] S3. Analyze the correlation between uncertainty sequences to obtain the signal uncertainty correlation matrix. The specific implementation is as follows: The correlation between the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence is analyzed, and their time axis alignment is performed. Time axis alignment utilizes the shared time stamp information of both sequences. Each data point in both the concentration and light scattering signal uncertainty sequences is time-stamped. The alignment operation arranges the data points according to the numerical value of their time stamps, ensuring that data points with the same time stamp occupy the same time position in the analysis. For example, a concentration signal uncertainty data point with a time stamp of 0 seconds is aligned with a light scattering signal uncertainty data point with a time stamp of 0 seconds. For each scattering angle channel of the light scattering signal uncertainty sequence, the time axis alignment operation for that channel's light scattering signal uncertainty sequence and concentration signal uncertainty sequence is performed separately.

[0028] This study calculates the degree of matching between the fluctuation patterns of the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence at different time offsets. The fluctuation pattern of the concentration signal uncertainty sequence is the curve shape of its value changing over time. Similarly, the fluctuation pattern of the light scattering signal uncertainty sequence is also a curve shape of its value changing over time. During the calculation, the time position of the concentration signal uncertainty sequence remains unchanged, while the light scattering signal uncertainty sequence is shifted forward or backward along the time axis, creating a time offset. For each time offset, the degree of matching between the shifted light scattering signal uncertainty sequence and the concentration signal uncertainty sequence in the overlapping time portion is calculated. The degree of matching is calculated by multiplying the concentration signal uncertainty sequence value at each time point of the overlapping portion by the light scattering signal uncertainty sequence value, and then summing the products at all time points. The sum is taken as the degree of matching at that time offset. The range of the time offset needs to be pre-defined and should encompass the maximum possible time difference between the signals. For example, the time offset range is set to -100 data points to +100 data points, and the data point interval is determined by the sampling time. When the sampling time is 0.02 seconds, the time corresponding to -100 data points is advanced by 2 seconds, and the time corresponding to +100 data points is delayed by 2 seconds. For each channel of the light scattering signal uncertainty sequence, the degree of matching between it and the concentration signal uncertainty sequence at each offset within the preset time offset range is calculated independently, resulting in a set of matching degree value sequences.

[0029] Based on the numerical sequence of matching degrees, find the maximum value in the sequence. The maximum matching degree is the largest value in the sequence. Record the time offset corresponding to the maximum matching degree. The magnitude of the maximum matching degree is used to determine the correlation strength between the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence. Determining the correlation strength requires setting a correlation strength threshold. The correlation strength threshold is obtained as follows: generate multiple sets of random number sequences of the same length as the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence, but independent of each other. Calculate the maximum matching degree among these random number sequences, statistically analyze the distribution of these maximum values, and take the 95th percentile of this distribution as the correlation strength threshold. For example, generate 1000 sets of random number sequences, calculate 1000 maximum matching degrees, sort these 1000 values ​​by size, and the 950th largest value is the correlation strength threshold. If the maximum matching degree is greater than the correlation strength threshold, a significant correlation is determined between the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence. The strength of the correlation is represented by the ratio of the maximum matching degree to the correlation strength threshold. If the maximum matching degree is less than or equal to the correlation strength threshold, no significant correlation is determined, and the correlation strength is recorded as 0. The correlation pattern is represented by the sign and magnitude of the time offset corresponding to the maximum matching degree. A positive time offset indicates that the fluctuation of the light scattering signal uncertainty sequence lags behind the fluctuation of the concentration signal uncertainty sequence.

[0030] Based on the determined correlation strength and association pattern, a signal uncertainty correlation matrix is ​​generated. The signal uncertainty correlation matrix is ​​a mathematical matrix structure. The rows of the matrix correspond to the concentration signal uncertainty sequence, and the columns correspond to the light scattering signal uncertainty sequence for each scattering angle channel. Each element in the matrix is ​​a tuple containing two numbers. The first number of the tuple is the quantified value of the correlation strength, i.e., the ratio of the maximum matching degree to the correlation strength threshold, or 0. The second number of the tuple is the time offset in the association pattern. The quantified value of the correlation strength and the time offset calculated between the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence for each scattering angle channel are filled into the corresponding tuple in the matrix. For matrix elements containing the concentration signal uncertainty sequence itself, the first number of the tuple is 1, and the second number is 0. The final generated signal uncertainty correlation matrix fully records the quantitative information of the correlation strength and time correlation between the concentration signal uncertainty sequence and all light scattering signal uncertainty sequence channels.

[0031] S4. Based on the signal uncertainty correlation matrix, determine the contribution of different elution time periods to the molecular weight distribution uncertainty, and synthesize the uncertainty envelope of the molecular weight distribution. Specifically, the implementation is as follows: Based on the signal uncertainty correlation matrix, the entire chromatographic elution process is divided into multiple consecutive elution time periods. The total elution time is determined by the recording time of the original concentration signal. The method for dividing the elution time periods is to observe the curve of the original concentration signal and identify all obvious chromatographic peaks. Each peak represents the elution of a group of lentinan components with similar molecular weights. The lowest point of the trough between two adjacent peaks, as well as the start and end points of the entire chromatogram, are selected as the boundary points of the elution time periods. For example, if the original concentration signal curve shows three main peaks, the start point of the chromatogram, the trough between the first and second peaks, the trough between the second and third peaks, and the end point of the chromatogram are selected as boundary points. These four boundary points divide the entire elution process into three consecutive elution time periods. Another method is to use a fixed time interval for uniform division, such as dividing the total elution time into elution time periods of 10 seconds each. The defined elution time segments are continuous on the timeline and cover the entire elution process. Adjacent elution time segments are connected end-to-end at their time boundaries, and each elution time segment has a clearly defined start and end point. The start and end points of each elution time segment are recorded in a time segment index list.

[0032] For each elution time period, the contribution weight of the elution time period to the molecular weight distribution uncertainty is calculated by combining the values ​​of the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence within the elution time period. For a specific elution time period, all data points with time markers between the start and end points of that elution time period are extracted from the concentration signal uncertainty sequence to obtain the concentration signal uncertainty subsequence for that elution time period. All data points with time markers between the same start and end points are extracted from the light scattering signal uncertainty sequence for each scattering angle channel to obtain each light scattering signal uncertainty subsequence for that elution time period. The sum of squares of all data points in the concentration signal uncertainty subsequence for that elution time period is calculated. For each scattering angle channel, the sum of squares of all data points in the light scattering signal uncertainty subsequence for that elution time period is calculated, and then this sum is multiplied by the correlation strength quantification value corresponding to the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence of that channel in the signal uncertainty correlation matrix. The correlation strength quantification value is the first number in the tuple of the signal uncertainty correlation matrix. The results of multiplication for all scattering angle channels are summed. The sum of squares of the concentration signal uncertainty subsequences is added to the weighted sum of squares of all scattering angle channels to obtain a comprehensive uncertainty energy value. This comprehensive uncertainty energy value serves as the original contribution measure of the elution time period to the global molecular weight distribution uncertainty. The same calculation is performed on all elution time periods to obtain the original contribution measure for each elution time period. The original contribution measures of all elution time periods are summed to obtain the total original contribution measure. The original contribution measure of each elution time period is divided by the total original contribution measure, and the ratio is used as the contribution weight for that elution time period. The contribution weight is a value between 0 and 1, and the sum of the contribution weights of all elution time periods equals 1. For example, if the original contribution measures calculated for three elution time periods are 200, 300, and 500, and the total original contribution measure is 1000, the contribution weights for the three elution time periods are 0.2, 0.3, and 0.5, respectively.

[0033] Based on the contribution weight of each elution time interval, the global uncertainty distribution is mapped to the corresponding molecular weight interval. The global uncertainty distribution is characterized by the concentration signal uncertainty sequence and the overall light scattering signal uncertainty sequence. Establishing the correspondence between elution time and molecular weight requires the calibration curve of lentinan in the gel permeation chromatography system. The calibration curve describes the linear relationship between elution volume and the logarithm of molecular weight. Under the condition of constant mobile phase flow rate, elution volume is proportional to elution time, therefore there is a linear relationship between elution time and the logarithm of molecular weight. According to the calibration curve, the molecular weight value corresponding to the midpoint of each elution time interval is calculated, which represents the median molecular weight of the eluted component within that elution time interval. Each elution time interval is mapped to a molecular weight interval centered on its median molecular weight. The width of the molecular weight interval is calculated by the slope of the calibration curve and the length of the elution time interval. For example, a elution period of 20 seconds, calculated using a calibration curve, corresponds to a molecular weight of 100,000 Daltons at the start of the elution period and 80,000 Daltons at the end. The molecular weight range for this elution period is 80,000 Daltons to 100,000 Daltons, with a median molecular weight of approximately 90,000 Daltons. The mapping operation assigns the contribution weight of each elution period to the corresponding molecular weight range. This contribution weight serves as a calibration factor for the uncertainty distribution amplitude within that molecular weight range. The global uncertainty distribution amplitude is represented by a baseline distribution, which is a normal distribution with a mean of 0 and a standard deviation of 1. The baseline distribution is scaled for each molecular weight range, with the scaling factor equal to the contribution weight of the elution period corresponding to that molecular weight range. For example, if the contribution weight for a certain molecular weight range is 0.3, then the uncertainty distribution amplitude within that molecular weight range is 0.3 times the amplitude of the baseline distribution.

[0034] The uncertainty envelope of the molecular weight distribution is synthesized by superimposing the uncertainty distributions mapped across all molecular weight intervals. The uncertainty envelope of the molecular weight distribution is a function defined along the molecular weight dimension, describing the possible total measurement uncertainty range at each molecular weight value. The synthesis process is performed point-by-point on a molecular weight basis. For a specific molecular weight value, it is determined which one or more molecular weight intervals it falls within. If the molecular weight value lies entirely within a molecular weight interval, the uncertainty envelope value for that molecular weight point is directly taken as the amplitude value calculated at that point from the scaled uncertainty distribution within that molecular weight interval. If the molecular weight value lies at the boundary between two molecular weight intervals, the amplitude values ​​of the molecular weight value in the uncertainty distributions corresponding to the two adjacent molecular weight intervals are calculated, and the maximum of the two amplitude values ​​is taken as the uncertainty envelope value for that molecular weight point. This search and calculation process is repeated for all molecular weight points to be examined. The molecular weight points to be examined can be set as points taken every 100 Daltons from the minimum to the maximum molecular weight, obtaining data pairs of molecular weight values ​​and uncertainty envelope values. These data pairs of molecular weight values ​​and uncertainty envelope values ​​are then connected to form a continuous uncertainty envelope curve. The uncertainty envelope of the molecular weight distribution is stored in the form of a data list, which contains a sequence of molecular weight values ​​and a corresponding sequence of uncertainty envelope values.

[0035] S5. Compare the uncertainty envelope of the molecular weight distribution with the expected elution volume-molecular weight relationship of the lentinan sample in the coupled system. Based on the comparison results, classify the reliability of the uncertainty envelope in different elution volume ranges to obtain an uncertainty envelope with reliability classification information. The specific implementation is as follows: The expected elution volume-molecular weight relationship curves for lentinan samples were obtained. These curves were obtained by analyzing the elution behavior of lentinan standards with known molecular weights in a gel permeation chromatography-multi-angle laser light scattering and differential refractive index (DRI) system. At least three lentinan standards with known and certified molecular weights were used. Each lentinan standard was injected into the GCI system at the same concentration and injection volume as the lentinan sample, under identical chromatographic conditions, including the same mobile phase composition, flow rate, column, and column temperature. The elution volume corresponding to the maximum value of the DRI signal peak for each lentinan standard was recorded. Data points were plotted on a two-dimensional coordinate system with the common logarithm of the molecular weight of each lentinan standard as the ordinate and the corresponding elution volume as the abscissa. Linear least squares was used to fit these data points, resulting in a straight line. The mathematical equation of this line is: the common logarithm of molecular weight equals the slope multiplied by the elution volume plus the intercept. This straight line is the linear calibration curve of elution volume versus logarithm of molecular weight. The slope, intercept, and linear correlation coefficient of the linear calibration curve were recorded. For example, using three lentinan standards with molecular weights of 10,000 Daltons, 50,000 Daltons, and 200,000 Daltons, the elution volumes were measured to be 15 mL, 13 mL, and 10 mL, respectively. Plotting the elution volume on the x-axis and the logarithm of molecular weight on the y-axis, a straight line with a slope of -0.2 and an intercept of 5.0 was obtained. This linear calibration curve represents the expected elution volume versus molecular weight relationship curve for lentinan samples in a gel permeation chromatography system coupled with multi-angle laser light scattering and differential refractive index detector.

[0036] The deviation of the central trend line of the uncertainty envelope of the molecular weight distribution from the expected elution volume-molecular weight relationship curve is calculated for each elution volume interval. The central trend line of the uncertainty envelope of the molecular weight distribution is the central line around which the uncertainty envelope is located. In the molecular weight dimension, the central trend line represents the most likely molecular weight distribution without considering uncertainty. The central trend line is extracted from the uncertainty envelope data of the molecular weight distribution. Specifically, for each molecular weight value in the uncertainty envelope data list, the arithmetic mean of the upper and lower boundaries of the uncertainty envelope corresponding to that molecular weight value is taken, and this arithmetic mean is used as the ordinate of the central trend line at that molecular weight value. The x-axis of the central trend line is converted from molecular weight to elution volume. The conversion is performed using the expected elution volume-molecular weight relationship curve. According to the equation of the expected elution volume-molecular weight relationship curve, each molecular weight value is substituted into the equation to solve for its theoretically corresponding expected elution volume. After the conversion, a central trend line is obtained with elution volume as the x-axis and molecular weight as the y-axis. The elution volume intervals need to be pre-divided. The method for dividing the elution volume intervals is to divide them along the elution volume dimension. For example, the range from the minimum to the maximum elution volume can be divided into 10 consecutive elution volume intervals. For each elution volume interval, the deviation between the observed molecular weight value of the central trend line within that interval and the expected molecular weight value corresponding to the expected elution volume-molecular weight relationship curve within that interval is calculated. The specific calculation steps are as follows: Several sampling points are uniformly selected within the elution volume interval, for example, the midpoint of the interval can be chosen as a sampling point. For each sampling point, the observed molecular weight value at that elution volume is obtained through linear interpolation based on the transformed central trend line. The elution volume value of the same sampling point is substituted into the expected elution volume-molecular weight relationship curve equation to calculate the expected molecular weight value at that sampling point. The relative deviation between the observed molecular weight value and the expected molecular weight value at that sampling point is calculated; the relative deviation is equal to the difference between the observed molecular weight value and the expected molecular weight value, divided by the expected molecular weight value. The arithmetic mean of the absolute values ​​of the relative deviations of all sampling points within the elution volume interval is calculated, and this arithmetic mean is defined as the deviation of the elution volume interval. For example, if the relative deviations of three sampling points within a certain elution volume range are calculated to be +0.05, +0.08, and -0.02, and the arithmetic mean of the absolute values ​​of these three relative deviations is 0.05, then the deviation of the elution volume range is 0.05.

[0037] Based on the magnitude of the deviation, a corresponding confidence level is assigned to the uncertainty envelope of each elution volume interval. The confidence level indicates the reliability of the uncertainty envelope within the corresponding elution volume interval. The number of confidence levels and the corresponding deviation threshold range for each confidence level are defined. For example, three confidence levels are defined: high confidence, medium confidence, and low confidence. Two deviation thresholds are set for these three levels: a high confidence threshold and a medium confidence threshold. The deviation thresholds are set based on statistical analysis of a large amount of historical lentinan sample test data. The deviation values ​​of a large number of lentinan samples in each elution volume interval are collected from historical data, and the distribution of these deviation values ​​is statistically analyzed. The high confidence threshold is set to the deviation value corresponding to the 30th percentile of this deviation value distribution, and the medium confidence threshold is set to the deviation value corresponding to the 70th percentile of this deviation value distribution. For example, if historical statistics show that the 30th percentile is 0.03 and the 70th percentile is 0.10, then the high confidence threshold is set to 0.03 and the medium confidence threshold is set to 0.10. The rule for assigning a confidence level to a elution volume range is as follows: if the deviation of the elution volume range is less than or equal to the high confidence threshold, then the elution volume range is assigned a high confidence level; if the deviation of the elution volume range is greater than the high confidence threshold but less than or equal to the medium confidence threshold, then the elution volume range is assigned a medium confidence level; if the deviation of the elution volume range is greater than the medium confidence threshold, then the elution volume range is assigned a low confidence level. For each defined elution volume range, the confidence level to which the elution volume range belongs is determined based on its calculated deviation and the preset high confidence and medium confidence thresholds.

[0038] The confidence level is combined as additional information with the uncertainty envelope of the molecular weight distribution to form an uncertainty envelope with confidence level information. The uncertainty envelope of the molecular weight distribution itself is a data list of molecular weight value sequences and corresponding uncertainty envelope value sequences. The combination operation adds a new confidence level code sequence to the data list of molecular weight value sequences and uncertainty envelope value sequences. Since the uncertainty envelope of the molecular weight distribution has already been linked to the elution volume intervals through coordinate transformation, each elution volume interval corresponds to a range of molecular weights. All data points in the data list whose molecular weight values ​​fall within a certain elution volume interval are marked with the confidence level assigned to that elution volume interval. The confidence level information is represented by numerical codes, for example, 3 for high confidence level, 2 for medium confidence level, and 1 for low confidence level. After marking, each molecular weight value data point has not only an uncertainty envelope value but also a confidence level code. This new data list containing the molecular weight value sequence, uncertainty envelope value sequence, and confidence level code sequence is the uncertainty envelope with confidence level information. The uncertainty envelope with confidence level information is stored in an electronic data file format for easy reading and use in subsequent steps. Its data structure clearly reflects the reliability differences of measurement results in different molecular weight ranges. After the binding is completed, the uncertainty envelope with confidence level information is ready for the final spectrum output.

[0039] S6. Based on the uncertainty envelope with confidence level information, output the molecular weight distribution fitting spectrum of the shiitake polysaccharide sample. The specific implementation is as follows: The central trend line of the molecular weight distribution is determined based on the uncertainty envelope with confidence level information. The uncertainty envelope with confidence level information includes a sequence of molecular weight values, a sequence of uncertainty envelope values, and a sequence of confidence level codes. The central trend line of the molecular weight distribution is calculated from the uncertainty envelope value sequence. Each element in the uncertainty envelope value sequence contains two values: an upper boundary value and a lower boundary value. For each molecular weight value in the molecular weight value sequence, the corresponding upper and lower boundary values ​​of the uncertainty envelope are read. The ordinate of the central trend line at that molecular weight value is calculated; the ordinate is equal to the sum of the upper and lower boundary values ​​of the uncertainty envelope divided by 2. This calculation is repeated for each molecular weight value in the molecular weight value sequence to obtain the sequence of ordinate values ​​for the central trend line. The sequence of ordinate values ​​for the central trend line and the sequence of molecular weight values ​​constitute the data points for the central trend line of the molecular weight distribution. The central trend line of the molecular weight distribution is stored in computer memory as a data list. For example, for a data point with a molecular weight of 100,000 Daltons, the upper boundary of the uncertainty envelope is 105,000 Daltons, the lower boundary of the uncertainty envelope is 95,000 Daltons, and the calculated ordinate of the central trend line is 100,000 Daltons.

[0040] The uncertainty envelope, containing confidence level information, is combined with the central trend line of the molecular weight distribution as a boundary line. This combination is performed within the plotting program. The plotting program creates a two-dimensional coordinate system, with molecular weight on the horizontal axis and signal intensity on the vertical axis. Data points from the central trend line of the molecular weight distribution are input into the plotting program, which connects these data points with a solid line to draw the central trend line. The upper boundary value sequence of the uncertainty envelope from both the molecular weight value sequence and the uncertainty envelope value sequence is input into the plotting program, which connects the data points with a dashed line to draw the upper boundary line of the uncertainty envelope. The lower boundary value sequence of the uncertainty envelope from both the molecular weight value sequence and the uncertainty envelope value sequence is input into the plotting program, which connects the data points with a dashed line to draw the lower boundary line of the uncertainty envelope. The plotting program fills the area between the upper and lower boundary lines of the uncertainty envelope with a semi-transparent color. The central trend line of the molecular weight distribution, the upper boundary line of the uncertainty envelope, the lower boundary line of the uncertainty envelope, and the semi-transparent filled area are displayed in the same graph.

[0041] Based on the confidence level information, the boundary lines are visually labeled. The confidence level information is stored in the confidence level code sequence. Visual labeling is achieved by modifying the visual attributes of the boundary lines. A mapping relationship is established between the confidence level codes and visual attributes. Visual attributes include line color and fill color. For example, the mapping relationship is established as follows: confidence level code 3 corresponds to a green line color and a semi-transparent green fill color; confidence level code 2 corresponds to a yellow line color and a semi-transparent yellow fill color; confidence level code 1 corresponds to a red line color and a semi-transparent red fill color. The upper and lower boundary lines of the uncertainty envelope are redrawn segment by segment. The confidence level code sequence is traversed to identify molecular weight intervals with consecutively identical confidence level codes. For each molecular weight interval, the corresponding numerical subsequence is extracted from the upper boundary value sequence of the uncertainty envelope and from the lower boundary value sequence of the uncertainty envelope. Based on the confidence level code corresponding to the interval, the mapping relationship is queried to obtain the line color and fill color. Using the obtained line colors, redraw the upper and lower boundary lines of the uncertainty envelope corresponding to this molecular weight range on the graph. Using the obtained fill colors, refill the area between the upper and lower boundary lines of the uncertainty envelope corresponding to this molecular weight range on the graph.

[0042] Generate and output a molecular weight distribution fitting graph of a lentinan sample, including a central trend line for molecular weight distribution, boundary lines with visual annotations, and a legend. Add axis labels to the graph, with the horizontal axis labeled as molecular weight and the vertical axis labeled as signal intensity. Add a title to the graph, stating "Molecular Weight Distribution Fitting Graph of Lentinan Sample." Add a legend to the blank area of ​​the graph. The legend contains a legend box containing multiple legend items. Each legend item consists of an example line segment and a text label. The first legend item shows a blue solid line labeled "Central Trend Line." The second legend item shows a green solid line labeled "High Confidence Range." The third legend item shows a yellow solid line labeled "Medium Confidence Range." The fourth legend item shows a red solid line labeled "Low Confidence Range." The legend box has a border. Integrate all graphic elements, including the central trend line for molecular weight distribution, boundary lines with visual annotations, axis labels, title, and legend, into a single complete image. Use the image export function of the plotting program to output the integrated image as an image file in PNG format. For example, output the image as a file named "Molecular Weight Distribution Fitting Map of Shiitake Polysaccharide Sample.png". The output image file is the final molecular weight distribution fitting map of the shiitake polysaccharide sample.

[0043] Example 2: Figure 2A schematic diagram of the structure of an intelligent fitting system for the molecular weight distribution of lentinan according to the present invention is provided. The intelligent fitting system for the molecular weight distribution of lentinan includes: The signal acquisition module is used to acquire the original concentration signal and original light scattering signal output by the gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector system when analyzing lentinan samples; The evaluation and processing module is used to evaluate the uncertainty of the original concentration signal and the original light scattering signal respectively, and obtain the corresponding uncertainty sequence. The correlation analysis module is used to analyze the correlation between uncertainty sequences and obtain the signal uncertainty correlation matrix; The envelope synthesis module is used to determine the contribution of different elution time periods to the uncertainty of molecular weight distribution based on the signal uncertainty correlation matrix, and to synthesize the uncertainty envelope of molecular weight distribution. The grading module is used to compare the uncertainty envelope of molecular weight distribution with the expected elution volume-molecular weight relationship of lentinan sample in the coupled system. Based on the comparison results, the confidence of the uncertainty envelope in different elution volume ranges is graded to obtain uncertainty envelope with confidence grading information. The spectrum output module is used to output a molecular weight distribution fitting spectrum of shiitake polysaccharide samples based on the uncertainty envelope with confidence level information.

[0044] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0045] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0046] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. Computer-readable storage media can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0047] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0048] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0049] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0050] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0051] If a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0052] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0053] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent fitting of the molecular weight distribution of lentinan, characterized in that, include: S1. Obtain the original concentration signal and original light scattering signal output by the gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector system when analyzing lentinan samples; S2. Evaluate the uncertainty of the original concentration signal and the original light scattering signal respectively to obtain the corresponding uncertainty sequence, including: Based on the noise statistical characteristics of the original concentration signal in the baseline region without sample elution, the concentration measurement uncertainty of each data point of the original concentration signal is evaluated, and a concentration signal uncertainty sequence is formed. Based on the noise statistical characteristics of the original light scattering signal in the baseline section without sample elution, the measurement uncertainty of light scattering intensity at each data point of the original light scattering signal is evaluated, and a light scattering signal uncertainty sequence is formed. S3. Analyze the correlation between uncertainty sequences to obtain the signal uncertainty correlation matrix, including: Align the concentration signal uncertainty sequence with the light scattering signal uncertainty sequence on the time axis; Calculate the degree of matching between the fluctuation patterns of the concentration signal uncertainty sequence and the fluctuation patterns of the light scattering signal uncertainty sequence at different time offsets; Based on the maximum matching degree and its corresponding time offset, the correlation strength and association pattern between the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence are determined. Based on the determined correlation strength and association pattern, a correlation matrix of signal uncertainty is generated; S4. Based on the signal uncertainty correlation matrix, determine the contribution of different elution time periods to the uncertainty of molecular weight distribution, and synthesize the uncertainty envelope of molecular weight distribution. S5. Compare the uncertainty envelope of the molecular weight distribution with the expected elution volume-molecular weight relationship of the lentinan sample in the coupled system. Based on the comparison results, classify the reliability of the uncertainty envelope in different elution volume ranges to obtain an uncertainty envelope with reliability classification information, including: Obtain the expected elution volume-molecular weight relationship curve for lentinan samples; Calculate the degree of deviation between the central trend line of the uncertainty envelope of the molecular weight distribution and the expected elution volume-molecular weight relationship curve in each elution volume interval; Based on the magnitude of the deviation, a corresponding confidence level is assigned to the uncertainty envelope of each elution volume range; The confidence level is combined with the uncertainty envelope of the molecular weight distribution as additional information to form an uncertainty envelope with confidence level information. S6. Based on the uncertainty envelope with confidence level information, output the molecular weight distribution fitting spectrum of the shiitake polysaccharide sample.

2. The intelligent fitting method for the molecular weight distribution of lentinan according to claim 1, characterized in that, S1 includes: Simultaneously acquire the concentration simulation signal output by the differential refractive index detector and the multi-angle light scattering simulation signal output by the multi-angle laser light scattering detector; The concentration simulation signal and the multi-angle light scattering simulation signal were converted into digital signals, which were used as the original concentration signal and the original light scattering signal, respectively.

3. The intelligent fitting method for the molecular weight distribution of lentinan according to claim 2, characterized in that, Converting the concentration analog signal and the multi-angle light scattering analog signal into digital signals includes: performing synchronous analog-to-digital conversion on the concentration analog signal and the multi-angle light scattering analog signal at a sampling rate of not less than twice the highest signal frequency of the detector.

4. The intelligent fitting method for molecular weight distribution of lentinan according to claim 1, characterized in that, S4 include: Based on the signal uncertainty correlation matrix, the entire chromatographic elution process is divided into multiple consecutive elution time periods; For each elution time period, the contribution weight of the elution time period to the molecular weight distribution uncertainty is calculated by combining the values ​​of the concentration signal uncertainty sequence and the light scattering signal uncertainty sequence within the elution time period. Based on the contribution weight of each elution time period, the global uncertainty distribution is mapped to the corresponding molecular weight range; The uncertainty envelope of the molecular weight distribution is synthesized by superimposing the uncertainty distributions obtained from mapping across all molecular weight ranges.

5. The intelligent fitting method for the molecular weight distribution of lentinan according to claim 1, characterized in that, Obtaining the expected elution volume-molecular weight relationship curve for lentinan samples involves: analyzing the elution behavior of lentinan standards with known molecular weights in the coupled system, and establishing a linear calibration curve of elution volume versus the logarithm of molecular weight.

6. The intelligent fitting method for molecular weight distribution of lentinan according to claim 1, characterized in that, S6 include: The central trend line of molecular weight distribution is determined based on the uncertainty envelope with confidence level information. The uncertainty envelope with confidence level information is combined with the central trend line of the molecular weight distribution in the form of a boundary line; Based on the credibility rating information, the boundary lines are visually labeled; Generate and output a molecular weight distribution fitting map of lentinan samples, including a central trend line of molecular weight distribution, boundary lines with visual annotations, and a legend.

7. A smart fitting system for the molecular weight distribution of lentinan, used to implement the smart fitting method for the molecular weight distribution of lentinan according to any one of claims 1-6, characterized in that, include: The signal acquisition module is used to acquire the original concentration signal and original light scattering signal output by the gel permeation chromatography coupled with multi-angle laser light scattering and differential refractive index detector system when analyzing lentinan samples; The evaluation and processing module is used to evaluate the uncertainty of the original concentration signal and the original light scattering signal respectively, and obtain the corresponding uncertainty sequence. The correlation analysis module is used to analyze the correlation between uncertainty sequences and obtain the signal uncertainty correlation matrix; The envelope synthesis module is used to determine the contribution of different elution time periods to the uncertainty of molecular weight distribution based on the signal uncertainty correlation matrix, and to synthesize the uncertainty envelope of molecular weight distribution. The grading module is used to compare the uncertainty envelope of molecular weight distribution with the expected elution volume-molecular weight relationship of lentinan sample in the coupled system. Based on the comparison results, the confidence of the uncertainty envelope in different elution volume ranges is graded to obtain uncertainty envelope with confidence grading information. The spectrum output module is used to output a molecular weight distribution fitting spectrum of shiitake polysaccharide samples based on the uncertainty envelope with confidence level information.

Images