Peak signal fitting method, device, equipment and medium
By automatically selecting the fitting model and adjusting the peak parameters one by one, the problem of low peak signal fitting efficiency in the existing technology is solved, and more efficient and accurate peak signal fitting is achieved.
Patent Information
- Application Number
- CN202610221915.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-25
- Publication Date
- 2026-03-24
AI Technical Summary
In existing technologies, peak signal fitting requires constant manual adjustment of parameters, resulting in low efficiency.
By automatically selecting the fitting model and adjusting the peak parameters one by one, and by using preset adjustment step size and residual sum of squares change value, the fitting process is optimized to achieve automated fitting.
It improves the efficiency of peak signal fitting, reduces the influence of subjective human factors, and enhances fitting speed and accuracy.
Smart Images

Figure CN121723154A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, specifically providing a peak signal fitting method, apparatus, device, and medium. Background Technology
[0002] In fields such as catalyst activity analysis, metal dispersion measurement, and material adsorption capacity assessment, chemisorption analyzers are used to analyze the chemical properties and catalytic performance of material surfaces. Their core functions include temperature-programmed reduction, temperature-programmed oxidation, and temperature-programmed desorption. The data processing typically involves acquiring peak signals, analyzing these signals, and then obtaining the analytical results.
[0003] In existing technologies, after fitting the peak signal, key parameters such as the peak signal's position, area, and half-width at half-maximum are identified. Then, based on these parameters, quantitative analysis of the material's surface characteristics is performed. Currently, operators select a fitting model based on experience, adjust parameters to obtain fitted data, and then calculate the residuals based on the fitted data and peak signals. This process is repeated until the residuals remain essentially constant before parameter adjustments are stopped. This parameter adjustment process requires continuous manual adjustments and consumes a significant amount of time calculating the fitted data, resulting in low efficiency in fitting peak data.
[0004] Accordingly, there is a need in this field for a new peak signal fitting method to solve the above problems. Summary of the Invention
[0005] In order to overcome the above-mentioned defects, this application is made to provide a solution or at least a partial solution to the technical problem in the prior art that in the process of fitting peak signals, it is necessary to manually adjust parameters continuously and spend a lot of time to calculate fitting data, resulting in low efficiency of fitting peak data.
[0006] Firstly, this application provides a peak signal fitting method, including:
[0007] The range of peak signals and peak parameters is obtained; wherein the peak signals are acquired by a chemisorption analyzer.
[0008] Based on the peak signal, a fitting model corresponding to the peak signal is obtained;
[0009] For one of the peak parameters, in any round, the nth adjustment step size of the nth peak parameter is obtained according to the range of the peak parameter;
[0010] Based on the nth adjustment step size, the (n-1)th peak parameter is adjusted to obtain the nth peak parameter;
[0011] Substitute the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal;
[0012] Based on the nth fitted signal and the peak signal, the nth residual sum of squares is obtained;
[0013] The change value of the residual sum of squares is obtained by comparing the nth residual sum of squares with the (n-1)th residual sum of squares;
[0014] Wherein, 1≤n≤N, and N corresponds to the round when the change value of the residual sum of squares is less than the preset change value, or N is greater than the preset round, or the Nth adjustment step size is less than the preset step size.
[0015] In one technical solution of the above-mentioned peak signal fitting method, after substituting the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal, the method further includes:
[0016] Based on the nth fitted signal, the fitting determination coefficient, peak area recovery rate, and peak parameter range are obtained;
[0017] The fitting determination coefficient is determined to be less than the preset fitting determination coefficient, and / or the peak area recovery rate is not within the preset peak area recovery rate range, and / or the peak parameter range does not meet the preset constraints;
[0018] The nth adjustment step size is adjusted in the opposite direction to obtain the reverse adjustment step size;
[0019] Based on the reverse adjustment step size, the nth peak parameter is adjusted to obtain the (n+1)th peak parameter.
[0020] In one technical solution of the above-mentioned peak signal fitting method, obtaining the peak signal includes:
[0021] Acquire the raw signal;
[0022] The original signal is denoised to obtain a denoised signal;
[0023] Remove the spurious peaks from the denoised signal to obtain the true peak signal;
[0024] The boundary of the real peak signal is obtained by performing boundary localization on the real peak signal;
[0025] Based on the peak boundaries of the real peak signal, the overlapping peaks in the real peak signal are obtained;
[0026] The overlapping peaks are split to obtain multiple independent peak signals;
[0027] The peak signal includes the plurality of independent peak signals and the plurality of peak signals in the real peak signal excluding overlapping peaks.
[0028] In one technical solution of the above-mentioned peak signal fitting method, obtaining a fitting model corresponding to the peak signal based on the peak signal includes:
[0029] Feature extraction is performed on the peak signal to obtain peak shape features;
[0030] Based on the peak shape characteristics, the peak shape type of the peak signal is obtained;
[0031] Based on the peak shape type, a fitting model corresponding to the peak shape type is obtained.
[0032] In one technical solution of the above-mentioned peak signal fitting method, the step of denoising the original signal to obtain a denoised signal includes:
[0033] Based on the original signal, the baseline drift type is obtained;
[0034] Based on the baseline drift type, a denoising model corresponding to the baseline drift type is obtained;
[0035] Based on the denoising model, the baseline drift of the original signal is removed to obtain the denoised signal.
[0036] In one technical solution of the above-mentioned peak signal fitting method, removing spurious peaks from the denoised signal to obtain the true peak signal includes:
[0037] Obtain the second derivative of all points in the denoised signal;
[0038] The second derivatives of all points that are less than a preset second derivative threshold are filtered to obtain a set of second derivatives;
[0039] Based on the set of second derivatives, the actual peak signals corresponding to the second derivatives in the set of second derivatives are obtained.
[0040] In one technical solution of the above-mentioned peak signal fitting method, the step of splitting the overlapping peaks to obtain multiple independent peak signals includes:
[0041] The overlapping peaks are decomposed to obtain multiple intrinsic mode functions;
[0042] Continuous wavelet transforms are performed on the multiple intrinsic mode functions to obtain multiple independent peak signals.
[0043] Secondly, this application provides a peak signal fitting device, comprising:
[0044] An acquisition module is used to acquire the range of peak signals and peak parameters; wherein, the peak signals are acquired by a chemisorption analyzer.
[0045] The processing module is used to obtain a fitting model corresponding to the peak signal based on the peak signal;
[0046] The fitting module is used to obtain the nth adjustment step size of the nth peak parameter in any round, based on the range of the peak parameter, for one of the peak parameters.
[0047] The fitting module is also used to adjust the (n-1)th peak parameter according to the nth adjustment step size to obtain the nth peak parameter;
[0048] The fitting module is also used to substitute the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal;
[0049] The fitting module is further configured to obtain the nth residual sum of squares based on the nth fitted signal and the peak signal;
[0050] The fitting module is also used to obtain the change value of the residual sum of squares based on the nth residual sum of squares and the (n-1)th residual sum of squares;
[0051] Wherein, 1≤n≤N, and N corresponds to the round when the change value of the residual sum of squares is less than the preset change value, or N is greater than the preset round, or the Nth adjustment step size is less than the preset step size.
[0052] Thirdly, this application provides a peak signal fitting device, including at least one processor and a memory; wherein,
[0053] The memory stores computer-executed instructions;
[0054] The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method described in any one of the first aspects.
[0055] Fourthly, this application provides a computer-readable storage medium storing a plurality of program codes adapted to be loaded and run by a processor to perform the method described in any one of the first aspects.
[0056] This application provides a peak signal fitting method, apparatus, device, and medium. The method specifically includes: acquiring the peak signal and the range of peak parameters; wherein the peak signal is acquired by a chemisorption analyzer; obtaining a fitting model corresponding to the peak signal based on the peak signal; for one parameter in the peak parameters, in any round, obtaining the nth adjustment step size of the nth peak parameter based on the range of the peak parameters; adjusting the (n-1)th peak parameter according to the nth adjustment step size to obtain the nth peak parameter; substituting the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal; obtaining the nth residual sum of squares based on the nth fitted signal and the peak signal; obtaining the change value of the residual sum of squares based on the nth residual sum of squares and the (n-1)th residual sum of squares; wherein 1≤n≤N, and N corresponds to the round when the change value of the residual sum of squares is less than a preset change value, or N is greater than a preset round, or the Nth adjustment step size is less than a preset step size, thereby improving the fitting efficiency of the peak signal. Attached Figure Description
[0057] The disclosure of this application will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Furthermore, similar numbers in the drawings are used to denote similar components, wherein:
[0058] Figure 1 This is a flowchart illustrating a peak signal fitting method according to an embodiment of this application.
[0059] Figure 2 This is a flowchart illustrating a second embodiment of a peak signal fitting method provided in this application.
[0060] Figure 3 This is a flowchart illustrating a third embodiment of a peak signal fitting method provided in this application.
[0061] Figure 4 This is a flowchart illustrating a fourth embodiment of a peak signal fitting method provided in this application.
[0062] Figure 5 This is a flowchart illustrating a fifth embodiment of a peak signal fitting method provided in this application.
[0063] Figure 6 This is a flowchart illustrating a sixth embodiment of a peak signal fitting method provided in this application.
[0064] Figure 7 This is a flowchart illustrating a seventh embodiment of a peak signal fitting method provided in this application.
[0065] Figure 8 This is a schematic diagram of the structure of a peak signal fitting device provided in an embodiment of this application;
[0066] Figure 9 This is a schematic diagram of the structure of a peak signal fitting device provided in an embodiment of this application;
[0067] Figure 10 This is a schematic diagram of a peak signal fitting system provided in an embodiment of this application.
[0068] List of reference numerals in the attached diagram:
[0069] 11: Acquisition module; 12: Processing module; 13: Fitting module; 21: Processor; 22: Memory; 31: Chemisorption instrument; 32: Computer. Detailed Implementation
[0070] Some embodiments of this application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of this application and are not intended to limit the scope of protection of this application.
[0071] In the description of this application, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and can also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "a" or "this" can also include plural forms.
[0072] In existing technologies, after collecting peak signals using a chemisorption analyzer, operators select a fitting model based on experience, then obtain fitting data by adjusting parameters, and obtain residuals based on the fitting data and peak signals. This process is repeated until the residuals remain essentially unchanged, at which point parameter adjustment is stopped. During the parameter adjustment process, manual adjustments are required continuously, and a significant amount of time is spent calculating the fitting data, resulting in the technical problem of low efficiency in fitting peak signals.
[0073] Based on this, in order to solve the above-mentioned technical problems, the technical concept of this application is to provide a new peak signal fitting method to improve the fitting efficiency of peak signals.
[0074] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.
[0075] Figure 1 This is a flowchart illustrating an embodiment of a peak signal fitting method provided in this application. Figure 1 As shown, specifically, the method includes:
[0076] Step S101: Obtain the range of peak signal and peak parameters.
[0077] In this embodiment, the peak signal is acquired by a chemisorption analyzer. Furthermore, the peak parameters associated with the peak signal include peak position, peak height, peak width, and mixing type coefficient.
[0078] The x-coordinate of the peak apex corresponds to the peak position. The peak height is the y-coordinate of the peak apex. The peak width is the width at half the peak height. The ranges for each peak parameter are as follows: the peak position range should conform to the known adsorbate desorption rules; the peak height range is greater than 0 and less than 1.2 times the initial peak height; the peak width range is greater than 0 and less than 3 times the initial peak width; and the mixing type coefficient range is greater than 0 and less than 1.
[0079] Step S102: Based on the peak signal, obtain the fitting model corresponding to the peak signal.
[0080] In this embodiment, through a large number of experiments, the correspondence between the peak signal and the fitting model is determined, and the correspondence is stored in the memory of the chemisorption instrument.
[0081] In this embodiment, for example, the fitting model includes Gaussian model, Lorentz model, Voigt model, and combinations of multiple models.
[0082] Step S103: For one of the peak parameters, in any round, based on the range of the peak parameters, obtain the nth adjustment step size of the nth peak parameter.
[0083] In this embodiment, the four peak parameters are adjusted one by one according to a preset peak parameter adjustment priority. Furthermore, for any given peak parameter, its range is redefined based on the maximum value of its range and its current value. This redefined range is then divided into 10 equal parts to obtain the nth adjustment step size. Specifically, the mixed-type coefficient among the peak parameters only needs adjustment when the fitted model is a Voigt model.
[0084] Preferably, the adjustment step size for peak height and peak width is 3% of the corresponding initial value. The adjustment step size for peak position and mixing type coefficient is 1% of the corresponding initial value.
[0085] The preset peak parameter adjustment priority is based on the order of the peak parameters' influence on the fitting effect in different fitting models. For example, for a Gaussian model, the preset peak parameter adjustment priority is peak height, peak width, peak position, and mixture type coefficient.
[0086] In this embodiment, during the first adjustment, the first peak parameter is the initial value of the peak parameter, and the first adjustment step size is a preset initial step size. Specifically, the initial value of the peak position is the abscissa corresponding to the measured peak apex of the peak signal; the initial value of the peak height is the ordinate corresponding to the measured peak apex of the peak signal; the initial value of the half-peak width is the peak width at half the measured peak height of the peak signal; if the fitting model is only the Voigt model, the initial value of the mixing coefficient is 0.5; if the fitting model is only the Gaussian model, the initial value of the mixing coefficient is 0; and if the fitting model is only the Lorentz model, the initial value of the mixing coefficient is 1.
[0087] Step S104: Adjust the (n-1)th peak parameter according to the nth adjustment step size to obtain the nth peak parameter.
[0088] In this embodiment, the (n-1)th peak parameter is added to the nth adjustment step size to obtain the nth peak parameter.
[0089] Step S105: Substitute the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal.
[0090] In this embodiment, for peak parameters that have not been adjusted, the initial value of the unadjusted peak parameter is substituted into the fitting model; for peak parameters that have been adjusted, the value of the adjusted peak parameter is substituted into the fitting model to obtain the nth fitted signal.
[0091] Step S106: Obtain the nth residual sum of squares based on the nth fitted signal and the peak signal.
[0092] In this embodiment, the nth fitted signal and the peak signal have the same number of discrete points. The residuals of the discrete points at the same time are calculated, and then the residuals of all points are added together to obtain the nth residual sum of squares.
[0093] In this embodiment, according to formula 1:
[0094] (1)
[0095] Obtain the nth residual sum of squares SSR n Where n is the number of rounds, m is the number of discrete points in the nth fitted signal, and y i Y is the value of the i-th peak signal of the n-th fitted signal.i Let be the value of the i-th peak signal.
[0096] In this embodiment, the preset peak parameter adjustment priority is based on the percentage contribution of each peak parameter to the residual sum of squares. Specifically, peak height is greater than 40%, half-maximum width is approximately 30%, peak position is approximately 20%, and the mixed-type coefficient is approximately 10%. Adjusting the peak parameters according to this preset priority can avoid multi-parameter oscillations and improve convergence efficiency.
[0097] In this embodiment, the step size can be adaptively adjusted according to the change of the residual sum of squares, so as to make the parameters quickly approach the optimal value and avoid local optima.
[0098] Step S107: Obtain the change value of the residual sum of squares based on the nth residual sum of squares and the (n-1)th residual sum of squares.
[0099] In this embodiment, according to formula 2:
[0100] (2)
[0101] The change in the sum of squared residuals, ΔSSR, is obtained. Where SSR... n SSR is the sum of squared residuals of the nth time. n-1 It is the sum of squares of the (n-1)th residual.
[0102] Where 1≤n≤N, and N corresponds to the time when the change in the sum of squared residuals is less than a preset change value, or N is greater than a preset number of rounds, or the Nth adjustment step size is less than a preset step size, the iteration stops. That is, by adjusting the Nth peak parameter, the Nth fitted signal remains basically unchanged, or N has reached 1000, or the Nth adjustment step size is 0.0001, at this point, the adjustment of the peak parameter stops.
[0103] In this embodiment, all peak parameters are adjusted according to steps S103 to S107. Once the last peak parameter is adjusted, the resulting fitted signal is the fitted signal for that peak signal.
[0104] In this embodiment, after acquiring the peak signal collected by the chemisorption analyzer, the fitting model corresponding to the peak signal is automatically matched from the correspondence between the peak signal and the fitting model. Then, each parameter in the peak parameters is adjusted one by one according to the adjustment step size. When the change value of the residual sum of squares is less than the preset change value, or N is greater than the preset number of rounds, or the Nth adjustment step size is less than the preset step size, the adjustment of the parameter is stopped, and the value of the parameter is obtained. After the last peak parameter is adjusted, the fitting signal of the peak parameter is obtained. Compared with the prior art, which requires manual selection of fitting model and adjustment parameters based on experience, this application automatically selects the fitting model and automatically adjusts the value of the peak parameter, thereby improving the fitting speed and avoiding errors in model selection and adjustment parameters caused by human subjective factors, thus improving the fitting efficiency of the peak signal.
[0105] Figure 2 This is a flowchart illustrating a second embodiment of a peak signal fitting method provided in this application. Based on the above embodiments, as... Figure 2 As shown, specifically, after step S105, the method further includes:
[0106] Step S201: Based on the nth fitted signal, obtain the fitting determination coefficient, peak area recovery rate, and peak parameter range.
[0107] In this embodiment, the coefficient of determination reflects the ability of the fitted curve to interpret the peak signal. Its value ranges from greater than 0 to less than 1. The closer the coefficient of determination is to 1, the better the fitting effect.
[0108] According to formula 3:
[0109] (3)
[0110] Obtain the fitting determination coefficient R. 2 Among them, SSR n Y is the sum of squares of the nth residual, m is the number of discrete points in the nth fitted signal, and Y is the sum of squares of the nth residual. i Let be the value of the i-th peak signal, and M be the average value of the peak signals.
[0111] Peak area typically corresponds to the total amount of adsorbate. Peak area recovery is the ratio of the area of the fitted curve to the area of the peak signal. Peak parameter range includes the range of peak position, peak area, and peak intensity distribution.
[0112] Step S202: Determine that the fitting determination coefficient is less than the preset fitting determination coefficient, and / or the peak area recovery rate is not within the preset peak area recovery rate range, and / or the peak parameter range does not meet the preset constraints.
[0113] The peak position must conform to the known adsorbate desorption rules; the peak area must be greater than 0 and less than the maximum adsorption capacity of the test material; the peak intensity distribution must conform to the distribution rules of active sites, where the quotient of the peak height and peak area of multiple peaks is consistent with the distribution rules of active sites.
[0114] In this embodiment, if at least one of the three conditions is met, then step S203 is executed.
[0115] Step S203: Adjust the nth adjustment step size in the reverse direction to obtain the reverse adjustment step size.
[0116] In this embodiment, the nth adjustment step size needs to be reduced to obtain the reverse adjustment step size.
[0117] For example, according to Formula 4:
[0118] (4)
[0119] The reverse adjustment step size Rstep is obtained. Where step... n This is the nth adjustment step size.
[0120] Step S204: Adjust the nth peak parameter according to the reverse adjustment step size to obtain the (n+1)th peak parameter.
[0121] By adjusting the step size in the reverse direction and adding the parameter of the nth peak, we obtain the parameter of the (n+1)th peak.
[0122] In this embodiment, based on the nth fitted signal, the fitting determination coefficient, peak area recovery rate, and peak parameter range are obtained; it is determined that the fitting determination coefficient is less than the preset fitting determination coefficient, and / or the peak area recovery rate is not within the preset peak area recovery rate range, and / or the peak parameter range does not meet the preset constraints; the nth adjustment step size is adjusted in reverse to obtain the reverse adjustment step size; based on the reverse adjustment step size, the nth peak parameter is adjusted to obtain the (n+1)th peak parameter. During the peak parameter adjustment process, if the adjusted peak parameter does not meet the constraints, the peak parameter is adjusted according to the reverse adjustment step size to improve the accuracy of parameter adjustment, thereby improving the accuracy of the fitted signal.
[0123] Figure 3 This is a flowchart illustrating a third embodiment of a peak signal fitting method provided in this application. Based on the above embodiments, as... Figure 3 As shown, specifically, in step S101, acquiring the peak signal includes:
[0124] Step S301: Acquire the raw signal.
[0125] In this embodiment, after the chemisorption instrument starts running, a chemisorption experiment occurs in the chemisorption instrument. Then, the detector measures the change in gas thermal conductivity or ion current and outputs an electrical signal proportional to the gas concentration. This electrical signal forms a peak curve, which is the original signal.
[0126] Step S302: Denoise the original signal to obtain a denoised signal.
[0127] In this embodiment, the noise in the original signal is mainly baseline drift. Therefore, the baseline drift in the original signal is removed to obtain a denoised signal.
[0128] Step S303: Remove the spurious peaks from the denoised signal to obtain the true peak signal.
[0129] In this embodiment, the peak of the denoised signal is detected, and peaks with peak values lower than a preset peak value are removed to obtain the true peak signal.
[0130] Step S304: Perform boundary localization on the real peak signal to obtain the peak boundary of the real peak signal.
[0131] In this embodiment, the boundary of the true peak signal can be determined by the zero point of the first derivative. Specifically, the starting boundary of the true peak signal corresponds to the zero point where the first derivative changes from negative to positive, and the ending boundary of the true peak signal corresponds to the zero point where the first derivative changes from positive to negative. That is, by solving for the first derivative at each point in the true peak signal, if the values of three consecutive first derivatives are less than 0, equal to 0, and greater than 0, then the point where the first derivative is 0 is the starting boundary of the true peak signal. If the values of three consecutive first derivatives are greater than 0, equal to 0, and less than 0, then the point where the first derivative is 0 is the ending boundary of the true peak signal.
[0132] Step S305: Obtain the overlapping peaks in the real peak signal based on the peak boundaries of the real peak signal.
[0133] In this embodiment, after determining the peak boundary of the true peak signal, the range of each peak signal in the true peak signal can be determined based on the peak boundary. Then, for each peak signal, the second derivative is calculated. If the second derivative of a peak signal has multiple extreme points, then the peak signal is an overlapping peak.
[0134] Step S306: The overlapping peaks are split to obtain multiple independent peak signals.
[0135] In this embodiment, cluster analysis can be used to split overlapping peaks, that is, to divide the data with the same characteristics in overlapping peaks, thereby obtaining multiple independent peak signals.
[0136] In this embodiment, the peak signal includes multiple independent peak signals and multiple peak signals other than overlapping peaks in the real peak signal.
[0137] In this embodiment, the original signal is acquired; the original signal is denoised to obtain a denoised signal; spurious peaks in the denoised signal are removed to obtain a true peak signal; the true peak signal is boundary located to obtain the peak boundary of the true peak signal; overlapping peaks in the true peak signal are obtained based on the peak boundary of the true peak signal; and overlapping peaks are split to obtain multiple independent peak signals, thereby achieving accurate splitting of overlapping peaks.
[0138] Figure 4 This is a flowchart illustrating a fourth embodiment of a peak signal fitting method provided in this application. Based on the above embodiments, as... Figure 4 As shown, specifically, one embodiment of step S102 includes:
[0139] Step S401: Extract features from the peak signal to obtain peak shape features.
[0140] In this embodiment, the peak shape characteristics of the peak signal include peak symmetry factor, full width at half maximum (FWHM), and peak height.
[0141] Step S402: Based on the peak shape characteristics, obtain the peak shape type of the peak signal.
[0142] In this embodiment, if the peak symmetry factor of the peak signal is equal to 1, then the peak shape type of the peak signal is a symmetrical peak; otherwise, it is an asymmetrical peak. If the ratio of the half-peak width to the peak height is less than or equal to 0.1, then the peak shape type of the peak signal is a narrow peak. If the ratio of the half-peak width to the peak height is greater than or equal to 0.3, then the peak shape type of the peak signal is a wide peak. If the ratio of the half-peak width to the peak height of a peak signal is greater than 0.1 and less than 0.3, and the peak shape of the peak signal is broadened as a whole and the outline is not smooth, then the peak shape type of the peak signal is a multi-peak superposition.
[0143] Step S403: Based on the peak shape type, obtain the fitting model corresponding to the peak shape type.
[0144] In this embodiment, there are preset peak types and corresponding fitting models, as shown in Table 1.
[0145] Table 1. Correspondence between peak shape type and fitting model
[0146] Peak type Fitting Model Symmetrical peak Gaussian model Narrow peak Lorentz model asymmetric peak Voigt Model Multi-peak superposition Hybrid model (Gaussian model + Lorentz model + Voigt model) Broad Peak Double Gaussian model / Double Lorentz model
[0147] Regarding Table 1, it should be noted that for symmetrical peaks, due to weak adsorption, the adsorbate desorption rate is uniform, resulting in a symmetrical peak shape; therefore, the fitting model for symmetrical peaks is a Gaussian model. For narrow peaks, due to strong adsorption, the active sites are uniform, resulting in a sharp peak shape; therefore, the fitting model for narrow peaks is a Lorentz model. For asymmetric peaks, due to moderate adsorption strength, adsorbate diffusion is limited, resulting in an asymmetric peak shape; therefore, the fitting model for asymmetric peaks is a Voigt model. For multiple peak superpositions, due to the coexistence of multiple active sites and significant differences in adsorption strength at different sites, the fitting model for multiple peak superpositions is a hybrid model composed of Gaussian, Lorentz, and Voigt models. For broad peaks, a single active site exists, but the active site distribution is wide, requiring splitting into two sub-peaks to characterize the site distribution; therefore, the fitting model for broad peaks is a double Gaussian model or a double Lorentz model.
[0148] In this embodiment, feature extraction is performed on the peak signal to obtain peak shape features; based on the peak shape features, the peak shape type of the peak signal is obtained; based on the peak shape type, a fitting model corresponding to the peak shape type is obtained, thereby achieving accurate matching between the peak signal and the fitting model, and thus improving the accuracy of the fitted signal.
[0149] Figure 5 This is a flowchart illustrating a fifth embodiment of a peak signal fitting method provided in this application. Based on the above embodiments, as... Figure 5 As shown, specifically, the implementation of step S302 includes:
[0150] Step S501: Obtain the baseline drift type based on the original signal.
[0151] If the baseline changes in a sloping manner and the original signal moves slowly over time, then the baseline drift type contained in the original signal is linear drift. If the baseline exhibits curvature or fluctuation, then the baseline drift type contained in the original signal is nonlinear drift.
[0152] Step S502: Based on the baseline drift type, obtain the denoising model corresponding to the baseline drift type.
[0153] In this embodiment, if the baseline drift type is linear drift, the denoising model is a polynomial fitting. If the baseline drift type is nonlinear drift, the denoising model is a locally weighted regression.
[0154] Step S503: Based on the denoising model, remove the baseline drift of the original signal to obtain the denoised signal.
[0155] In this embodiment, if polynomial fitting is used, the order of the polynomial is selected. For example, the order 1 or 2 can be selected. Then, the least squares method is used to fit the order 1 or 2 polynomial with the original signal as the dependent variable and time as the independent variable to obtain a polynomial.
[0156] If local weighted regression is used, firstly, for each data point in the original signal, a neighboring point within a local window is selected as the center. Then, a weighting function is used to assign weights to the points within the window. Specifically, the closer the point is to the center point, the higher the weight is, in order to reduce the influence of distant points. Secondly, a low-order polynomial is fitted within the window using weighted least squares to obtain a local regression model. Finally, the weights are adjusted through multiple iterations to reduce the influence of outliers and make the results more robust.
[0157] In this embodiment, the baseline drift type is obtained based on the original signal; a denoising model corresponding to the baseline drift type is obtained based on the baseline drift type; the baseline drift of the original signal is removed based on the denoising model to obtain a denoised signal, thereby improving the denoising effect of the original signal.
[0158] Figure 6 This is a flowchart illustrating a sixth embodiment of a peak signal fitting method provided in this application. Based on the above embodiments, as... Figure 6 As shown, the specific implementation of step S303 includes:
[0159] Step S601: Obtain the second derivative of all points in the denoised signal.
[0160] In this embodiment, the denoised signal is a discrete signal containing multiple points. The second derivative of all points is obtained by taking the derivative twice for each point in the denoised signal.
[0161] Step S602: Filter the second derivatives of all points that are less than the preset second derivative threshold to obtain the set of second derivatives.
[0162] In this embodiment, if the second derivative at a certain point is less than 0, then that point is a peak. Furthermore, the larger the absolute value of the second derivative, the sharper the peak. If the second derivative at a certain point is greater than 0, then that point is a trough.
[0163] In this embodiment, the points whose second derivatives are less than a preset second derivative threshold constitute the second derivative set.
[0164] Step S603: Based on the set of second derivatives, obtain the actual peak signal corresponding to the second derivative in the set of second derivatives.
[0165] In this embodiment, the point corresponding to the set of second derivatives is a significant peak, and the peak signal corresponding to the significant peak is the true peak signal.
[0166] In this embodiment, the second derivatives of all points in the denoised signal are obtained; the second derivatives of all points that are less than a preset second derivative threshold are filtered to obtain a set of second derivatives; based on the set of second derivatives, the real peak signal corresponding to the second derivatives in the set of second derivatives is obtained, thereby suppressing the false peaks caused by noise.
[0167] Figure 7 This is a flowchart illustrating a seventh embodiment of a peak signal fitting method provided in this application. Based on the above embodiments, as... Figure 7 As shown, the specific implementation of step S306 includes:
[0168] Step S701: Decompose the overlapping peaks to obtain multiple intrinsic mode functions.
[0169] In this embodiment, the overlapping peaks are first mirrored to reduce boundary effects. Then, the number of modes, penalty factor, and convergence criterion are set, and the center frequency of each mode is initialized. In each iteration, the spectrum of each mode is updated through a Wiener filter, and the center frequency is updated according to the mode energy distribution to ensure that the mode bandwidth is minimized. Then, the Lagrange multipliers are adjusted to force the constraint to be satisfied. The iteration is repeated until the mode spectrum change is less than a preset threshold or the maximum number of iterations is reached. Finally, multiple intrinsic mode functions are obtained.
[0170] Step S702: Perform continuous wavelet transform on multiple intrinsic mode functions to obtain multiple independent peak signals.
[0171] In this embodiment, multiple independent peak signals are obtained by performing continuous wavelet transforms on multiple intrinsic mode functions through multi-scale enhancement.
[0172] In this embodiment, the overlapping peaks are decomposed to obtain multiple intrinsic mode functions; continuous wavelet transforms are performed on the multiple intrinsic mode functions to obtain multiple independent peak signals, thereby achieving accurate separation of strongly overlapping peaks.
[0173] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of this application, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the scope of protection of this application.
[0174] Furthermore, this application also provides a peak signal fitting device.
[0175] Figure 8 This is a schematic diagram of a peak signal fitting device provided in an embodiment of this application. Figure 8As shown, the peak signal fitting device in this embodiment mainly includes an acquisition module 11, a processing module 12, and a fitting module 13. In some embodiments, one or more of the acquisition module 11, processing module 12, and fitting module 13 can be combined into a single module. In some embodiments, the acquisition module 11 can be configured to acquire the peak signal and the range of peak parameters; wherein the peak signal is acquired by a chemisorption analyzer. The processing module 12 can be configured to obtain a fitting model corresponding to the peak signal based on the peak signal. The fitting module 13 can be configured to, for one parameter of the peak parameters, obtain the nth adjustment step size of the nth peak parameter in any round based on the range of the peak parameters. The fitting module 13 can also be configured to adjust the (n-1)th peak parameter according to the nth adjustment step size to obtain the nth peak parameter; the fitting module 13 can also be configured to substitute the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal. The fitting module 13 can also be configured to obtain the nth residual sum of squares based on the nth fitted signal and the peak signal. The fitting module 13 can also be configured to obtain the change value of the residual sum of squares based on the nth residual sum of squares and the (n-1)th residual sum of squares. Wherein, 1≤n≤N, and N corresponds to the round when the change value of the residual sum of squares is less than a preset change value, or N is greater than a preset round, or the Nth adjustment step size is less than a preset step size.
[0176] The above-mentioned peak signal fitting device is used to perform Figures 1 to 7 The peak signal fitting method embodiments shown are similar in technical principle, the technical problems solved, and the technical effects produced. Those skilled in the art can clearly understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the peak signal fitting device can be referred to the content described in the embodiments of the peak signal fitting method, and will not be repeated here.
[0177] Those skilled in the art will understand that all or part of the processes in the method of the above-described embodiment can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.
[0178] Furthermore, this application also provides a peak signal fitting device.
[0179] Figure 9 This is a schematic diagram of a peak signal fitting device provided in an embodiment of this application. Figure 9 As shown, the peak signal fitting device includes a processor 21 and a memory 22. The memory 22 can be configured to store a program for executing the peak signal fitting method of the above-described method embodiments, and the processor 21 can be configured to execute the program stored in the memory. This program includes, but is not limited to, a program for executing the peak signal fitting method of the above-described method embodiments. For ease of explanation, only the parts related to the embodiments of this application are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of this application. The peak signal fitting device can be a control device device comprising various electronic devices.
[0180] Furthermore, this application also provides a peak signal fitting system.
[0181] Figure 10 This is a schematic diagram of a peak signal fitting system provided in an embodiment of this application. Figure 10 As shown, the peak signal fitting system includes a chemisorption analyzer 31 and a computer 32. The chemisorption analyzer 31 is used to collect peak signals and then send the peak signals to the computer 32. The computer 32 performs fitting processing on the peak signals to obtain the fitted signal.
[0182] Furthermore, this application also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to this application, the computer-readable storage medium can be configured to store a program that performs the peak signal fitting method of the above-described method embodiments. This program can be loaded and run by a processor to implement the peak signal fitting method. For ease of explanation, only the parts related to the embodiments of this application are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of this application. The computer-readable storage medium can be a storage device device comprising various electronic devices. Optionally, in the embodiments of this application, the computer-readable storage medium is a non-transitory computer-readable storage medium.
[0183] Furthermore, it should be understood that since the various modules are only provided to illustrate the functional units of the device described in this application, the physical devices corresponding to these modules may be the processor itself, or a part of the processor's software, hardware, or a combination of both. Therefore, the number of modules shown in the figures is merely illustrative.
[0184] Those skilled in the art will understand that the various modules in the device can be adaptively split or combined. Such splitting or combining of specific modules will not cause the technical solution to deviate from the principles of this application; therefore, the technical solutions after splitting or combining will fall within the protection scope of this application.
[0185] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. Without departing from the principles of this application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of this application.
Claims
1. A peak signal fitting method, characterized in that, include: The range of peak signals and peak parameters is obtained; wherein the peak signals are acquired by a chemisorption analyzer. Based on the peak signal, a fitting model corresponding to the peak signal is obtained; For one of the peak parameters, in any round, the nth adjustment step size of the nth peak parameter is obtained according to the range of the peak parameter; Based on the nth adjustment step size, the (n-1)th peak parameter is adjusted to obtain the nth peak parameter; Substitute the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal; Based on the nth fitted signal and the peak signal, the nth residual sum of squares is obtained; The change in the sum of squares of the nth residual and the (n-1)th residual are obtained. Wherein, 1≤n≤N, and N corresponds to the round when the change value of the residual sum of squares is less than the preset change value, or N is greater than the preset round, or the Nth adjustment step size is less than the preset step size.
2. The method according to claim 1, characterized in that, After substituting the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal, the method further includes: Based on the nth fitted signal, the fitting determination coefficient, peak area recovery rate, and peak parameter range are obtained; The fitting determination coefficient is determined to be less than the preset fitting determination coefficient, and / or the peak area recovery rate is not within the preset peak area recovery rate range, and / or the peak parameter range does not meet the preset constraints; The nth adjustment step size is adjusted in the opposite direction to obtain the reverse adjustment step size; Based on the reverse adjustment step size, the nth peak parameter is adjusted to obtain the (n+1)th peak parameter.
3. The method according to claim 2, characterized in that, The acquisition of the peak signal includes: Acquire the raw signal; The original signal is denoised to obtain a denoised signal; Remove the spurious peaks from the denoised signal to obtain the true peak signal; The boundary of the real peak signal is obtained by performing boundary localization on the real peak signal; Based on the peak boundaries of the real peak signal, the overlapping peaks in the real peak signal are obtained; The overlapping peaks are split to obtain multiple independent peak signals; The peak signal includes the plurality of independent peak signals and the plurality of peak signals in the real peak signal excluding overlapping peaks.
4. The method according to claim 3, characterized in that, The step of obtaining a fitting model corresponding to the peak signal based on the peak signal includes: Feature extraction is performed on the peak signal to obtain peak shape features; Based on the peak shape characteristics, the peak shape type of the peak signal is obtained; Based on the peak shape type, a fitting model corresponding to the peak shape type is obtained.
5. The method according to claim 3 or 4, characterized in that, The denoising process of the original signal to obtain a denoised signal includes: Based on the original signal, the baseline drift type is obtained; Based on the baseline drift type, a denoising model corresponding to the baseline drift type is obtained; Based on the denoising model, the baseline drift of the original signal is removed to obtain the denoised signal.
6. The method according to claim 5, characterized in that, The step of removing spurious peaks from the denoised signal to obtain the true peak signal includes: Obtain the second derivative of all points in the denoised signal; The second derivatives of all points that are less than a preset second derivative threshold are filtered to obtain a set of second derivatives; Based on the set of second derivatives, the actual peak signals corresponding to the second derivatives in the set of second derivatives are obtained.
7. The method according to claim 6, characterized in that, The process of splitting the overlapping peaks to obtain multiple independent peak signals includes: The overlapping peaks are decomposed to obtain multiple intrinsic mode functions; Continuous wavelet transforms are performed on the multiple intrinsic mode functions to obtain multiple independent peak signals.
8. A peak signal fitting device, characterized in that, include: An acquisition module is used to acquire the range of peak signals and peak parameters; wherein, the peak signals are acquired by a chemisorption analyzer. The processing module is used to obtain a fitting model corresponding to the peak signal based on the peak signal; The fitting module is used to obtain the nth adjustment step size of the nth peak parameter in any round, based on the range of the peak parameter, for one of the peak parameters. The fitting module is also used to adjust the (n-1)th peak parameter according to the nth adjustment step size to obtain the nth peak parameter; The fitting module is also used to substitute the nth peak parameter and other peak parameters into the fitting model to obtain the nth fitted signal; The fitting module is further configured to obtain the nth residual sum of squares based on the nth fitted signal and the peak signal; The fitting module is also used to obtain the change value of the residual sum of squares based on the nth residual sum of squares and the (n-1)th residual sum of squares; Wherein, 1≤n≤N, and N corresponds to the round when the change value of the residual sum of squares is less than the preset change value, or N is greater than the preset round, or the Nth adjustment step size is less than the preset step size.
9. A peak signal fitting device, comprising at least one processor and a memory; wherein, The memory stores computer-executed instructions; The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method of any one of claims 1 to 7.
10. A computer-readable storage medium storing a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by a processor to perform the method of any one of claims 1 to 7.
Citation Information
Patent Citations
Phase modulation code rate estimation method based on peak position
CN106506414A
Gaussian multi-peak fitting-based spectrum analysis algorithm
CN107462535A
A trailing peak processing method and device
CN109697320A
Nuclear magnetic resonance spectrogram analysis method and system
CN121141733A
Methods of automated spectral peak detection and quantification without user input
US20100100336A1
