CN-121994734-A - Method for detecting activity quality of peptide milk powder based on wet production

CN121994734ACN 121994734 ACN121994734 ACN 121994734ACN-121994734-A

Abstract

The invention belongs to the technical field of data processing, and in particular relates to a method for detecting the activity quality of peptide milk powder based on wet production, which comprises the following steps of S1, obtaining fluid time sequence data of a production node and an original reflection spectrum sequence of a spray drying node, and performing smooth filtering operation on the original reflection spectrum sequence to obtain a smooth spectrum after standardized treatment; S3, constructing an operation index with causal logic based on background characteristics of the smooth spectrum and hydrodynamic characteristics of fluid time sequence data to finish deep correction operation on the apparent chemical concentration and output real activity retention concentration with concentration attribute and activity attribute. The invention effectively eliminates optical measurement errors and accurately evaluates the actual biological potency.

Inventors

DU LEI
Zhong Xuli
XU SHANWANG
LI BINGHUA
HU ZHI
SHI YIHUA

Assignees

陕西圣泉乳业科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260409

Claims (10)

1. The method for detecting the activity quality of the peptide milk powder based on the wet production is characterized by comprising the following steps of: s1, acquiring fluid time sequence data of a production node and an original reflection spectrum sequence of a spray drying node, and performing smoothing filter operation on the original reflection spectrum sequence to obtain a smoothed spectrum after standardized treatment; S2, introducing the smooth spectrum into a pre-configured partial least squares regression algorithm model, and calculating and outputting the apparent chemical concentration of the peptide component in the current batch of milk powder; s3, based on background characteristics of a smooth spectrum and hydrodynamic characteristics of fluid time series data, constructing an operation index with causal logic to finish deep correction operation on apparent chemical concentration, and outputting real activity retention concentration with concentration attribute and activity attribute; And S4, the industrial control host receives and records the real activity retention concentration value in real time, compares the real activity retention concentration value with a preset quality outline index base line to judge a detection result and execute closed-loop control of equipment.
2. The method for detecting the activity quality of the peptide milk powder based on the wet production according to claim 1, wherein the steps of obtaining the fluid time series data of the production node and the original reflection spectrum sequence of the spray drying node, performing a smoothing filter operation on the original reflection spectrum sequence to obtain a smoothed spectrum after the normalization processing, and include: collecting a fluid temperature time series and a fluid pressure time series at the heater pipeline and the high-pressure pump outlet valve; collecting an original reflection spectrum sequence of the peptide milk powder within a continuous wavelength range at a powder outlet of the fluidized bed; and calling a smoothing filter algorithm to perform first-order difference derivation and polynomial smoothing operation on the original reflection spectrum sequence to output a smoothed spectrum.
3. The method for detecting the activity quality of the peptide milk powder based on the wet production according to claim 1, wherein the step of introducing the smoothed spectrum into a pre-configured partial least squares regression algorithm model to calculate and output the apparent chemical concentration of the peptide component in the current batch of milk powder comprises the following steps: reducing the dimension of the high-dimension spectral absorbance matrix by a singular value decomposition technology and extracting mutually orthogonal latent variable features; Calculating a linear regression weight matrix between the latent variable characteristics and the peptide mass concentration; and calculating the smooth spectrum through a linear regression weight matrix to obtain the apparent chemical concentration of the peptide component in the current milk powder.
4. The method for detecting the activity quality of peptide milk powder based on wet production according to claim 1, wherein constructing an operation index with causal logic based on background characteristics of a smoothed spectrum and hydrodynamic characteristics of fluid time series data to complete a deep correction operation on apparent chemical concentration and output a true activity retention concentration with both concentration attribute and activity attribute comprises: the absorbance deviation absolute value in the non-absorption reference frequency band is extracted to construct a spectrum scattering drift factor, and the calculation relation is as follows: in the formula, Is a dimensionless spectrum scattering drift factor, Is the first in the preset frequency band The point of the wavelength of the light, For the total number of discrete wavelength points contained within the band, For the normalized smoothed spectral absorbance measured, Is ideal reference absorbance without agglomeration.
5. The method for detecting the activity quality of peptide milk powder based on wet production according to claim 4, wherein the construction of the operation index with causal logic to complete the deep correction operation of the apparent chemical concentration based on the background feature of the smoothed spectrum and the hydrodynamic feature of the fluid time series data, and the output of the true activity retention concentration with both concentration attribute and activity attribute, further comprises: Carrying out nonlinear integral fusion on a dynamic temperature variable and a dynamic pressure variable on a time scale to construct a composite stress index, wherein the calculated relational expression is as follows: in the formula, In the form of a composite stress index, As a total number of discrete time steps, In order to be able to achieve the actual fluid temperature, For the conformational de-rotation critical temperature threshold, As the absolute pressure of the fluid is, For a standard atmospheric calibration constant for eliminating the pressure dimension, Is the sampling time interval constant.
6. The method for detecting the activity quality of the peptide milk powder based on the wet production according to claim 5, wherein the construction of the operation index with causal logic to complete the deep correction operation of the apparent chemical concentration based on the background feature of the smooth spectrum and the hydrodynamic feature of the fluid time series data, and the output of the true activity retention concentration with both the concentration attribute and the activity attribute, further comprises: the coupling apparent chemical concentration, the spectrum scattering drift factor and the composite stress index are used for constructing the real activity retention concentration, and the calculated relational expression is as follows: in the formula, The concentration is preserved for the true activity and, In order to obtain the apparent chemical concentration, Compensating the gain constant for dimensionless optical amplification, Is a dimensionless molecular stress sensitivity loss constant, For the maximum allowable margin stress constant of the system, Is a dimensionless spectrum scattering drift factor, Is a composite stress index.
7. The method for detecting the activity quality of the peptide milk powder based on the wet production according to claim 1, wherein the industrial control host receives and records the true activity retention concentration value in real time, compares the true activity retention concentration value with a preset quality outline index base line to judge a detection result and execute closed-loop control of equipment, and comprises the following steps: triggering a feedback control procedure when the true activity retention concentration measurement value of the continuous plurality of sampling batches shows monotonically decreasing signs and approaches a lower control limit; the opening degree of an exhaust baffle of the spray drying tower is finely adjusted by sending command voltage to field control equipment so as to reduce exhaust temperature; the output frequency of the high-pressure pump frequency converter is synchronously regulated down so as to reduce physical shearing work in the fluid conveying process.
8. The method for detecting the activity quality of the peptide milk powder based on the wet production according to claim 4, wherein the absolute value of absorbance deviation in a non-absorption reference frequency band is extracted to construct a spectrum scattering drift factor, and the deviation measure is output by accumulating and averaging the absolute difference between the actually measured absorbance and the ideal absorbance, so that the severity of physical masking of the target polypeptide molecule by an external coarse structure is characterized.
9. The method for detecting the activity quality of the peptide milk powder based on the wet production according to claim 5, wherein in the process of constructing a composite stress index by carrying out nonlinear integration fusion on a dynamic temperature variable and a dynamic pressure variable on a time scale, fluid temperature data which is obtained by taking a maximum screening function and is not up to a damage threshold value is introduced, forward temperature difference accumulation is generated for the fluid temperature data which exceeds the damage threshold value, and the absolute pressure of the fluid is subjected to nonlinear mapping processing through a natural logarithmic function so as to be matched with a marginal decreasing rule in fluid mechanics.
10. The method for detecting the activity quality of the peptide milk powder based on the wet process production according to claim 6, wherein in the construction of the true activity retention concentration by coupling the apparent chemical concentration, the spectral scattering drift factor and the composite stress index, the optical masking is performed on the apparent chemical concentration by using the spectral scattering drift factor to reversely compensate and reduce the physical absolute total concentration, and the biological activity removal is performed on the basis of the natural exponential decay mapping item of the composite stress index to obtain the true activity retention concentration representing the biological potency.

Description

Method for detecting activity quality of peptide milk powder based on wet production Technical Field The invention belongs to the technical field of data processing, and particularly relates to a method for detecting the activity quality of peptide milk powder based on wet production. Background With the continuous development of the functional dairy industry, the prepared milk powder added with the active small molecular peptide component plays an important role in the market. The modern dairy enterprises commonly adopt wet processes to produce the high-quality milk powder. The complete link of the wet process comprises core procedures such as liquid phase batching, concentrated milk heating, high-pressure pump conveying, fluidized bed spray drying and the like. After drying and molding, in order to realize rapid quality detection of the concentration and activity of peptide components in the dry powder at the end of a production line, the industry generally adopts near infrared spectrum analysis technology to carry out nondestructive measurement, and concentration values are obtained by collecting spectrum sequences and combining partial least squares regression model inversion. The conventional physicochemical detection means have the defect of optical scattering baseline drift caused by microstructure dissimilarity under the specific engineering scene of wet production. The small molecular peptide is extremely easy to generate hydrophobic agglomeration with whey protein and the like under the coupling action of thermal stress and mechanical shearing force in the stage of heating and high-pressure pumping of concentrated milk subjected to a wet process. The physical state change greatly increases the surface roughness of the finally dried milk powder particles, and when near infrared light beams are irradiated on the surfaces of the coarse particles, a strong Mie scattering effect is induced, so that the reflection spectrum base line is seriously and non-linearly lifted upwards in the whole wave band. The conventional smoothing filter algorithm can only filter high-frequency clutter, and cannot inhibit low-frequency baseline drift caused by particle morphology change, so that the regression model misjudges an enhanced scattered light path as absorption attenuation, and serious concentration measurement omission and negative calculation deviation are caused. The existing spectrum detection technology has a technical blind area for representing disjoint physical and chemical concentration parameters and biological activity. The underlying detection principle of the partial least squares regression model is highly dependent on the chemical absorption peaks of peptide bonds in specific bands for concentration mapping, whereas the biological potency of a polypeptide substance depends on its complete three-dimensional spatial conformation. The instantaneous high temperature environment and high pressure fluid shear force in the wet process can lead to irreversible unwinding of the spatial structure of part of peptide chains, and the actual biological activity is lost, but the physical structure of peptide bonds on the main chain is not broken. The existing detection means only depends on transient spectrum characteristics of the end product, and cannot fuse physical parameter histories born by materials in the whole process life cycle, so that a large number of inactive components which have lost activity are mixed in the final inversion output concentration value, and the effective bioactivity retention rate of the product in delivery cannot be objectively reflected. Disclosure of Invention The invention provides a method for detecting the activity quality of peptide milk powder based on wet production, which aims to solve the technical problems that the optical measurement omission error is caused by physical agglomeration of microscopic particles in the wet processing of peptide milk powder, and the apparent detection concentration and the true biological titer are disjointed due to the fact that the complex process comprehensive stress induces the spatial conformation inactivation of peptide. The invention provides a method for detecting the activity quality of peptide milk powder based on wet production, which comprises the following steps: s1, acquiring fluid time sequence data of a production node and an original reflection spectrum sequence of a spray drying node, and performing smoothing filter operation on the original reflection spectrum sequence to obtain a smoothed spectrum after standardized treatment; S2, introducing the smooth spectrum into a pre-configured partial least squares regression algorithm model, and calculating and outputting the apparent chemical concentration of the peptide component in the current batch of milk powder; s3, based on background characteristics of a smooth spectrum and hydrodynamic characteristics of fluid time series data, constructing an operation index with causal