CN-121980321-A - Method and system for predicting delamination stability of dispersant suspension system

CN121980321ACN 121980321 ACN121980321 ACN 121980321ACN-121980321-A

Abstract

The invention relates to the technical field of monitoring of a dispersing agent suspension system, and discloses a method and a system for predicting delamination stability of the dispersing agent suspension system, wherein the method comprises the steps of collecting an original time sequence turbidity data stream of the dispersing agent suspension system, calculating a microscopic agglomeration index after a sliding window treatment, and generating probability density distribution; the method comprises the steps of performing asymmetric Gaussian fitting on probability density distribution to separate characteristic components, calculating attenuation of an interface coupling factor, further constructing a healthy phase space of a suspension system, mapping the index into real-time state points, calculating a Markov distance and direction angle sequence, finally judging dominant risk types and predicting macroscopic layering time nodes according to the sequences, and capturing micro dynamics change and interface failure signals in advance when macroscopic turbidity values are pseudo-stable by constructing a multidimensional phase space depth mining micro characteristic.

Inventors

Request for anonymity
Request for anonymity
Request for anonymity
Request for anonymity

Assignees

中科微点技术有限公司

Dates

Publication Date: 20260505
Application Date: 20260403

Claims (11)

1. A method for predicting delamination stability of a dispersant suspension system, the method comprising: Collecting an original time sequence turbidity data stream of a dispersing agent suspension system, performing sliding window segmentation on the original time sequence turbidity data stream to obtain a local data segment sequence, and calculating a microscopic agglomeration index sequence based on the local data segment sequence to form a probability density distribution sequence; performing asymmetric Gaussian fitting on the probability density distribution sequence to separate a main peak component and a tailing component, and calculating an interface coupling factor attenuation sequence based on the main peak component and the tailing component; and constructing a healthy phase space of the suspension system, mapping a microcosmic aggregation index sequence and an interface coupling factor attenuation sequence to the healthy phase space of the suspension system to form a real-time state point sequence, calculating to obtain a Markov distance sequence and a direction angle sequence based on the real-time state point sequence, judging the current dominant risk type according to the direction angle sequence, predicting a time node of macroscopic layering according to the Markov distance sequence, and outputting a full-period stability prediction report.
2. The method for predicting the delamination stability of a dispersant suspension according to claim 1, wherein said method for calculating a microscopic agglomeration index sequence based on a sequence of local data segments comprises: and performing time-frequency transformation on the local data fragment sequence to generate a time-frequency voiceprint image sequence, extracting a high-frequency Brownian energy sequence from the time-frequency voiceprint image sequence, and calculating a microscopic agglomeration index sequence.
3. The method for predicting the delamination stability of a dispersant suspension system according to claim 2, wherein the method for generating the time-frequency voiceprint sequence comprises the steps of: And performing short-time Fourier transform on each local data segment in the local data segment sequence, generating a corresponding time-frequency voiceprint image, and summarizing to form a time-frequency voiceprint image sequence, wherein each pixel point of the time-frequency voiceprint image represents the energy density of the corresponding time and frequency.
4. A method for predicting the delamination stability of a dispersant suspension system according to claim 3, wherein the method for extracting the high-frequency brown energy sequence comprises: and defining a high-frequency characteristic interval corresponding to the brownian motion of the suspended particles in each time-frequency voiceprint of the time-frequency voiceprint sequence, carrying out integral operation on the energy density in the high-frequency characteristic interval to obtain a high-frequency brownian energy value, and summarizing to form the high-frequency brownian energy sequence.
5. The method for predicting the delamination stability of a dispersant suspension system according to claim 4, wherein the method for generating a microscopic agglomeration index sequence comprises: And marking the first local data segment in the local data segment sequence as an initial reference data segment, taking the high-frequency Brownian energy value corresponding to the initial reference data segment as a reference value, and calculating the attenuation proportion of each high-frequency Brownian energy value in the high-frequency Brownian energy sequence relative to the reference value to obtain a microscopic agglomeration index sequence.
6. The method for predicting the delamination stability of a dispersant suspension system according to claim 5, wherein the method for generating the probability density distribution sequence comprises: and carrying out statistical analysis on all turbidity values in each local data segment in the local data segment sequence, calculating to obtain probability density distribution corresponding to the local data segment, and summarizing the probability density distribution corresponding to all the local data segments according to time sequence to form a probability density distribution sequence.
7. The method for predicting the delamination stability of a dispersant suspension system according to claim 6, wherein the method for generating the sequence of attenuation of the interfacial coupling factor comprises: Calculating offset distances between the centroid positions of the corresponding tailing components and the centroid positions of the main peak components aiming at each probability density distribution in the probability density distribution sequence to obtain centroid offset, and summarizing to form a centroid offset sequence; and calculating the attenuation proportion of each centroid offset in the centroid offset sequence relative to the offset reference value by taking the centroid offset corresponding to the initial reference data fragment as the offset reference value to obtain an interface coupling factor attenuation sequence.
8. The method for predicting the delamination stability of a dispersant suspension system according to claim 7, wherein the healthy phase space of the suspension system is represented by a microcosmic agglomeration index as a horizontal axis variable and an interfacial coupling factor attenuation as a vertical axis variable.
9. The method for predicting the delamination stability of a dispersant suspension system according to claim 8, wherein the method for generating a mahalanobis distance sequence and a direction angle sequence comprises: Defining a coordinate origin as an ideal uniform state origin in a healthy phase space of a suspension system, calculating the mahalanobis distance from each real-time state point in a real-time state point sequence to the ideal uniform state origin, summarizing to form a mahalanobis distance sequence, calculating the deviation direction angle of each real-time state point in the real-time state point sequence relative to the ideal uniform state origin, and summarizing to form a direction angle sequence.
10. The method for predicting the delamination stability of a dispersant suspension system according to claim 9, wherein said method for determining the current dominant risk type based on the sequence of direction angles comprises: acquiring a deviation direction angle corresponding to a current moment in a direction angle sequence, and judging that the current dominant risk type is a small molecular failure type risk when the deviation direction angle of the current moment is smaller than a preset angle threshold value; and when the deviation direction angle of the current moment is larger than or equal to a preset angle threshold value, judging that the current dominant risk type is a spreading failure type risk.
11. A dispersant suspension system delamination stability prediction system for use in implementing a dispersant suspension system delamination stability prediction method according to any one of claims 1-10, said system comprising: The microscopic agglomeration calculation module is used for collecting an original time sequence turbidity data stream of the dispersing agent suspension system, carrying out sliding window segmentation on the original time sequence turbidity data stream to obtain a local data segment sequence, calculating a microscopic agglomeration index sequence based on the local data segment sequence and forming a probability density distribution sequence; The interface coupling calculation module is used for executing asymmetric Gaussian fitting on the probability density distribution sequence to separate a main peak component and a tailing component, and calculating an interface coupling factor attenuation sequence based on the main peak component and the tailing component; And the stability prediction module is used for constructing a healthy phase space of the suspension system, mapping a microcosmic aggregation index sequence and an interface coupling factor attenuation sequence to the healthy phase space of the suspension system to form a real-time state point sequence, calculating based on the real-time state point sequence to obtain a Mahalanobis distance sequence and a direction angle sequence, judging the current dominant risk type according to the direction angle sequence, predicting a time node of macroscopic layering according to the Mahalanobis distance sequence, and outputting a full-period stability prediction report.

Description

Method and system for predicting delamination stability of dispersant suspension system Technical Field The invention relates to the technical field of monitoring of dispersing agent suspension systems, in particular to a method and a system for predicting delamination resistance stability of a dispersing agent suspension system. Background Dispersing agent suspension systems, such as pesticide suspending agents, paint color pastes and the like, are heterogeneous thermodynamically unstable systems formed by uniformly dispersing solid particles in a liquid medium by virtue of the action of a dispersing agent. In the long-term storage and transportation process, the physical stability of suspended particles is maintained, irreversible agglomeration, sedimentation and layering are prevented, and the key of guaranteeing the drug effect release and the service performance of the product is provided. Currently, the monitoring of the stability of a suspension system is changed from simple terminal quality detection to full-period dynamic prediction, and the aim is to pre-judge the degradation trend of the system in advance by real-time monitoring data so as to perform formula adjustment or inventory intervention in time. In order to evaluate the stability of suspension systems, various detection methods have emerged in the prior art. For example, chinese patent publication No. CN120197407B discloses a method and system for optimizing dispersion stability of herbicide suspension, which accelerates degradation process of the suspension by simulating various environmental stress conditions (such as temperature, humidity, water hardness), measures turbidity data in the degradation process in real time, evaluates dispersion stability by using turbidity change rate, and focuses on solving problem of influence evaluation of environmental factors on stability. In addition, the Chinese patent application with publication number of CN121275571A discloses a method for detecting the performance of a macromolecular dispersing agent for producing polymer polyol, which focuses on the production and manufacturing links, and constructs multidimensional indexes by collecting dispersing time, process energy consumption, particle size distribution (D50 and D90) and rheological parameters, thereby realizing intelligent diagnosis and optimization of the performance of a dispersing agent production process. However, the above-mentioned techniques rely mainly on significant changes in macroscopic physical quantities as a basis for determination, with hysteresis and dead zones in the face of the early occurrence of the micromechanics evolution of the dispersant suspension system. Specifically, when the small molecular encapsulation ability of the dispersant begins to fail, "blocking" between suspended particles tends to begin at the microscopic level, where the particles, although they are brought closer together by van der Waals forces to form tiny agglomerated precursors, remain suspended in the liquid medium without macroscopic settling. From the light scattering principle, the microscopic agglomeration causes the equivalent particle diameter of the particles to be slightly increased, so that the light scattering cross section is increased, and the turbidity reduction effect caused by the weak reduction of the local particle number density is offset to a large extent. The antagonism of this physical mechanism results in the overall turbidity value remaining "pseudo-steady" for a considerable period of time, even with an abnormally slight rise. The existing monitoring method is easy to be deceived by the appearance of the numerical level, and the erroneous judgment system is in the stable period, so that microscopic instability signals such as brownian motion frequency reduction and the like are ignored. Once the microscopic agglomeration accumulation breaks through the critical point, the system can rapidly break and macroscopically delaminate, and then an alarm is triggered, so that the best opportunity for redispersion treatment or formula remediation is usually missed, and economic loss and product waste are caused. Disclosure of Invention In order to overcome the defects in the prior art, the invention provides a method and a system for predicting the delamination stability of a dispersing agent suspension system, and aims to solve the problem that the traditional monitoring means are shielded by the phenomenon of pseudo stability of macroscopic turbidity. According to the invention, by constructing a two-dimensional healthy phase space containing microscopic agglomeration indexes and interface coupling factor attenuation, and utilizing the hidden Brownian motion change and interface wetting failure characteristics in the Mahalanobis distance and direction angle depth deconstructing time sequence data, the microscopic instability risk of a suspension system is accurately identified and early-warned at an early stage of invisible na